The long answer is a graphical organization of documents and data that work much like a database, but without the overhead. It’s a table of contents, an index, and a filing system all in one. And my AI executive assistant puts things there, and, more importantly, links them in a way that lets them be found by any AI session that needs it.

What my second brain holds is the scaffolding for everything about my life and business, with the exception of what goes into StorytellerOS which is where my manuscripts, social media, newsletters, blog posts, ads and sales data goes. Storyteller is my office. Second brain is the operating manual.

And one of the most important parts of the operating manual are Skills.

Skills are written instructions for the jobs I do every week, stored in my second brain, and used by whichever AI happens to be behaving that day. There are “official” places inside the apps that skills are stored, but what I’ve done is told Claude and ChatGPT to look in second brain for the skill they need and use it, so I only have to keep one set of skills updated in one place. Some might call this efficient. I will remind you I am fundamentally lazy, which is why I automate everything.

Memory and skills do different jobs

Memory is what the AI needs to know about you. Claude keeps a profile about who you are, what you do, your background, and topics like recent work, tools, voice, areas, etc.

A skill is how a particular job gets done. It says what the job is, what goes into it, what a finished result looks like, and where the AI’s authority ends and yours begins.

Both Claude and ChatGPT have memory in their settings. The idea of it is great. The application and real-world use of it is… not great. It remembers the wrong things very often, or not at all.

I store both memory and skills in second brain, and I’ve switched the settings in both Claude and ChatGPT to write anything worth remembering there instead of the internal memory. The results are superior.

Why a skill beats a good prompt

Most of us started with prompts. We moved on to optimizing prompts next, and told the AI to “act as” a role, like a marketer, or a social media guru. Previous versions of AI loved those frameworks, but modern versions ignore it. It also eats a lot of tokens.

A prompt assumes the AI remembers things it doesn’t, and it leaves out the dozen small decisions you make without noticing you’re making them. Paste the same prompt on two different days and you’ll get two different results, then spend twenty minutes repairing whichever one is worse.

A skill carries those decisions with it. It points at your voice guide and your book list so the AI reads them fresh every time. It says “two hashtags maximum on Instagram” in writing, so you never again have to scroll past the seventh one and feel your eye twitch.

Personal Skills, Project Skills, and Imported Skills

Every skill I own lives in my second brain, whatever app it started in. Skills show up from three directions, and two of them will quietly scatter your work across accounts if you let them.

Personal skills are the ones you write for yourself. You describe a job you do every week, the AI turns that into instructions, and you refine them every time it gets creative. In Claude, a personal skill applies to your whole account, which is what you want for anything you’d use more than once.

Project skills are the sneaky ones. When you’re working inside a project in Claude, the default is to create skills scoped to that project. They work beautifully there and don’t exist anywhere else. Unless you specifically ask for a personal skill, you’ve built a tool that lives in one room of the house and can’t find the door. I’ve watched people rebuild the same social post skill three times across three projects and assume the AI was just being inconsistent.

Imported skills come ready-made. The AI companies build some into their apps, and plugins arrive with their own. The Reader Radius plugin, for example, ships with skills for creating brand guides, creating social posts, building trial reels, and batching campaigns. You can see them all here: https://github.com/chellehoniker/claude-code-author-automations/tree/main/skills

You can’t move imported skills into your second brain because they’re the instructions that come with the app, but you can supplement their skills with your own to be sure that your wishes are followed.

That way, you can move between AI companies, and you can move between accounts inside the same company. I run two Max plans on Claude because I refuse to be told my afternoon is over by a token counter, and both accounts read the same skills from the same folder.

Mine live in Obsidian as plain text files, with the book and marketing details they depend on in StorytellerOS. Notion works, and so does a folder on your computer. Obsidian adds the graphing part that makes it a better choice than Notion, but in the end, what matters is that it’s yours.

How I use one: create a month of Pinterest Posts

One of my skills is to batch create Pinterest Pins. It takes me about 15 minutes each month to do this.

The skill tells it to:

1. Pull the blurb, series position, and buy links from StorytellerOS, where each title lives with its marketing copy and retailer links.

2. Read the voice guide for that pen name. Lola MacDougall and Chelle Honiker do not share a voice, and I intend to keep them from meeting.

3. Draft thirty posts with a different angle each: the premise, a character, a setting detail, a reader quote, the trope, a series nudge, and a direct buy post.

4. Create video pins and image pins

5. Show me everything in one page for review.

6. Use Reader Radius to schedule all the posts.

What goes into a good skill

“Help me with social posts” gives the AI an impressive amount of room to embarrass you. A useful skill answers the questions a smart new hire would ask on day one:

• The job. One sentence on what this skill produces.

• The audience. Who reads the result, and which pen name or brand it belongs to.

• The inputs. Where the source material lives: your book list, your voice guide, last month’s best-performing posts.

• The format. Length, structure, platform rules, and everything you always or never do.

• An example. One result you were happy with, so the AI can see the target instead of guessing at it.

• The handoff. Where the draft goes and what needs your eyes before it goes anywhere.

You don’t need all of that on the first pass. Write what you know, run it, and add whatever the AI gets wrong. One bad draft teaches you more than an afternoon spent imagining every way it could go sideways, and the AI will find ways you didn’t imagine anyway.

Plan on some upkeep

Skills need maintenance. A new model reads an instruction differently than the last one did, a platform changes its character limits, and a job you did one way in March turns out to need a step you didn’t know about in October.

When the AI keeps making the same mistake, fix the skill. Fixing the draft by hand every week is how you end up doing the job yourself while also paying for an assistant.

Your homework

Pick the social media job you’re most tired of doing, or most tired of correcting.

1. Start from a ready-made skill, like the Reader Radius social post skill, or ask your AI to draft one based on how you’d explain the job to a new assistant.

2. Save a copy in your second brain and point your AI at that copy.

3. Read the instructions and fix anything that doesn’t sound like you.

4. Add one example post you’ve written and liked.

5. Run it on three different books, and add each correction to the skill.

By the third run you’ll know whether you’ve built something useful, and your AI will have stopped trying to work “discovering the truth” into a post about a bakery.

Bring it to class

Every Wednesday at Noon Central, I go live and break down some social ideas and skills - it’s free and you’re welcome to see it in action!

Register here: https://us02web.zoom.us/meeting/register/BdYI5JwxRx6MHIR9hn13YA

I’m also teaching a three-hour second-brain session at Author Nation, where skills are a big piece of the build. Register here: https://www.authornation.live/Addedworkshops

Chelle

P.S. Next week is voice guides - creating your own social media, brand guide, copywriting guide and style guide so your assistants sound like you!

Every AI tech bro on the internet is currently building a second brain and would love to tell you about it at length on their podcast or YouTube channel. Annoyingly, they have a point. There are several good reasons to build one, and my current favorite is being able to tell Claude to piss off.

Lately, Claude and I had been disagreeing about who was in charge. I’d ask for something specific, and it would decide it knew better. Having access to my business files seems to have given it the impression we were partners. We aren’t splitting the expenses, so I’m unclear on how we got there.

I kept having to explain to a computer that I’d already thought about what I wanted before I asked for it. That’s an irritating conversation to have with a person, and having it with software I voluntarily invited into my life AND PAY FOR adds a special layer of embarrassment. I have enough people telling me what to do without a bot piling on.

So I switched to ChatGPT, pointed it at my second brain, and kept working. Mine is split between Obsidian and StorytellerOS, and the new tool could read all of it. I’d already written down how my business works and how I want things done, so I got to act on my irritation right away, without first calculating how long it would take to train another AI from scratch.

I’m not married to any of these tools

I’ll say it again. All of these companies are in an arms race. They are going to keep releasing new models, and I’m going to keep using whatever works for me at the moment. I don’t plan on a long-term commitment to a product that might behave differently after its next update.

Some of you have read a lot of Claude tutorials from me, so this email may feel like I showed up to the family reunion with a different date. Those tutorials were based on what worked at the time. I’ll keep telling you when something works and when I’ve moved on to something better. I’m comfortable with that level of inconsistency in my relationship with a software company.

The competition has produced some extraordinary capabilities, and I’m also watching for regressions, which is the polite technical term for an update making something worse. A model can get better at difficult work and somehow get worse at an instruction it used to follow without complaint. A skill you spent time getting right may need revising because the new version interprets it differently.

The release announcements focus on the improvements, naturally. You’ll usually discover the rest while trying to finish something you’d told yourself would take ten minutes.

Start with a folder

“Build a second brain” can sound like an invitation to spend a weekend arranging folders when you should be doing the work inside them. I enjoy an unnecessarily elaborate system as much as anyone, so I’m probably the wrong person to supervise you near a blank dashboard.

What matters is getting the information out of your head and out of any one AI app, in a form you can use again. In my setup, that’s written context about my business and instructions for how I want work handled.

I keep mine in two places. Obsidian holds my notes, decisions, and AI instructions as plain text files on my own computer, which I can open with or without any app at all. It’s exceptional for this because notes link to each other and an AI can follow those connections, though I’ll admit it takes more setup than most people want on a Tuesday night. StorytellerOS holds the author side of the business: series, manuscripts, retailer links, pen names, newsletters, and social posts. I built it because my author business was scattered across more tabs than I care to admit. Whichever AI I’m using that week can connect to both.

Plenty of people do this in Notion, and plenty more just use files and folders on their computer. Any of those work if you can open your documents without the AI.

You can start with a folder. The name of the folder is not a decision that requires a research phase.

What to put in it

Some of it is straightforward. For each series, list the books in reading order with links to where readers can buy them. (Mine sit together in StorytellerOS, so I’m not hunting down the right buy link at midnight.) Add enough explanation that an assistant knows which book you’re promoting. You probably don’t want it inviting new readers to start with book four, right after someone they were supposed to get attached to in book one has been murdered.

If you write under more than one pen name, make the separation explicit. Your sweet romance readers and your monster romance readers may have different expectations for a newsletter surprise, and leaving the AI to guess which audience you mean is an unnecessarily exciting way to spend release day. I keep each pen name walled off in its own space, and I’d recommend some version of that wherever you keep yours.

The harder part is the stuff you’ve stopped noticing you know. You’ve decided how often you’re comfortable emailing readers. You remember why you quit a particular promotion. You probably have a perfectly good reason for sending people to one retailer in certain situations, and another for skipping a popular tactic that doesn’t fit your books. Write those decisions down with enough context to be useful, or you’ll keep getting sensible suggestions you already considered and turned down, and you’ll spend the time you meant to save explaining yourself to a chatbot.

Show it how you write

“Friendly and conversational” will get you approximately nowhere. Give the AI examples of things you’ve actually written, tell it what sounds off in its drafts, and save the corrections you keep having to make. The ones that make you want to delete the paragraph and do it yourself are the most important to keep.

An assistant needs to see how you explain an idea, including where you stop explaining because your readers are intelligent adults who can follow a thought on their own.

(I have feelings about this. You may have noticed.)

My voice samples and past newsletters live in the same place I write and send the new ones, which means the assistant drafting this week’s email is working from the real thing. The more detail you capture, the less you rely on an app remembering a conversation correctly. You also end up with a record you can read. If an instruction is wrong, you can fix it yourself without trying to work out what the bot thinks it knows about you.

Memory and skills

Memory is the information about your business and the decisions you’ve made. A skill is a set of instructions for doing one particular job, like preparing your weekly newsletter.

“Help with my newsletter” leaves an impressive amount of room for interpretation. A useful skill tells the assistant who the audience is, where the voice guide lives, how long the finished draft should be, where to save it, and what still needs your review.

It can be rough on the first pass. You’ll find out what’s missing when you use it, and that’s a lot more helpful than spending three days imagining every possible exception.

Plan on some upkeep. A new model might keep misreading one of your instructions, or a tool might change how it connects to another app, and you’ll need to revise that part. Keep the current version with the rest of your materials. If I’ve spent an afternoon working out how to get a good result, I want to carry that work into whatever tool I use next.

Keep a copy you can actually open

An app assuring you it will remember something is easy to mistake for a filing system.

You should be able to open your documents without opening the AI app, and they should make sense when you read them. Keep a backup somewhere else, too, so an unavailable service doesn’t take your only copy with it. I follow my own advice here. My Obsidian files sit on my computer and get backed up separately, and I export from StorytellerOS too, because I’d feel ridiculous writing this newsletter otherwise.

If your instructions only exist inside a project in one AI app, find out how to export them while the app is working and you still have access. Then read the export. A huge archive of conversations might have what you need buried in it somewhere, and I don’t want to spend the morning of a launch looking for it.

Your current AI can help turn those conversations into usable documents. Ask it to pull together what you’ve decided about a particular task, and then check its work. It may leave out a correction or include an idea you talked about and never agreed to, and you’ll want to catch that before it becomes instructions for the next assistant.

Reading it over also has a way of showing you what you never actually wrote down, or where your own instructions contradict each other. I would love to tell you every communication problem starts with the bot. Unfortunately, I have met us.

Why this matters for your business

Keeping the only copy of your memory and skills inside one AI app is a recipe for disaster. We’re handing these services more of our daily work, so their business decisions can turn into our business problems quickly.

Prices can go up. You might still like the product, but liking it doesn’t mean you have to pay whatever it asks. If you can move your information, you have a real choice when you decide whether the new price makes sense. If you can’t, you’re comparing the new monthly bill to the time it would take to rebuild everything you’ve taught the app, and that’s a miserable little calculation after months of getting things the way you want them.

Terms can change, too. Something you were fine with at signup might not be acceptable anymore, and I want to make that call based on the terms without worrying that leaving will cost me my own working instructions.

Services also go down. The company may be doing everything it can to fix an outage, and your release email still has to go out. If the only copy of your process is in that AI app, you’re stuck trying to remember which links go where and rebuilding the instructions that kept your drafts from sounding like a motivational speaker wandered into your inbox. My release emails and social posts go out from StorytellerOS, so when one AI is having a bad day, I connect a different one and keep going while the first one sorts out its emergency.

Switching still takes some work

A skill may need adjusting for the new app. Connections to your other services might have to be set up again, and you’ll need to check the results, because a new assistant reading your instructions won’t necessarily follow them perfectly. You’ll have the source material and the decisions you already made, though, and that’s a much more manageable job than trying to remember everything you’ve ever told an AI.

You can use the same material with more than one tool if you like doing different jobs in different apps, and you won’t have to maintain several competing biographies of yourself. It’s useful for humans, too. If you hire an assistant someday, you can give them a login to where everything already lives, which beats handing over your inbox and a hopeful plan to remember the important bits.

If you only do one thing

If this has you mentally adding “document my entire existence” to a list you already resent, start with the recurring job where you’re most tired of correcting the AI.

1. Ask your AI to draft instructions for that job using the context it has.

2. Read the draft as if you were handing it to someone on their first day. Add the decisions it missed and remove anything it made up, before a confident guess turns into company policy.

3. Save the corrected version outside the AI app, along with an example of a result you were happy with. (Obsidian, StorytellerOS, Notion, or a plain folder all work. Pick the one you’ll actually open again.)

4. Give it to a different AI and ask it to do the same task.

You’ll see what carries over and what still needs explaining. You might decide your original tool does a better job, and that’s good information to have. You’ll still own the document.

Where to get help

There are new free classes in the Author Automations Hub for your burning AI questions. If the idea makes sense and you get stuck trying to make it work with your own files, bring that with you. The part where the instructions seem reasonable and the software stays deeply uncooperative is a very familiar activity around here.

We’re also planning an AI Rodeo in San Antonio next year, and I’ll share more as plans come together. I’m looking forward to getting a room full of authors working on this together. It’s reassuring to watch somebody else’s supposedly helpful assistant need a little adult supervision.

Someone drops a screenshot of their copyright page into the chat that’s some variation of: “No part of this book may be used to train artificial intelligence.” Forty people pile in with “Stealing this!” Everyone exhales and goes back to writing, feeling like they just hung a clove of garlic on the windowsill and the vampires will have to find another house.

I get it. But you won’t find that line in my books, and it isn’t because I’m reckless. We’re standing guard over a battlefield that was cleared months ago while the real action moved down the street.

The fight over AI training is mostly over. Your new book doesn’t need to be in a model’s training data to show up inside an AI answer today.

While authors were pasting anti-training disclaimers into Vellum, modern AI moved on to retrieval. That isn’t a stealth attack or a legal loophole. It’s a permission slip that has had your signature on it since the day you clicked “Agree” on your retailer dashboard. We all did.

I covered the authorship side of this back in June for Indie Author Magazine: who owns AI-assisted drafts, what the Copyright Office actually wants disclosed, and how to register your work properly. This is the other half of the story, which is where training law actually landed, how retrieval works when a reader asks a chatbot about your plot, and why the publishing contracts we already signed cover it.

The Training War Has a Scoreboard (and Models Don’t Hold Your Manuscript)

The copyright-page disclaimer rests on the idea that an AI model swallows your book whole, keeps the PDF on a hard drive somewhere, and regurgitates it on request like a well-read cat.

The courts spent two years taking that idea apart.

Training leaves patterns, not files. In Bartz v. Anthropic (2025), the court ruled that training models on lawfully acquired books is transformative fair use. The machine reads your text once alongside millions of other books, nudges a few billion mathematical weights by a hair, and discards the original. In Kadrey v. Meta (2025), top technical experts couldn’t force Llama to cough up more than 50 consecutive words of the plaintiffs’ books. Your villain’s monologue is not sitting in a database waiting to be lifted.

The $1.5B Anthropic settlement was about piracy, not training. When final approval landed in July 2026, authors got checks of roughly $3,000 per registered title because Anthropic downloaded books from torrent sites like LibGen. In that same case, the judge said plainly that buying physical paperbacks, shredding them, scanning them, and training on them was legal fair use. The pirated files were the crime, and the training wasn’t.

New books aren’t in training sets anyway. Unless your novel is quoted word-for-word on thousands of websites the way Harry Potter is (which hit roughly 42% memorization in stress tests), a trained model holds zero recoverable manuscript text from your work. Your debut cozy mystery is not, statistically speaking, Harry Potter.

So the training fight is scored, and your brand-new book can still show up in a detailed AI answer tomorrow.

The Real Engine: From Training to Retrieval

While everyone was arguing about training datasets, the tech world quietly shifted to Retrieval-Augmented Generation, or RAG, because nobody in tech has ever named anything well.

Training is a massive, static event done up front. Retrieval happens live, the moment a user types a prompt.

When a reader asks an AI tool a specific question about your book, the system doesn’t rummage through the model’s memory. It fetches the relevant pages directly from a live file or database, hands those paragraphs to the AI as a temporary cheat sheet, writes a summary from that cheat sheet, and then forgets it ever saw the thing.

Nothing was trained and no model learned your plot points, but your text was retrieved, processed, and served on demand like room service for chatbots.

You see this on the open web every day with Google AI Overviews and Perplexity. For your ebooks, though, the database being searched isn’t scraped off the open web by some random crawler. It lives on the retailer server where you uploaded it, on purpose, with a cover and a blurb attached.

You Already Authorized the Index

Your manuscript isn’t floating around for web bots to trip over. It sits in retailer databases, and retailers don’t need to crawl your site because they already hold the pristine, high-res file you handed them under your distribution contract.

Go open Section 5.5 of the KDP Terms and Conditions, which I promise is riveting. In exchange for putting your book in the store, you granted Amazon an irrevocable, nonexclusive license to:

“reproduce, index and store Books... and reformat, convert and encode Books.”

Index. Store. Convert. Encode.

That clause wasn’t written to pull a fast one on anyone. It was written so Amazon could host your file, run standard search, build “Look Inside” previews, and format your book for different devices. But under the hood, interactive tools like Ask This Book on Kindle aren’t relying on AI training or fair use at all. They’re RAG retrieval with a conversational interface bolted on.

Amazon’s own technical breakdown says as much. The feature uses your manuscript purely as temporary prompt context, it’s classified as an extension of standard search functionality, and it runs as an always-on feature across the store with no author opt-out.

Amazon didn’t need to ask us for new AI training permissions because retrieval is indexing, and we all said yes to indexing the day we published our first title.

Fighting the Right Problem

Back in 2023, the Authors Guild drafted a model clause reserving the right to “reproduce and/or use the Work for purposes of training artificial intelligence technologies.” Thousands of indies pasted it onto page two, and traditional publishers slotted it in beside their standard copyright notices.

That clause was built for 2023. It addresses whether a model can ingest text to learn up front, and it says nothing about real-time search, indexing, or retrieval prompts, because RAG wasn’t what anyone was staring at when it was drafted. It’s a good lock on a door nobody uses anymore.

Natural-language disclaimers don’t override fair use in the US anyway. Over in Europe, the Higher Regional Court of Hamburg (OLG Hamburg, Dec 2025) ruled that human-readable copyright notes don’t satisfy the legal requirement for machine-readable opt-outs. So in at least one major jurisdiction, the sentence isn’t even legible to the thing it’s aimed at.

Where That Leaves Us

This isn’t a call to panic or yank your books off retail shelves. It’s a call to see the board clearly so we stop spending energy on disclaimers whose main job is making us feel better.

Training law has settled boundaries. Training on legally acquired copies leans hard toward transformative use, while sourcing from pirate archives remains flatly illegal.

Retrieval is modern search. Tools that answer questions about your book in real time run on database permissions and search indexing rather than model training.

Standard retail terms already cover it. Distribution agreements like KDP §5.5 grant retailers the right to index files for interactive search, and you agreed to that long before anyone said “chatbot” out loud.

Direct sales are the one place you hold the pen. If you don’t want a platform indexing your files or running interactive search against your text, selling direct on your own site is the only space where you write the license terms yourself. Everywhere else you’re a guest.

We don’t need to keep re-fighting the 2023 war over whether a model “swallowed” our words. We just need to understand that when we hit publish, we agreed to let retailers index our files, and in 2026, search indexes talk back.

My original piece on copyright registration and ownership is over at Indie Author Magazine: AI Copyright for Indie Authors: Where We Stand Now. One update to that article: the $1.5B Anthropic settlement it references received final judicial sign-off on July 20, 2026.

Last week I walked you through getting the desktop app set up, why Cowork lives there instead of the web, and what a terminal is for those of you who remember installing Oregon Trail from a floppy disk. (The number of you who DM’d me about dying of dysentery was deeply validating.)

This week we’re getting into the features that actually make Cowork powerful enough to justify the subscription. There are three concepts to understand: Connectors, Skills, and Plugins. They sound more technical than they are, and by the end of this newsletter you’ll know what each one does, how they relate to each other, and which ones matter for where you are right now.

Connectors: How Claude Talks to Your Other Apps

A connector is exactly what it sounds like — a bridge between Claude and another app you use. When you connect Google Drive, Claude can read and search your files without you copying and pasting content into a chat window. When you connect Gmail, it can search your email, find threads, and draft replies. When you connect Google Calendar, it can check your schedule and add events.

Without connectors, Claude is stuck in the same copy-paste loop you’ve been in with every chatbot: you manually bring information to it, and you manually take information back out. Connectors let Claude go get what it needs and put things where they belong.

To set one up, open the Claude desktop app, go to Customize, and you’ll see a directory of official connectors — Google Drive, Gmail, Google Calendar, Zoom, Notion, Slack, Stripe, Zapier, and a list that keeps growing. Click “add,” sign into your account, and you’re connected. Claude can now read from and write to that service.

If you’ve been reading this newsletter for a while, you might remember my deep dives on MCP servers. Connectors and MCP servers are the same thing with different names. MCP stands for Model Context Protocol (you absolutely do not need to remember that) — it’s just the technical term for how Claude connects to other apps. When Anthropic builds an official connector and puts it in that directory, they’re packaging up an MCP connection with a friendly button and some security vetting so you don’t have to think about the plumbing.

Beyond the official connectors, Claude let’s you also configure Personal connectors, which are available by first connecting a Marketplace link provided by the plugin developer, or uploading a plugin. My preference is a Marketplace link because the plugin can be configured to remain in sync with the developer.

Web Connectors vs. Desktop Connectors (Yes, They’re Different, and Yes, It’s Confusing)

This trips people up, so I want to lay it out clearly — and acknowledge that even I find this part messy.

There are two types of connectors: web connectors and desktop connectors. Web connectors run on the company’s servers in the cloud. Desktop connectors install something on your local machine and run from there. Both types can be set up inside the Claude desktop app through Cowork, which is where the confusion starts — you’re in the desktop app, but some of the connectors you’re adding are web-based and some are local.

The practical difference matters more than the technical one. Web connectors generally work the same on every machine because they’re running on someone else’s server. Desktop connectors can be finicky because they sometimes require additional software installed on your computer (things like Node or Bun, which are developer tools that most authors have never heard of and shouldn’t need to care about). Whether a desktop connector works smoothly can also depend on whether you’re on Mac or Windows, because the two operating systems don’t always play nice with the same configuration.

I learned this the hard way with authorautomations.social. We originally built it as a desktop connector, and it worked great on Mac because 99% of Macs ship with the software we used for the connector. Then Windows users started testing it from behindVPNs and we hit inconsistencies that were frustrating for the users to troubleshoot. So we rebuilt it as a web connector, and the cross-platform headaches went away. We lost a tiny bit of functionality, but it was worth the trade-off.

My prediction (and keep in mind that we are still in the genuinely early days of all of this): most companies you’d want to connect to will have official web connectors soon. The direction this is heading is toward click-to-connect web connectors that work regardless of your operating system, and desktop connectors will become more of a power-user option for custom setups. In the meantime, if you run into a connector that’s giving you grief during setup, it’s probably a desktop connector with a dependency your machine is missing. Drop a note in the Discord and we’ll help you sort it out.

The One-Account Limitation (And How to Get Around It Safely)

One thing worth knowing about official connectors: most currently support one account per service. So if you have three Google accounts (I have approximately 4,000, which is a me problem), the official Google Drive connector lets you connect exactly one of them.

For most authors with a single Google account, this is a non-issue. For those of you juggling a personal account and a business account, or multiple pen name accounts, it’s a real limitation — and this is where third-party connectors enter the picture.

There are developers out there building connectors that solve specific problems like this, and some of them are genuinely good. The challenge is that this space is the wild west right now, and you need to know how to vet what you’re installing before you give it access to your Google account (or your email, or your files, or anything else you care about).

For example, if you Google “MCP server for multiple Google accounts,” one of the first results is a listing on mcpmarket.com for a multi-account Google connector. No shade to the developer — it might be a perfectly fine product. But when I look at the details, the developer has one follower, and the underlying software repository has one star. That tells me almost nobody has reviewed this code, and I’m not comfortable connecting my Google accounts through it.

Compare that with the google_workspace_mcp connector by Taylor Wilsdon, which is what I use and what I walked through at the AI Summit (including a separate video step-by-step I recorded after because everyone asked for it!). That repository has 2,300+ stars and over 700 developers contributing to the code. That level of community involvement means a lot of eyes have looked at this software, a lot of people are actively using it, and problems get found and fixed quickly.

When you’re evaluating any third-party connector, look at the GitHub repository behind it. 99% of developers use Github. Check the star count (think of these as votes of confidence from other developers), check how many contributors are involved, and check when it was last updated. A connector with thousands of stars and recent activity is a very different proposition than one with a single star from the person who built it. You’re potentially handing these tools a lot of power — treat that decision the way you’d treat giving someone a key to your house.

Skills: Teaching Claude to Do Things Your Way

Connectors give Claude access. Skills tell Claude what to do with that access and how to do it the way you want.

A skill is a saved set of instructions — think of it as a recipe card for a specific task. It includes what the task is, how you want it done, what tools to use, what your preferences are, and what standards to apply. When you activate a skill, Claude follows those instructions instead of making its own assumptions about how you’d like things handled.

I have over 100 skills at this point, and that’s genuinely where I think everyone should be heading. Every repeatable task in your business is a candidate for a skill, because every skill you create is one less time you have to explain yourself from scratch.

My Brevo Mastermind Newsletter skill drafts my weekly newsletter to my mastermind group with the right tone, format, and sign-off every time. It also scans my Notion for things I’ve added throughout the week that I wanted to remember to tell them about, because my neurospicy brain is not going to hold onto the thing I really, really wanted to mention by the time Thursday rolls around. I just toss it into Notion whenever I think of it and let the skill find it when it’s time to draft.

My Update WooCommerce Products skill, combined with a project in Cowork, knows that my covers are in a specific folder on my computer and my book copy is in Airtable. Because both are connected, it goes and gets everything it needs — the cover image, the description, the metadata — and adds the new book to my online store. (You could do the same thing with Shopify, Payhip, or whatever you use to sell direct.)

My Email Triage skill has my instructions for how to sort, prioritize, and respond to email across all my accounts, and we go 1x1 twice a day (with a scheduled task we’ll talk about in future posts because those deserve their own detail, and a small parade to celebrate them) and respond, archive, defer based on rules in the skill and detail from my second brain in Obsidian and Notion.

Each skill tells Claude exactly how I want that type of work done, and that specificity matters more than you’d think. Without a skill, I can absolutely say “go get my book copy out of Airtable” in a regular chat, and Claude will try — but it might take 15 attempts of hunting and pecking around my tables and fields before it finds the right one. That eats up your tokens (which cost money or count against your plan limits) and your time. With a skill, I’ve pre-defined which Airtable base, which table, and which field to pull from, so Claude goes straight to it on the first try.

Start with one. Pick the task you do most often — maybe it’s drafting social media posts, or formatting your newsletter, or summarizing reader feedback — and start a conversation with Claude in Cowork. Say something like “interview me about how I do this task, what tools I use, and how I want the output to look.” Claude will ask you questions, and when you’re done, it’ll save a skill you can reuse every time you do that task.

The part I love most about skills is that you don’t have to go back into the settings to update them. If you’re chatting with Cowork throughout the day and you realize a skill is missing something, just say “hey, add this to my copywriting skill as a non-negotiable rule.” Claude updates the skill for you, right there in the conversation. You’re teaching it incrementally, the same way you’d train a real assistant — one correction at a time, in the flow of actual work. Change your mindset from “I need to go edit a settings file” to “I need to tell my assistant to update their notes.”

Plugins: Where Skills and Connectors Come Together

Plugins are bigger than a single skill or a single connector. They’re bundles that can combine skills, connectors, agents, hooks, monitors, themes, and other server functions into a package that works together for a specific purpose.

The easiest way to understand this is through an example. AuthorAutomations.social has a plugin that bundles two things: skills for creating carousels, Reels, and TikToks with trending music from inside Cowork, and the web connector that posts them to your social accounts when you’re done. The skill is the framework — it works with Claude on how something gets created, what format it takes, what dimensions to use. The connector is the pipe — it delivers the finished product to where it needs to go. The plugin wraps both of those together so you’re not configuring them separately.

I think plugins are where this ecosystem gets really interesting over the next year, especially for business intelligence. I’ve suggested that Author Nation build a plugin, for example, that will have the framing and theme of this year’s conference built in (speakers were announced yesterday, and Joe Solari explained the theme and how the philosophy is adapting in this YouTube Live) along with a way to incorporate what each speaker presented.

But Chelle, you ask, isn’t this just a chatbot? No. It’s a business intelligence source incorporated into your own business.

This of it like this: You’re working on your Facebook Ads. Claude knows your analytics and spend from a connector to StorytellerOS (or Meta, or another connector). It knows your personal objectives from your second brain, Notion. And now it knows the things you learned at Author Nation from the smartest people in the industry, and it can create new ads or update your existing ones based on that intelligence, and your personal analytics and objectives.

It’s early days for plugins, but it gives you a sense of where plugins are heading — pre-built packages of expertise and context that you can connect to Claude.

Imagine a stoicism skill from Ryan Holiday, or the “Badass” series from Jen Sincero packaged as a plugin you could connect to your business planning workflow. Those don’t exist yet, but they should, and I’d be genuinely surprised if we don’t see things like that within the next six months. The infrastructure for it is already here — someone just needs to build them. (And if you’re reading this and thinking “I could build that,” come talk to me in the Vibe Coding Discord.)

How It All Fits Together

Think about the relationship between these three things this way:

Connectors (really just MCP servers or API connectors — same thing, different name) are the roads between Claude and your apps. They handle the “can Claude reach this service” question. Official connectors in the directory are just connections with a friendly button and some security vetting on top.

Skills are the driving instructions. They handle the “once Claude gets there, what does it do and how does it do it” question. Without skills, Claude uses its own judgment about how to handle tasks. With skills, it follows your playbook.

Plugins are the full trip planned out — a bundle of roads and driving instructions (and sometimes other components) packaged together for a specific purpose. AuthorAutomations.social’s plugin gives you both the creative skills and the posting connector in one install.

Where This Is Going Next Week

Next week we’re talking about Projects — how to set up your workspace so you give Claude access to what it needs and keep it away from what it doesn’t. There’s a heavy emphasis on security, access, and knowledge management, because the way you organize your Claude environment determines how useful (or how chaotic) the whole experience is. We’ll cover options for “second brains” including Obsidian, Notion, and Google Workspace, and I’ll walk you through how I have my executive assistant structured so it knows where everything lives across my business.

In the meantime, if you haven’t downloaded the Claude desktop app yet, this is the week to do it. Everything we’re building toward in this series lives there.

Reply and tell me: what’s the first app you’d want Claude to connect to, and what would you have it do there? I read every one.

—Chelle

Last week, over 700 of you showed up for the AI for Author Business Summit. I had planned to pre-record some sessions to save my voice, but Anthropic kept shipping updates faster than I could finish a slide deck, so we did the whole thing live. My voice survived. Barely. Theater kids don’t quit, y’all — the show goes on.

By the end of Day One, people were connecting Airtable to Claude Cowork and chatting about their 300+ backlist and series bible details. By Day Two, attendees were chatting with their email, setting up scheduled tasks to update their websites, and a few brave souls had hacked it to update their Scrivener docs, and clean their Mailerlite email lists. Someone else hooked up their Facebook Ads and rebuilt their entire strategy with new images and video they created. (I love you people.)

And then there were the social media shifts. They learned how to connect AuthorAutomations.social inside Claude Cowork and create and schedule all their posts from a chat window —including images, videos, infographics, carousels, and yes, Trial Reels (which seems to be moving the needle, according to some authors chatting about it in Wide for the Win!)

The shift we talked about all week was the move from chatting with AI to having AI take action on your behalf — with your permission, your guardrails, and your rules. That’s what agentic means, and it landed for a lot of people. Let me remind you if feel like you’re behind — only 2% of the world is actually using AI in their business. 2%. You’re officially an early adopter living in the future.

I’m going to break a lot of what we covered in this newsletter over this series, but if you want the cheat code (and save weeks of waiting for me to write) then replay packages for all 15 sessions + access to Discord are available at https://aisummit.indieauthortraining.com.

We also have three things coming up on the calendar. On April 29 at 11am Central, I’m doing an overview of StorytellerOS for anyone who wants to see the full platform in action. On May 5 from 1-3pm Central, we’re running a hands-on session on Claude Cowork for Ads Management. And every Wednesday at 12 noon Central, I’m hosting an Author Automations Social Getting Started session if you want help setting up the social scheduler and connecting it to your workflows.

Register for any or all of the webinars at https://webinars.indieauthortraining.com

Last week we broke down the differences between Claude Chat, Cowork, and Code.

One of the biggest questions after that was a version of “okay, I get that there are three Claudes, but where do I actually start?” This newsletter is the answer. We’re going to walk through the differences between the desktop app and the web version, talk about what a terminal is (for those of you who need it), and then set up the two concepts that make Claude Cowork and Claude Code genuinely powerful: Skills and Connectors. Next week, we’ll go deeper into both of those.

Desktop App vs. Web: Start With the Desktop

If you only take one thing from this newsletter, make it this: download the Claude desktop app and use that instead of the website, if you can.

The web version at claude.ai is fine for chatting. You can ask questions, paste in text, get answers — it works the same way it always has. But when you move to Cowork (the part where Claude can actually do things on your computer), the desktop app is where everything lives. Cowork in the desktop app can read files and folders on your machine, create and edit documents in place, connect to your tools through connectors, run scheduled tasks, build “live artifacts” (which are dashboards, but better), and access the skills you build to teach it how you work.

The web version can’t do most of that because it doesn’t have access to your local files. It runs in what’s called a sandbox up in the cloud, which means it can’t see your desktop, your documents folder, or any of the configuration you’ve done on your machine. Your skills don’t transfer to the web automatically either — you can manually set them up there, but it’s not reliable and you’ll spend more time troubleshooting than actually getting work done.

During the summit, someone asked about the pros and cons of desktop vs. web, and my answer was pretty simple: if you’re just chatting, either one works. If you want Claude to actually do things for you, desktop app every time.

To get it, go to claude.ai, look in the bottom left corner for “Get apps and extensions,” and download it for your Mac or PC. You’ll want at least the $20/month Pro plan to access Cowork, and if you end up being a power user (you’ll know when you get there), the $100/month plan gives you a lot more capacity.

About That Desktop App I Used to Hate

I need to make a confession here because some of you have been reading this newsletter long enough to remember that I was not exactly a fan of the Claude desktop app a month ago, and I didn’t recommend Cowork at all. It was buggy. Cowork was unimpressive. I had real concerns about whether Anthropic was going to get this right.

They got it right.

The improvement in the last four to six weeks has been dramatic, and I’m not using that word casually. The desktop app went from something I tolerated to something I rely on for hours a day. I see the “Relaunch to update” notification on my screen so often that I’m starting to wonder if Anthropic’s engineering team sleeps.

They are shipping updates at a pace I have never seen from a software company in nearly three decades of working in technology, and every update makes it noticeably better.

If you tried the desktop app a month ago and walked away frustrated, go back. The version you tried and the version that exists today are almost unrecognizable from each other, and the version that exists tomorrow will probably be better than both.

Terminal Access: For When You Need Claude Code

Most of you will live in Cowork and never need to think about this section. But if you’re on a version of Windows that won’t run the desktop app, or if you want to graduate to Vibe coding with Claude Code (the more powerful sibling that can connect to your servers, databases, and basically anything with a login), you’ll need to use what’s called a terminal.

Quick aside: Yes, there is a version of Claude Code on the Desktop App. Yes, it’s pretty good, and you can use it for a lot of cool things. But not #allthethings. So if you’re not yet able to use Cowork, or want to unlock Code’s full potential, read on for hooking up the version that’s called Claude CLI (Command Line Interface).

For the GenX kids: you already know what this is, even if the word “terminal” doesn’t ring a bell. Remember DOS? Remember that black screen with the blinking cursor where you typed C:\> dir to see your files and C:\> OREGON.EXE to die of dysentery on the trail to Oregon? That’s a terminal. It’s a text-based way to talk to your computer by typing commands instead of clicking on icons. Your Mac has one built in (it’s called Terminal, creatively enough), and Windows has a couple of options called Command Prompt and PowerShell.

I use an app called Warp instead of the built-in terminal, and I recommend it if you’re going down this road. Warp is a modern terminal that doesn’t look like it was designed in 1987 — it has autocomplete, it’s easier to read, and it doesn’t make you feel like you accidentally hacked into NORAD. It’s free, it works on Mac (and Windows support is available too), and it makes the Claude Code experience significantly less intimidating. It has a built-in AI you can chat with to ask it to help you do things like, “Hey, install Claude Code for me” and then “Be sure I can run Claude Code everywhere on my computer.” and “Hey, can you make me a latte?”

Give Anthropic like maybe 20 minutes and that one might be a reality.

If you get stuck, drop a message in the Discord and we’ll help you get sorted. Or grab one of the upcoming sessions where we’ll walk through it live.

Projects: Your First Move in the Desktop App

Once you have the desktop app installed, the first thing to do is create a Project. Projects are how you bundle related conversations, files, and context together so Claude remembers what you’re working on across multiple chats.

Think of it this way: without a project, every conversation with Claude starts from scratch. It doesn’t remember your last chat, your preferences, or what you told it about your book series yesterday. With a project, you create a container where Claude has ongoing context about a specific area of your work.

You might create a project for your book launch, another for your newsletter, another for your backlist marketing. Inside each project, you can attach reference documents — your brand guide, your blurb, your series bible, whatever Claude needs to give you useful answers without you re-explaining everything each time.

During the summit, I walked through how my own projects are organized, and the main takeaway was that projects are the difference between Claude being a generic chatbot and Claude being an executive assistant that actually knows your business. If you’re just getting started, skip regular chat entirely and go straight to creating one project for the area of your business where you spend the most time.

The Two Things That Make Cowork Powerful

Cowork on its own is already useful — you point it at a folder, describe what you want, and it gets to work. But the features that turn it from a nice-to-have into something that can genuinely run parts of your business are Skills and Connectors.

Connectors are how Claude talks to your other tools. When you go into Customize in the desktop app, you’ll see a directory of approved connectors — Google Drive, Gmail, Google Calendar, Notion, Slack, Stripe, Zapier, n8n, and a growing list that gets longer every week. Connecting one is as simple as clicking “add” and signing into your account. Once connected, Claude can read from and write to that service without you having to copy and paste between windows.

If you’ve been reading this newsletter for a while, you might remember my deep dives on MCP servers. Connectors are the friendly, click-to-install version of that same concept. MCP stands for Model Context Protocol (you don’t need to remember that), and it’s the bridge that lets Claude talk to your apps without needing to understand the technical details of each one. The official connectors in the directory have been vetted by Anthropic, so they’re the safest place to start. For those of you who want to go further and connect things that aren’t in the official directory (like authorautomations.social), you can set up personal connectors, which we covered extensively during the summit and which I’ll walk through in detail next week.

One limitation worth knowing: the official connectors currently only support one account per service. So if you have multiple Google accounts (I have approximately 4,000, which is a me problem), you’re limited to connecting one through the official connector. There are workarounds for this using personal connectors and Claude Code, and I’ll cover those in a future installment.

Skills are how you teach Claude to do things your way. A skill is basically a set of instructions — a recipe card, if you will — that tells Claude how to handle a specific type of task with your preferences, your voice, your tools, and your standards baked in.

I have over 100 skills at this point, which is admittedly excessive. I have a developmental editor skill, a line editor skill, a continuity specialist skill, a pen name architect skill, a series architect skill, a copywriting skill, and on and on. Each one contains specific instructions about how I want that task done, which tools to use, where to put the output, and what standards to apply.

You don’t need 100 skills. You need one. Pick the task you do most often — maybe it’s drafting social media posts, or summarizing your reader feedback, or formatting your newsletter — and start a conversation with Claude in Cowork. Say “interview me about how I do this task, what tools I use, and how I want the output to look.” Claude will ask you questions, and when you’re done, you’ll have a skill you can save and reuse every time you do that task.

The beautiful part about skills (and this came up several times during the summit) is that you don’t have to go back into the settings menu to update them. If you’re chatting with Cowork throughout the day and you realize the skill is missing something, just say “hey, add this to my copywriting skill as a non-negotiable rule.” Claude updates the skill for you. You’re teaching it incrementally, the same way you’d train a real assistant — one correction at a time, in the flow of actual work.

Where This Series Is Headed

Next week, I’m going deeper on Skills and Connectors with specific walkthroughs for authors. I’ll show you how to set up your first connector (we’ll start with Google Drive since nearly everyone uses it), how to create a skill from scratch by having Claude interview you, and how the two work together to turn Cowork from a fancy chatbot into something that actually handles repeatable tasks in your business. I’ll also touch on a big one —how to manage some security and risk with connectors and skills.

If you came to the summit, next week’s newsletter will feel like a practical follow-up to what you saw in the MCP Servers and Skills session. If you didn’t make it to the summit, you’ll have everything you need to get started.

Reply and tell me: what’s the first task you’d want to teach Claude to do your way? I read every one.

—Chelle

I’ve been getting the same question from about forty different directions since Ireland: “What is Claude Code and why won’t you shut up about it.” (Fair.) So I made two things.

A 45-minute video where I break down all three Claude tools, do a live demo, and walk through the weekend I handed Claude Code my entire Notion task backlog and sat in my easy chair while it worked through 100+ tasks. It’s 45 minutes of me being thorough and slightly unhinged. That one’s for the people who learn by watching someone push buttons and gasp.

This newsletter, which covers the same ground in about eight minutes of reading. No buffering required.

If you consume both, there’s overlap — that’s on purpose, not early-onset memory loss. (Though ask me again in a few years.) Pick whichever fits your capacity.

Most of you have used Claude. You’ve opened claude.ai, typed a question, gotten an answer. Maybe you’ve used it to brainstorm plot points, draft a blurb, or ask why your Mailerlite automation isn’t firing. (It’s always the trigger. Always.)

That version of Claude — the chatbot in a browser — is one of three tools Anthropic makes now. The other two are where things get interesting for running your author business. And almost nobody in the indie author space is explaining them clearly, because the people who use them are mostly developers writing for other developers.

I’m an old-school programmer — DOS, COBOL, C++ back in the Y2K and Code Red Mountain Dew era. I stepped away from coding for a long time, and Claude Code brought me back. Not as a developer, exactly, but as a product manager who can build things again. That’s a distinction worth understanding, and I’ll come back to it later in this series. For now, let me translate the three tools into author-friendly language.

Hey, real quick: if you’re also into vibe coding, I have a Discord Channel to come hang out with a nerd herd: https://discord.gg/MZKACmjSjs

The Three Claudes, in Plain English

Claude Chat is the one you know. You type, it responds, you copy-paste the output somewhere useful. It lives in your browser or the Claude app. Think of it as a very smart coworker who can only communicate through sticky notes — you hand one over, they hand one back.

Claude Cowork is the one most of you should care about right now. It launched in January 2026 and it lives in the Claude desktop app. You open it, point it at a folder on your computer, and say “do this.” It reads your files, creates new ones, edits existing ones, and works through multi-step tasks on its own — checking in with you along the way. If Claude Chat is a coworker who communicates through sticky notes, Cowork is a coworker who has a key to the office and can actually sit down at the desk and do the work.

Claude Code is Cowork’s more powerful (and more intimidating) sibling. It runs in your terminal (the black screen with the blinking cursor that makes you feel like you’re in a 90s hacker movie). It can write code, run commands, create files, connect to your databases, manage your servers — anything you could do from a command line, Claude Code can do for you. This is the tool I used to build AmandaBot, authorautomations.social, and most of the infrastructure that runs my business. It is extremely powerful and it looks like The Matrix. (It’s also how I scrapped my entire Ireland presentation and rebuilt it in an afternoon, which is a story for another installment.)

Here’s why that distinction matters: in Chat, you are the bottleneck. You copy text in, you copy text out, you paste it somewhere, you format it, you save it. Every step is a manual handoff. In Cowork, Claude handles the handoffs. You describe the outcome, and it produces deliverables — actual files, in your actual folder, ready to use. In Code, you can do everything in Cowork, and more.

When You’d Use Which

This is the part I wish someone had spelled out for me six months ago instead of making me piece it together from developer Twitter threads.

Use Chat when you need a conversation. Brainstorming, Q&A, feedback on a blurb, “is this sentence grammatically correct or am I losing my mind.” Anything where the back-and-forth is the point.

Use Cowork when you have a task with a tangible output. You want a document created. Files organized. A spreadsheet built from scattered information. A batch of social media posts drafted and saved. Anything where you’d normally say “I need to sit down for an hour and just power through this” — that’s a Cowork job.

Use Code when you want to clone yourself. Chat answers questions. Cowork handles tasks. Code is where you build the systems that run your business while you’re not looking — the connectors that link Claude to your email, your website, your social platforms. The skills that teach it how you want things done. The agents that handle recurring work on your behalf. The hooks that trigger workflows automatically. This is how I built AmandaBot, and it’s how I run a publishing company, a training business, a magazine, and a social media platform from my iPad while traveling across two continents. Code is the infrastructure layer, and you don’t need a CS degree to use it. (I don’t have one. I have stubbornness and a COBOL background from the Reagan administration, which is arguably worse.)

Most authors will live in Chat and Cowork. That’s completely fine. Code is where you go when you’re ready to build the systems that let you stop doing the work and start overseeing it — and we’ll get there later in this series.

What This Actually Looks Like: A Real Example in Cowork

Full disclosure: my own book organization lives in StorytellerOS now, so I don’t actually need Cowork for this particular task anymore. But I wanted to test it with something familiar so I could show you what the experience feels like — and a messy book launch folder is about as universal as it gets.

I had a folder on my desktop with everything for an upcoming book launch. Cover images in three sizes. A blurb document. A series overview. A few versions of taglines I’d been noodling on. Comp titles in a text file. Release date, preorder links, ISBNs — scattered across multiple docs because I am a chaotic creature who organizes by vibes.

I opened Cowork, pointed it at the folder, and said: “Read everything in here and create a complete book launch media kit. I need a one-sheet with the cover, blurb, comp titles, buy links, and author bio. Then create a separate file with 20 social media posts in my voice — mix of announcement posts, countdown posts, and reader-question posts. Save everything back into this folder.”

Cowork read every file. It figured out which cover was the final version (because I’d named it “FINAL-final-v3-USE-THIS” like a professional). It pulled the blurb, matched it to the comp titles, grabbed the links, and produced a clean one-sheet as a markdown file and 20 social posts in a separate document. Saved them right into my launch folder.

I reviewed the posts, tweaked maybe four of them, and moved on with my life.

The same task in Chat would have meant: open each document, copy the contents, paste them into the chat window one at a time (hoping I don’t hit the context limit), explain what I want, get the output back as text in a chat bubble, then copy it out and paste it into a new document, format it, and save it myself. Every handoff is manual. Every step is me.

In Cowork, I described the outcome and went to make coffee. (The espresso machine at home works. Unlike certain Regus locations I could name.)

Getting Started

Cowork is available now in the Claude desktop app on Mac and Windows. You’ll need a Claude Pro subscription ($20/month) at minimum — it’s also available on Team and Enterprise plans.

To try it:

1. Download the Claude desktop app from claude.ai if you don’t have it

2. Open the app and click “Cowork” in the sidebar

3. Choose a folder to give Claude access to (start with something low-stakes — a folder of drafts, your book metadata, your launch files)

4. Give it a task in plain English

5. Watch it work and approve or redirect as it goes

Start small. “Organize these files by project” or “read this manuscript and create a chapter-by-chapter summary” or “build me a spreadsheet of all the ISBNs, titles, and publication dates from these documents.” Once you see what it can do with your own files, you’ll start seeing Cowork-shaped tasks everywhere.

Full Disclosure: It’s New, and It Shows

Cowork is genuinely useful and I would not be writing a multi-part series about it if I didn’t believe that. I also need you to know that it is sometimes buggy as hell.

It doesn’t always do what you ask it to do. The Chrome browser takeover feature is a bit of a nightmare and will eat your tokens like chocolate chip cookies the day before you start Keto. It can be frustratingly dumb about connecting to all your tools, and while the interface is friendly enough, the troubleshooting and initial setup might make you want to toss your computer into the nearest body of water. (If you’re near a lake, I recommend a good overhead throw. Very cathartic.)

I’m telling you this now so you don’t hit your first wall and think you’re doing it wrong. You’re not. The tool is young. These are all pitfalls I’ll go over in depth across this series and at the AI Summit — what to expect, what to work around, and what to just wait for Anthropic to fix.

What’s Coming in This Series

This is Part 1. Here’s where we’re headed:

Part 2: Cowork for Your Social Media — I’ll walk through using Cowork to generate and schedule social content for your books, including how it connects to authorautomations.social and the Claude plugin that lets you manage your posting calendar in plain English.

Part 3: Building Your First Claude Skill — Skills are reusable instructions that make Claude better at specific tasks — and they work in Cowork, not just Code. I’ll show you how to create one for your own writing or business workflow.

Part 4: Connectors — Giving Claude Access to Your Tools — Connectors are how Claude talks to the rest of your business. Email, Google Drive, your calendar, your social platforms. I’ll walk through what’s available, how to set them up, and which ones are worth your time right now.

Part 5: Clone Yourself with Claude Code — This is where you go from doing the work to building the systems that do the work for you. Connectors, skills, agents, and the infrastructure that lets you run a business from your iPad. I’ll show you what’s possible and where the on-ramp is. (I promised you I’d redo the walkthrough I recorded a few weeks ago with the new features, and I will.)

Part 6: The Agentic Author Stack — How all of this fits together. Chat, Cowork, Code, MCP servers, authorautomations.social, and the tools that connect them. The full picture of what it looks like to run an author business with AI agents handling the operations.

If there’s something specific you want me to cover, reply to this email. I read every one of them. (AmandaBot does too, but she’s polite enough not to spoil the answers.)

The summit is April 21–22. Day 1 is free. If you’ve been on the fence, this series will give you a head start on everything we’ll cover in the paid sessions on Day 2. Details at aisummit.indieauthortraining.com.

I'm back in Texas this week, reunited with my espresso machine and a desk that doesn't require a subscription. Part 2 drops in a few days and not weekly. In the meantime, go download the Claude desktop app and point Cowork at your messiest folder. You'll either be impressed or you'll want to throw your laptop in a lake. Either way, you'll have something to tell me about.

I’m writing this from Canterbury, which is where I usually post up after a string of conferences on this side of the Atlantic. (After Author Nation it’s my mom’s place in Lake Tahoe. After Europe it’s Canterbury. I have decompression spots the way some people have lucky socks.) This is where I process what I’ve learned, synthesize it, and figure out what it means for the work I’m doing.

I have a lot to process this time. And some of it changes this newsletter.

Ireland Was AI-Positive in a Way I Haven’t Seen Before

At Ireland’s Publishing Show, several sessions — including mine — were explicitly labeled as AI sessions. Which meant the room was self-selected: people who showed up had moved past “should we” a long time ago and were ready to get into the how. We skipped the throat-clearing and went straight to the good stuff, which was honestly a relief after years of opening every talk with kindergarten rules: no biting, no hitting, no crying, and eyes on your own paper.

One speaker, Matthias Kadott, did something I loved: he created a brand new pen name specifically as a proof of concept for the conference. Started from scratch, wrote a book, deployed it to market including AI translations — and netted over €7,000 in a single month. As a demo. With receipts.

I don’t know about you, but I looked at that number and thought: okay, the landscape has genuinely shifted. The tools are here. The playbook is different. And people are already running it.

I Radically Changed My Own Presentation

I was originally planning to teach my usual Make.com and Zapier workflows for managing social media. That’s been my bread and butter for two years. It’s what this newsletter was built on.

I scrapped it (with permission) and taught social media using agentic AI instead.

It felt a little strange, honestly. Like showing up to Thanksgiving with a completely different casserole and hoping nobody asks what happened to the old one. But I couldn’t in good conscience stand in front of a room and teach a three-step Make.com scenario when I know there’s a faster, more powerful way to do the same thing — and when I’ve already built it.

I still have Make.com and Zapier workflows running in my business. They’re solid for scheduled data transfers and on-demand tasks. I’m not throwing them away. But when I’m reaching for a tool now, my first choice is agentic AI and vibe-coded apps. That shift happened gradually over the last year and then very quickly over the last few months.

The Thing I Built That Replaced the Workflow

Okay, here’s the concrete version.

I used to have a three-step Make.com workflow that required Airtable as a data source to create social media content from my books. It did the job. I taught it at conferences and in the Hub. People built their own versions of it.

I replaced it with authorautomations.social.

It creates 30 days of content in minutes — TikTok videos, carousels, captions — from your own books. Posts to 13 social media channels in your voice, with your content. The whole Airtable-to-Make pipeline that used to power this is gone. You don’t need it anymore.

It’s free for Founder subscribers and $29/month for everyone else.

And here’s the part that still makes me grin every time I demo it: there’s a Claude plugin for it. Which means you can sit in Claude and say, “Create 20 TikToks and post them to TikTok over the next week to promote Book 1 before Book 2 launches.” Claude reads the context on your own machine, generates the content, and uses authorautomations.social to schedule and post it.

You describe what you need in plain English. It happens. While you go write. Or nap. Or finally text back the friend you've been "meaning to reply to" since November. (Just me?)

This Newsletter Is Changing

I’ve been wrestling with this for a while, and Ireland made it click.

Agentic AI is the single greatest opportunity for indie authors to build million-dollar creative careers. It can manage the entire operational side of your publishing business — marketing, social, newsletters, sales tracking, metadata, all of it — while you write. “Bots do the boring so you can do the brilliant” was always the tagline. It’s just that now the bots are significantly less boring and significantly more capable.

I’m not pivoting into writing-with-AI territory. There are stronger voices in that space and I’m happy to point you toward them. But Author Automations is now where you come to learn how to run your author business using agentic AI.

The name probably needs to change at some point. But 2026 is the year of contraction and I am exercising restraint for possibly the first time in my life. (Joanna Penn did just tell me I have a “popcorn brain,” so the restraint is relative.)

What Stays

The Hub isn’t going anywhere. Plugins, extensions, templates — all still there, all still accessible. I’ll still post workflows for Make.com, N8N, and Zapier when they’re the right tool for the job. There will just be fewer of them, because most of the time I’m reaching for something else now.

What’s Coming

A lot of webinars, and virtual live events. I’ll be back in Austin next week where the bandwidth is much better, and my espresso machine is about to get a workout. I’ll post a full calendar when I’m back, but for now you can plan on:

What’s all the fuss about Claude Code & CoWork? Claude Code and Cowork keep coming up and nobody's explaining them well. I'm going to fix that — with a walkthrough built for authors, not developers (I recorded this a few weeks ago but things have changed and I want to update it with the new cool stuff)

authorautomations.social Deep Dive + the Claude Plugin — A full walkthrough of the app, including the Claude plugin that lets you manage your social media with natural language commands. I’ll show you exactly what it looks like to tell Claude to promote your book and watch it go do it.

Agentic AI for Book Launches — I’m going to map out what a launch looks like when AI agents are handling newsletter swaps, research, social scheduling, email sequences, ad monitoring, and sales tracking. Real examples from real launches, not a PowerPoint fantasy.

AI for Your Author Business Summit Preview — The summit is April 21–22. Day 1 is free. Day 2 is the deep dive into agentic AI and Claude Code. I’ll be sharing what to expect and how to get the most out of it.

The mix going forward will be tutorials and industry commentary — practical how-tos alongside the bigger conversations about where this industry is headed.

One More Thing

I won’t be mad if these changes mean it’s time for you to unsubscribe.

I mean that. If you came here for Zapier recipes and Make.com walkthroughs and that’s still what you need, I respect that completely. The Hub content isn’t disappearing and those tools still work. But this newsletter is going somewhere different now, and I’d rather tell you that directly than slowly drift and leave you wondering what happened.

For those of you sticking around: the conversations I had in Ireland, at LBF, at ALLi, and in Savannah all confirmed something I’ve felt for months. The authors who figure out agentic AI in the next twelve months are going to operate at a completely different level. I want this to be the place where you figure it out.

Canterbury coffee remains deeply mid. But the WiFi works and nobody’s asking me to be on a panel, so I’ll take it.

—Chelle

Exactly six years ago, I was sitting in a B&B in Northern Ireland trying to figure out how to get back to the States. The world had other plans, and I decided it was safer to stay put. I stayed for four months. During that weird, suspended stretch of time, a group of author friends came together to write, share meals, and share lives.

Six years later, we’re still doing it. I’m currently doing bougie couch surfing through their guest rooms and kitchens, and Princess Elowyn — the most brilliant child ever™ — gets to FaceTime with her aunties while Mimi “checks on her castles”. Irish, Scottish, English, Canadian, German, and Kuwaiti aunties. Best. Job. Ever.

But between the home-cooked dinners and the pub crawls, which are surprisingly hard with a cane and a torn ligament, I’ve also been at the Savannah popup, London Book Fair, ALLi’s Indie Author Lab, and now Ireland’s Publishing Show. And I can tell you this: every indie author I’ve talked to is having some version of the same conversation.

Discoverability and longevity.

Those two words came up everywhere. In panels. In hallways. Over dinner. In those hushed side conversations where someone grabs your elbow and says, “Okay, but seriously. What do I do?”

So this week, I want to tell you what I’m seeing out here — and where I think authors need to be having these conversations instead of spiraling alone.

(Spoiler: not Threads.)

Authors Are Tired in a Very Specific Way

Authors have always been tired. That’s not new.

But this is a specific flavor of tired: I’m doing seventeen jobs instead of seven and the goalposts keep moving tired.

Tony Lee said something at the Author Nation conference in 2025 that has stuck with me ever since: we’re publishers that write, not writers that publish. I quote it constantly now because that framing changes how you think about all of this.

If you’re running a publishing business, the path to longevity stops looking romantic and starts looking operational. The authors I keep seeing thrive over the long haul aren’t superhuman. They’re the ones who have systemized their workflows, delegated where they can, and automated what they can. The ones trying to manually carry every moving piece themselves are flaming out by year three.

Translations. Wide distribution. Ads on platforms that change their interface every time you finally learn the old one. Social media across channels you can barely keep straight anymore. Direct sales. Email sequences. Print logistics. Newsletter swaps. And now everyone’s standing on stages telling authors they also need to learn AI.

(I say this with full awareness that I am one of the people standing on those stages. The irony is not lost on me.)

At the Savannah popup, I sat with authors who are executing at a really high level — gorgeous covers, deep backlists, consistent releases — and the conversations weren’t about craft. They were about noise.

The rising volume of content on every platform. The suspicion. The friendly fire. Authors being accused — sometimes falsely — of using tools or processes they didn’t use. A lot of people are guarded right now. About their workflows, their teams, their tech, their everything. Because they just don’t want the smoke.

I get it.

But it’s also making it harder for authors to learn from each other, which is exactly what we need more of right now.

London Felt Different This Year

At London Book Fair, something had shifted.

For years, indie publishing showed up as the scrappy younger sibling. Trad was the default, and we were still proving ourselves. This year, that wasn’t the vibe anymore. We had a much bigger presence. A much louder voice. More panels, more attendees, more genuine curiosity from the trad side about what we’re doing and how we’re doing it.

That doesn’t mean the anxiety wasn’t there. Retailer changes came up constantly — in panels, in the exhibitor halls, at the pub afterward. People were speculating about policy changes, algorithm shifts, AI detection, discoverability. Whether the platform that built most of our careers is still the one we can count on long term. Nobody had clear answers.

But the energy wasn’t defeatist. It was more like: okay, we’re here now, we’re not going anywhere, and we’re going to figure this out together.

One bright spot that kept surfacing: translations.

That conversation has moved way past should I translate? That’s settled. Authors are earning real money in translated markets now. Tools like ScribeShadow, plus things like James Blatch’s AI for Translations course, are shrinking the speed-versus-accuracy tradeoff that used to make people hesitate.

The conversation now is: how do I do this well? Which languages? Which markets? How do I maintain quality at scale?

(For what it’s worth, I’m firmly in the “test two or three languages before you go full United Nations” camp. Scaling something that doesn’t work just gives you more of what doesn’t work, faster.)

And underneath all of it — translation, discoverability, direct sales, ads, AI — the question that kept surfacing was the only one that actually matters:

What do readers want?

What are they reaching for? What are they buying? What are they sticking with? And how do we make sure we’re in the mix when they go looking?

That’s the question that should be driving every strategic decision, and I was genuinely encouraged to hear it being asked that directly.

ALLi Had the Most Useful Answer

ALLi’s Indie Author Lab was where the conversation got the most grounded.

Orna Ross, Melissa Addey, and Joanna Penn were all refreshingly clear: figure out your own path.

Get clear on your priorities. Figure out where your energy actually comes from. Simplify your workflows around that. Do the right things for your business, on purpose, and let go of the rest.

That message landed because it wasn’t about adding more. It was about subtracting.

Direct sales and human interaction came up over and over. Authors selling direct — through their own stores, at live events, on TikTok, through email, through actual relationships with readers — are the ones seeing the needle move. Not because it’s trendy, but because it creates something we should all be thinking about more seriously:

Ownership of the reader relationship.

Because if your entire business depends on an algorithm you don’t control, you don’t have a business. You have an arrangement that works until it doesn’t.

I left ALLi thinking about how many authors I know — myself included — who keep adding tactics instead of removing friction. Simplification is harder to sell from a stage than the next shiny thing. But it’s probably the advice most of us actually need right now.

Side note: the honor of the day was Joanna Penn telling me I have a “popcorn brain” like hers— ideas always popping. And yes, I did try to register the domain name before she’d stepped off the stage. To be fair, I do have a lot of ideas. But 2026 is the year of contraction, so I let it go. (For now. Don’t look at me like that.)

The Upload Fee Debate

Speaking of James Blatch, he wrote a Substack piece recently that got the comments section so spicy it should’ve come with heartburn meds.

His idea: Amazon KDP should charge $300 per title to upload a book.

The fee would be refundable — basically a reverse advance you’d earn back after hitting $500 in royalties. The goal would be to kill the churn-and-burn model of low-effort AI upload farms flooding the platform with junk.

He’s not wrong that we have a problem.

The three-books-per-day upload limit Amazon quietly introduced was basically an admission that the floodgates were open. A speed bump doesn’t fix a flood.

James raised the question in good faith, and I think it sparked exactly the kind of industry debate we need more of.

But I don’t think $300 is the answer.

Fees to publish aren’t new. BookVault charges them. Ingram used to charge them. Professional organizations like ALLi and IBPA often offer ways to offset costs like that. The concept itself isn’t foreign.

KDP’s entire original value proposition was that you didn’t need permission or upfront capital to publish. A $300 toll booth changes that equation — the number itself may just be too high to avoid unintended consequences. The idea has merit. The execution needs more conversation, which he absolutely said from the jump.

My bigger issue is that this is still treating the symptom.

The flood of low-quality AI content is fundamentally a discoverability problem.

Amazon has the data to solve discoverability. They know sell-through. They know completion. They know returns. They know which books readers start, finish, review, and buy through. They absolutely have the tools to surface books people actually engage with instead of books that just got uploaded in bulk.

They could solve a lot of this without charging authors a dime. Universally, I’d love to have the conversation with all retailers about giving us more access to data. I doubt that’s coming.

That said — I’m not entirely against some kind of friction. Something that makes churn-and-burn uploads less profitable without crushing legitimate authors.

Why I’m Still Optimistic

Despite all of this, I still believe a solo author can build a million-dollar career.

I believed it before AI, and I may believe it more now.

Not because AI is magic. But because the operational load that used to require hiring three to five people can increasingly be handled by well-built automations and AI agents.

And I mean actually handled.

This is why I built StorytellerOS.

I kept running into the same problem: I’d open a dashboard and think what the heck am I looking at? Sales from KDP, Draft2Digital, and my direct store in three different places. Social media performance scattered across eleven channels. Newsletter stats in FluentCRM. Book metadata in Airtable. Ad spend in four platforms. None of it talking to each other, all of it requiring me to manually pull numbers together before I could make a single decision. Classic business intelligence problem — and AI is genuinely good at those.

So I built the thing I needed. StorytellerOS has studios for writing, marketing, social media, and sales — and agentic AI running through all of it. It manages exports, drafts newsletters, schedules social, monitors sales data, and pings me when something looks off. A weird drop in page reads, a spike in refunds, an ad campaign bleeding money while I’m not looking — I used to catch that stuff during a ninety-minute Monday morning spreadsheet ritual fueled by an inadvisable amount of caffeine. Now it just happens in the background while I write. Or stare out the window. (At least I’m staring productively, like a character in an Austen novel.)

The authors who figure out how to set up systems like this — whether they build their own, borrow someone else’s, or use something like what I’ve built — are going to have a real operational advantage over the next few years. Because the author who spends four hours a week on admin and the author who spends zero —but still has the data and strategy they need —are going to have very different creative output by December.

The conversation in this industry has moved past should authors use AI. Most people have landed somewhere on that by now. The real question is how do we use it without losing our voice, our judgment, or our ethics?

That’s a much better question.

And yes, I also hold the uncomfortable stuff. The environmental costs are real. AI slop is making discoverability worse. Guardrails matter. Intent matters. You can use these tools thoughtfully and still think critically about their costs.

Where These Conversations Are Actually Happening

I’m speaking at Ireland’s Publishing Show this week, and I’m genuinely excited to get into these questions with people who are thinking about them in practical, nuanced ways.

(Also excited to not be on an airplane for five consecutive minutes, but that’s a separate newsletter.)

If you want to go deeper, the AI for Your Author Business Summit is April 21–22, and that’s right around the corner, so at least 11,000 things will change between now and then.

Day 1 is free and packed with practical, implementable takeaways. Whether you’ve never touched an AI tool or you’ve already got workflows duct-taped together in Make.com, you’ll leave with something useful you can put to work the same week.

Day 2 is where we go full nerd: agentic AI, Claude Code, building agents connected to your actual data, and automating the operational grind that keeps you from your desk. I’m bringing examples from authors with real businesses, real revenue, and real readers — not hypothetical “what ifs.”

And if you’re wide — or even just wide-curious — the Wide for the Win community is closing in on 24,000 members, and the conversations happening in there are exactly the ones I think need to happen more: respectful, useful, long-game thinking.

I’ll be back next week with dispatches from Ireland and probably some strong opinions about Irish coffee.

Fair warning.

—Chelle

I’m in city one of a four-week stretch across four cities and two continents. Publishing popup here, London Book Fair, ALLi Author Lab, Publishing Show in Ireland, then back to the US sometime around the end of March. And this trip I’m doing one European airline rated carry-on and a backpack and staying in places with laundry facilities. Because I absolutely will NOT use a coffee maker to wash my unmentionables. #Gross

My business will keep running the entire time. Not because I hired a team of ten, but because I spent the last year building systems with clear boundaries — and the boundaries are the part nobody teaches you.

What Actually Happened This Morning

I want to show you what this looks like in practice, not the polished “I automated everything and now I sip espresso on a yacht” version, but the real one where I’m dictating newsletter drafts on my iPad in a co-working space with no functioning coffee. (Did I mention the coffee machine? I’m going to keep mentioning the coffee machine.)

One of our writers at Indie Author Magazine emailed to say her author page URL contained her personal email address and she was getting spam because of it. The URL looked like `/author/herauthorgmail-com/` — not great.

My old workflow for something like this:

Read the email and feel mildly guilty about it. Mentally calculate the effort involved (research Ghost’s database structure, SSH into the server, find the MySQL credentials, change the slug, add a redirect so indexed links don’t break, restart the service, reply to the writer). Assign it a priority somewhere between “important but not urgent” and “I’ll get to it when I have a free afternoon.” Punt it. Repeatedly. For weeks, probably months, because there’s always something more urgent. Feel guilty about the punting. Punt the guilt too.

I know this about myself. The task itself might take an hour of focused work, but the mental overhead of context-switching into sysadmin mode — remembering how Ghost’s database works, finding the right credentials, getting my brain into the technical headspace — that’s the real cost. Not the doing. The getting ready to get ready to do the thing.

Instead, AmandaBot handled it. For those of you who haven’t met her, AmandaBot is a semi-autonomous AI agent that lives in Telegram on my phone. I text her like I’d text a human assistant, and she handles things. Not theoretical things. Real things, today, while I’m sitting in a Regus with no functioning coffee. (Last mention, I promise. I’m lying. It won’t be the last mention.)

She saw the email, understood what needed to happen, checked with me, connected to the server, changed the slug in the database, added a permanent redirect from the old URL to the new one, restarted the service, verified everything worked, and emailed the writer back to let her know it was done. If you’ve emailed me in the last few months, there’s a decent chance AmandaBot is the one who replied. She has better response times than I do. (Lower bar than you’d think.)

The whole thing took about two minutes. I didn’t look up a single credential. I didn’t SSH into anything. I didn’t open a terminal.

You might be saying to yourself that this is WAY too technical than you will ever need.

But the same concept applies. You’re in flow state writing, and then you need to stop and look up an ASIN for someone real quick. You get an email from a reader asking where audiobook 3 is, because 4 just dropped. You’ve been putting off figuring out DMARC for your newsletter, or updating your website with new books. Every one of those is a context switch — and every context switch costs you twenty or thirty minutes of momentum you have to rebuild afterward.

Two minutes versus an hour is nice, sure. But the real value is that I never had to leave whatever I was actually working on, load an entirely different set of technical knowledge into my brain, do the thing, and then try to remember where I left off. Multiply those switches across a dozen small tasks per week and you’ve lost most of your productive writing time to stuff that isn’t writing. (And you wonder why the book isn’t done.)

The Part That Actually Matters

Here’s what I’ve noticed about the AI conversation in publishing: most of the education focuses on what AI can do. Make images. Write copy. Generate ideas. Chat, copy, paste back to your other window.

And this is where agentic AI — Claude Cowork, Claude Code, Manus, even Notion’s AI features — gets interesting. Instead of bouncing between your chat window and whatever you’re actually working in, these tools operate inside your workflow.

I was talking with a friend yesterday about her processes, and she has a human assistant. I asked her how she delegates things, and it’s remarkably similar to what I do with AmandaBot. She creates a document with instructions (a standard operating procedure), lists the repeated tasks she wants done (skills), and defines the desired outcomes (success metrics). With agentic AI, you do the same thing. Frameworks, expected outcomes, and measurements for success. The setup process is almost identical — the difference is that one of them doesn’t need health insurance. (Sorry, AmandaBot.)

What we’re not talking about enough is how to set up the boundaries. How to give an AI agent enough access to be genuinely useful without giving it enough rope to cause real damage. How to structure your business so that an agent can handle some tasks while being completely locked out of others. How to build trust incrementally, the same way you would with a human.

AmandaBot operates with the same permissions I’d give a competent intern on their first week. They can read emails and respond to straightforward requests. They can make technical changes to websites and systems I own. They can look up financial information — checking whether a payment went through, pulling a report — but they can’t move money, process refunds, or change billing settings. They can’t send emails to my whole list — drafts, yes. Send, no. They can’t publish content without my review. They can’t make promises on my behalf that involve commitments I haven’t approved.

That boundary hasn’t moved, and I don’t plan to move it until I’ve seen enough consistent behavior to trust a wider scope. Same way you’d promote a real employee — gradually, based on demonstrated reliability. (AmandaBot’s annual review is pending. I’ll let her know how it goes.)

If You Want to Learn How to Build This

That’s actually what I’m spending the first two sessions of Day 2 of the AI for Author Business Summit on in April. The first session, “Clone Yourself with Claude Code,” is about the philosophy — what makes a good candidate for delegation versus what you should keep your hands on. How to think about risk and reversibility. The stuff that determines whether your AI assistant actually helps or eventually sends an email you have to apologize for.

The second session, “MCP Servers & Skills: How to Control Your Empire,” is the mechanical how. I’ll show you exactly how I built AmandaBot’s permission system — which tools she can access, which accounts she can read versus write, how I structured the guardrails so they’re specific enough to be meaningful but flexible enough that she can actually get work done. It’s the session I wish someone had taught me before I spent months figuring it out through trial and error. (So many errors. So much trial.)

Day 1 of the summit is completely free, and you’ll walk away with real tools on your machine. Day 2 is the paid upgrade for authors who want the full automation layer. Details and registration at aisummit.indieauthortraining.com.

Get Thee to A Conference

James Blatch and Cissy Mecca said something on The Self Publishing Show this week that I want to echo: there is nothing better than in-person conferences.I may be a little biased because I’m here with them, but it’s still true.

The person sitting next to you at lunch who mentions the one tool that solves the problem you’ve been stuck on for six months. The hallway conversation where someone describes their workflow and you realize you’ve been overcomplicating yours by about 400%. The energy of being in a room full of people who take this business as seriously as you do. You can’t automate serendipity. (I’ve tried.)

If you’re on the fence about attending a conference this year, go. Large conferences like Author Nation and SPS Live are amazing for learning new skills and meeting with your business partner vendors. Small conferences with like-minded authors are great for accountability and personal leveling up. Budget for it like you’d budget for ads or a cover designer — treat it like a business expense, not a splurge. And if you’re at London Book Fair or the Publishing Show in Ireland, come find me. I’ll be the one looking for an outlet and pretending to like tea when it’s offered so I don’t start an international incident through sheer rudeness.

What You Can Do This Week

Before your next trip — conference, vacation, even a long weekend — write down every task that would pile up while you’re gone. Not the big projects. The small maintenance stuff. The emails that need responses, the social posts that need scheduling, the website updates you keep meaning to make. (You know the ones. They’ve been on your list since October.)

Pick one and automate it before you leave. Just one. Set up a workflow, hand it to an AI tool, schedule it in advance — whatever gets it off your plate. When you get back and that one thing is handled, you’ll understand why I keep talking about this stuff. It’s not about replacing yourself. It’s about freeing up the version of yourself that actually writes books.

I’ll be writing from London next week, from the Regus in Kensington that better have a functioning coffee machine. Decaffeinated Chelle is just not who you want to hear from, promise.

Chelle

I sat down to write this week’s newsletter and realized I have no fewer than six things to tell you about, which means one of two things: either I’ve been wildly productive or I’ve lost all sense of pacing. (It’s both. It’s always both.)

So here’s what’s in this issue:

• Founder and Pro Subscribers now have free access to authorautomations.social (the social scheduler I built for authors, not marketing bros)

• We have a Discord now (finally)

• There are five new workflows on the Hub

• I recorded a full video walkthrough of StorytellerOS so you can actually see the thing I’ve been building instead of just hearing me talk about it, and

• I’m announcing the AI for Author Business Summit — which deserves its own section below because I have feelings about it and also logistics

• Apparently I’m coming to Boston because you guys voted weird

I considered splitting this into multiple emails. Then I remembered you’re authors. You read 300-page books in a sitting. You can handle a long newsletter.

As the cool YouTubers say, Let’s get into it.

AI for Your Author Business — Free Virtual Summit, April 21–22

This one gets top billing because there’s a countdown clock ticking.

I’m hosting a two-day virtual summit on April 21–22 called AI for Your Author Business, and Day 1 is completely free. Not “free with an asterisk” free. Not “free but we held back the good stuff” free. Actually free, and you’ll leave with working tools on your machine.

Here’s why I’m doing this: I don’t teach writing with AI, unless you count copywriting. The entirety of my business is automating all the stuff that keeps me from doing the creative. This summit focuses on using AI to run your author business — marketing, operations, content creation, email, social scheduling, all of it. The stuff that eats your writing time.

Day 1 (Free — April 21, 10am–3:30pm CDT) is AI Marketing Quick Wins. You’ll walk away with a publishing command center in Airtable (the same template I use daily), marketing copy techniques that actually sound like you wrote them, social graphics and a book trailer you made yourself, and an automated workflow that turns one post into a week of social content. Everything is built live on screen so you can follow along.

Day 2 (Paid Upgrade — April 22, 10am–4pm CDT) is Clone Yourself: The Agentic AI Day. This is where it gets wild. I’ll walk through Claude Code installation and setup live (step by step, no developer background needed), MCP connections that let Claude Code talk to your Gmail, Google Drive, calendar, and Airtable, my actual inbox zero system across six email accounts, newsletter workflows that cut production time to 10 minutes, WordPress management without touching the dashboard, and a full content repurposing pipeline where one newsletter becomes a YouTube video, cross posted on Medium and my own sites, indexed in Google in less than 24 hours, and 26 pieces of social content — posted automatically.

Tickets:

• Free — Day 1 access + 30-day replay + StorytellerOS Lite Airtable template

• $97 Standard — Both days + 30-day replays + StorytellerOS Lite template

• $297 VIP — Both days + 365-day replays + Founder Access to Author Automations (includes weekly calls) + the full StorytellerOS Airtable template with financial tracking, reviews, project management, tasks, and story bible

Register here: https://aisummit.indieauthortraining.com

Day 1 is free and I’m not sandbagging. If you only ever attend the free day, you’ll still leave with more functional tools than most paid workshops deliver. Day 2 is for authors who want the full automation layer — the systems that let AI handle marketing operations so you can get serious hours back in your week.

StorytellerOS — Now You Can Actually See It

I’ve been talking about StorytellerOS for months. Describing features, explaining the architecture, generally waving my hands around while saying “trust me, it’s cool.” Which is not exactly a compelling product demo.

So I recorded a full video walkthrough:

The short version for anyone who hasn’t been following along: StorytellerOS is the platform I built because I got tired of running my author business across 47 browser tabs. It started as a tool for me and my writing group (we’ve met almost every day since 2020), and it grew into something bigger.

Project Studio is your publishing command center. All your book files, ASINs, ISBNs, pricing, links, translations, and audiobooks in one place. Enter an ASIN or ISBN and it fetches everything it can. Import Word docs, use the writing studio with goals and time tracking, and manage tasks and finances without switching apps.

Marketing Studio is your customer hub. Email subscribers, lists, newsletters, SMS, WhatsApp, blog posts to WordPress or Shopify, and review management — including automated review requests.

Social Studio is a content calendar, post creation tool, and scheduler for thirteen platforms with full image, carousel, and video creation.

And the part I know some of you are wondering about: StorytellerOS works fully without any AI. Every feature, every tool, no AI required. For those who choose to use it, you bring your own API keys, pick your own model, and your data stays yours. We don’t store it on our servers. The only things in our encrypted database are your Stripe billing details and your encrypted API keys. That’s it.

The video covers all of this in detail. Go watch it, then come tell me what you think in the (brand new) Discord.

Paid Subscribers Now Have Access to authorautomations.social

Author Pro and Founder-tier subscribers now have full INCLUDED access to authorautomations.social, the social media scheduling platform I built specifically for authors.

If you missed the original launch last fall: authorautomations.social is a standalone social media scheduler that posts to Facebook, Instagram, Threads, X, TikTok, Pinterest, LinkedIn, and Bluesky from a single dashboard. You log in, connect your accounts, queue up your posts, and walk away.

I built it because I needed a way to post to socials from make.com with a queue instead of hand-scheduling every post on every platform. You want different messages for different platforms because your LinkedIn audience does not care about your character’s dramatic coffee shop confrontation (but your BookTok audience absolutely does).

The part that makes this especially powerful for anyone already using the Hub: authorautomations.social has an API that works with Make.com, Zapier, and n8n. You can wire it directly into your existing workflows. Your RSS feed picks up a new blog post, your workflow formats it for each platform, and authorautomations.social schedules and posts it without you opening a browser. Connect it to your book launch workflow and your cover reveal goes out to eight platforms simultaneously while you’re doing literally anything else.

Paid subscribers get it included with their existing plan. If you’re not a paid subscriber but you want a solid social scheduler without the automation stuff, you can grab authorautomations.social as a standalone product for $39/month at https://authorautomations.social. It does everything Buffer and Hootsuite do, except it was built by someone who actually publishes books for a living.

Or, and I’m not joking, Substack is an absolute weirdo about pricing changes, so I haven’t increased anything here. I’ll let you do the math because I hate math.

Author Automations Now Has a Discord

I’ve wanted a real-time chat space for this community for a while. The Hub’s community forum works well for longer discussions and troubleshooting threads, but sometimes you want to ask a quick question, share something you found, or commiserate about why Make.com changed its interface again.

The Author Automations Discord server is live and open to everyone — free subscribers, paid subscribers, mastermind members, and anyone who wandered in from a conference talk. There are channels for automation help, general tech chat, and the kind of random conversations that happen when you put a bunch of nerdy authors in a room together.

Join here: https://discord.gg/rWYYdCSPHk

It’s open, it’s free, and you’re welcome whether you’ve been automating for years or you just learned what Zapier is last Tuesday.

Full disclosure, I’m traveling a LOT in the next month, so you’ll have to talk to one another more than me. Grace is in there to moderate, so please play nice. I’ll be in Savannah for a mastermind, then London for London Book Fair, and the Alliance of Independent Authors Author Lab, then Ireland for the Publishing Show, so say hello if you’re around!

Five New Workflows on the Hub

Every workflow from this round is ready to import into your own n8n setup, and they’re on the Hub now with full documentation and tier tags.

1. Daily Podcast Summary & Email Digest This workflow subscribes to any podcast feed and sends you a daily email with AI-generated summaries of new episodes, including key takeaways, timestamps, and action items. If you’re trying to keep up with publishing industry podcasts (or craft podcasts, or marketing podcasts, or that one true crime show you’re embarrassed about), it watches your subscriptions and sends you a morning digest that takes about two minutes to read instead of the 60 it took to record. I’ve been using this for about a dozen feeds and it has genuinely changed how I consume audio content. Instead of a backlog of episodes I feel guilty about, I scan the summaries and only listen to the ones that are actually relevant to what I’m working on.

2. AI Book Keyword & Category Research Finding the right keywords and categories for your book shouldn’t take all weekend. This workflow uses AI to generate seed keywords from your book description, expand them into long-tail variations, and organize everything in a spreadsheet that’s ready for your metadata on Amazon, Draft2Digital, or anywhere you publish. You paste in your blurb, and it hands you back a structured spreadsheet of keyword candidates you can actually use instead of staring at the Amazon category tree until your eyes cross.

3. Reader Feedback Sentiment Analysis This workflow collects reader feedback from a Google Form, runs it through AI sentiment analysis, and logs everything in a color-coded Google Sheet so you can spot trends, fix pain points, and double down on what readers love. The color-coding alone makes this worth installing — you can glance at the sheet and immediately see where things are going well and where something needs your attention without reading every individual response.

4. Automated Thank-You Emails with Personalized Book Recommendations When a reader fills out your feedback form, buys a book, or subscribes to your list, this workflow sends them a personalized thank-you email with book recommendations based on what they’ve already read or purchased. It pulls from your catalog, matches genres and themes, and drafts the email for you. The whole thing runs without you touching it, but you can set it to hold outgoing messages for your review before they send if that makes you more comfortable (I would, at least at first).

5. RSS Feed Monitor — Track Industry News with Images This workflow monitors any RSS feed, grabs new posts with their images, and delivers them to you, which makes it perfect for tracking competitors, industry blogs, podcast feeds, or your own site for broken posts. I use it to monitor about 30 feeds across the publishing industry, and it’s caught a few of my own broken posts before readers noticed (which is exactly the kind of embarrassment prevention I’m willing to automate).

All five are available in the Hub now and they’re all free — no paywalls.

That’s it. That’s the newsletter of a thousand announcements. If you made it this far, you deserve a coffee and possibly a nap. Go register for something and get on Discord.

See you there,

Chelle

I need your help picking cities for the rest of 2026. I'm planning more 3-day intensives (more on those in a sec) and I want to go where you actually are — not where I think you are.

I’m not being coy about this — I will literally book a venue based on your answers. So if you’ve been waiting for something closer to home, now’s when you speak up.

What’s a 3-Day Intensive, Anyway?

Since I mentioned it, let me explain what you’d actually be voting to bring to your city.

I usually travel to 8-10 countries a year (over-caffeinated, usually on a train), but this year I’m sticking closer to home for the rest of 2026.

I’ve started running small-group intensives wherever I land. You show up with a laptop. I show up with every automation, template, and system I’ve built over the past five years. We spend three days building your version — your dashboard, your social media automation, your AI brand voice, your newsletter pipeline, all of it — and you don’t leave until it’s tested and running.

Eight people max. I build alongside you the entire time. If something breaks, we fix it. If you don’t understand something, I explain it while probably also drinking my fourth espresso. If you want to add something weirdly specific to your workflow. One attendee wanted an agent that watches her Kindle sales rank, cross-references it with her ad spend, and sends her a Slack message that says either "keep going" or "pull the ads" with the math attached. We got it working between espressos on day two.

We figure it out.

I have the last two International Intensives coming up in March and I’d really love to fill the remaining seats with people from this community:

London: March 7-9 • 3 spots left — Shannon, Ireland: March 14-16 • 4 spots left

Grab an Intensive Spot

The Year of Support That Comes With It

This is the part that makes the pricing kind of absurd. Every intensive seat includes a full year in my Mastermind — Your fractional CTO and operations brain I cap at 40 members. I’ve opened it up until the end of February.

The Mastermind year includes:

Weekly strategy calls (Thursdays, 10 AM Central) where you can bring literally anything — automations, websites, email deliverability, AI tools, why your WooCommerce cart is doing something unhinged at 2 AM. I’ve probably seen it before. I’m your IT Team.

VIP calendar access for 90-minute co-working sessions and private strategy calls. You book time, we work through your stuff together.

Intimate conversations with some of the smartest people in the industry who come hang out with us for an hour. Ads specialists, Book Box geniuses, Intellectual Property Attorneys. These aren't webinars with 500 people on mute — they're small-room conversations where you can actually ask your questions and get real answers.

Software licenses worth over $300/month — Late, Placid, Meetn, OnlyPrompts, Storylink, Amelia, Elementor Pro, and FluentBoards. All included for the year. (I negotiated these so my members don’t have to cobble together subscriptions like some kind of digital duct tape situation.)

Every paid webinar I teach — Claude Code for Non-Developers, Airtable for Authors, WordPress with AI, whatever I come up with next. You just show up. Two per month starting in March.

Half off every course I sell on Indie Author Training — Airtable, Notion, Claude Code, FluentCRM (beginning and advanced), AI for Marketing, Productivity and Task Management, Direct Sales for Authors, and more.

All my workflows — the actual Make.com, Zapier, and n8n automations running my business right now. Documented, tested, yours to install.

Direct access to me and my connections. No ticketing system. You text, I answer. Usually within the day, unless I’m on a train somewhere in Europe pretending to be offline (I’m never actually offline).

One price includes the three-day intensive AND all of that for a year. I’ll let you do the math.

All the details are at chellehoniker.com/intensive or chellehoniker.com/mastermind if you don’t want the in-person thing and jump straight to the mastermind. It’s cool either way. And if you want to talk it through before committing, grab time at chellehoniker.com/meet. No pressure, no pitch — just a conversation about whether it’s the right fit.

Now go vote so I know where to book my next flight.

Chelle

Last week I mentioned that I knocked off about four months of work in a weekend using Claude Code. I migrated 32,000 images, audited a Ghost website, updated plugins across 60+ sites, cleaned up AWS storage buckets, migrated courses, and fixed bugs on StorytellerOS while I was eating lunch. The robot handled the thing AND told the customer it handled the thing and then emailed me that it had handled the thing. Genius.

That kind of time savings has been incredible for my task list. It’s also left me thinking deeply about the line between knowledge work and creative work—and where AI belongs on either side of that line. I don’t have all the answers. But here’s what sparked my thinking this week.

Dario Amodei, CEO of Anthropic (the company behind Claude), laid out his predictions in a January 2026 essay called The Adolescence of Technology. He describes what he calls “powerful AI”—essentially a “country of geniuses in a datacenter”—that’s smarter than Nobel Prize winners across most fields, can work autonomously for days or weeks at a time, and could be running in millions of instances within 1–2 years. He’s also publicly predicted that AI could displace half of all entry-level white collar jobs in 1–5 years. I buy most of that—not because AI is magical, but because a lot of knowledge work is painfully predictable.

Knowledge Work vs. Creative Work

Knowledge work gets graded on correctness. You take inputs, apply rules, and produce a deliverable that can be checked against a standard. Did you follow the process? Is the output accurate? Congrats, here's your paycheck. It's work that lives inside a flowchart, and agents love flowcharts. They don't get bored, they don't get distracted, and they don't decide to alphabetize the spice rack halfway through a quarterly report. (We've all been there. The spice rack was calling to you. The AI doesn't hear the spice rack.)

Creative work gets graded on whether a stranger feels something, and there’s no checklist for that. You can hit every craft beat correctly and still produce something that lands with a thud, because resonance isn’t a process you can document. I’ve read plenty of books that did everything “right” and I couldn’t tell you a single character’s name a week later.

AI is extremely good at correctness. Left to its own devices, it’s mediocre at resonance—and that gap is where the real work lives. Some writers are closing that gap brilliantly, using AI as a collaboration partner while infusing every page with their own voice, taste, and weird creative instincts. But if you talk to those writers for five minutes, you’ll hear the same thing: it’s hard. It requires knowing your voice so well that you can spot when AI has smoothed it away. It requires taste sharp enough to reject perfectly competent prose because competent isn’t the same as alive. The writers doing this well aren’t saving effort—they’re redirecting it from drafting into curation, revision, and creative direction, which are skills that take real development.

Without that human layer, AI can write a technically solid book right now. It can do structure. It can do tropes. It can do competent prose that offends no one. It can also crank out 90,000 words that feel like wallpaper. Pretty wallpaper, sure. Still wallpaper.

The difference between AI-assisted books that work and AI-assisted books that don’t isn’t the AI. It’s how much of you ended up on the page. Because readers don’t stick around for wallpaper. They stick around for voice.

Voice is not “good writing.” Voice is taste. It’s what you notice and what you ignore. It’s what you’re willing to say out loud. It’s how your brain moves across the page. It’s the difference between “this is fine” and “I have to text my friend about this character immediately.” AI doesn’t have that. You do. The collaboration works when you bring enough of yours to the table that a reader would never know (or care) how the words got there.

Competent Prose Has the Same Problem as Tribute Bands

AI can mimic taste the way a cover band can mimic Depeche Mode. There’s a tribute band called Depeche Tribute that’s genuinely good—they’ve got the sound down, they play the songs right, and I enjoy listening to them. But when I hear them, I don’t feel anything beyond “hey, this is fun.”

When I hear the actual Depeche Mode, it's a random Friday night and I'm drinking Boone's Farm Tickle Pink at the Flats in Lake Arrowhead, California with 50 of my best friends, all of us keeping one eye on the keg that someone's brother that goes to USC got us, and one eye out for cops ready to bust us for underage drinking. I'm not just hearing the music—I'm back in 1985, feeling every dumb, wonderful thing I felt that Sophomore year.

The cover band can’t give me that. They’re playing the notes. They’re not carrying the meaning.

AI-generated fiction—left entirely on its own—has the same problem. It can produce competent prose that hits the right beats and follows the right structures. You might enjoy reading it. But there’s no lived experience underneath it, no accumulation of taste and memory and opinion that makes one author’s voice different from another’s. It’s the notes without the meaning.

I know this newsletter will lose some subscribers because writing with AI is divisive, and some will read this as encouragement. It’s not.

I don’t teach writing with AI. That’s not my lane and I’m not pretending it is. My lane is information, education, marketing automation, and business process optimization, with a sprinkle of with a sprinkle of how to scale without sacrificing sleep, sanity, or the things that made you start writing in the first place.

Educators don’t get to skip the uncomfortable parts just because they’re uncomfortable. I’m not here to tell you what to do with AI and your creative work. I’m here to tell you what I’m seeing, and what I’m seeing is worth talking about honestly.

What I’m seeing is this: some writers are using AI as a collaboration partner and producing work that resonates. They’re bringing their own Boone’s Farm moments to the process—their memories, their opinions, their willingness to cut a perfectly good sentence because it doesn’t sound like them—and using AI to get to a draft faster so they can do the real work of making it theirs. That’s a legitimate creative process. It’s also genuinely difficult, because it means knowing your voice well enough to hear when it’s missing.

So no, I’m not losing sleep over AI replacing novelists. I’m losing sleep over authors burning their best creative energy on admin that machines are about to bulldoze. I'm losing sleep over the stuff that's actually fixable.

Use Agents for the Business Work. Let Them Earn Their Keep.

Give AI the stuff you already resent doing: inbox sorting and triage, drafting and scheduling your newsletter, writing and reformatting book descriptions for twelve different retailers, updating backmatter links across your entire catalog every time you release a new book, generating keywords and categories for a launch, building a series page on your website, pulling your sales data into a spreadsheet that actually makes sense, turning one blog post into a thread, a carousel, a short-form video script, and a pin. Formatting and metadata and distribution. The stuff that keeps you at your desk at 11pm even though the actual book is done.

If an assistant could do it, an agent will do it. Probably faster. Definitely without asking you where the Google Drive folder is.

Then keep the story decisions with the human who has a point of view. Keep the parts that make readers care: what the book is actually about under the plot, what your characters want and what you refuse to give them, what you find funny or sharp or tender or unforgivable, the emotional experience you’re building on purpose, the choices that make your work feel like you and not like “a book in this genre.”

A lot of books are going to get replaced by AI-generated content, because they’re text, not storytelling. AI Agents will eat that market like it’s free snacks. But this isn’t new. Kris Austin, CEO of Draft2Digital, told me his company was fighting off literally thousands of smoothie recipe books a decade ago—people chasing quick money with low-quality content designed to ride whatever trend was hot. Before that it was keto books. Before that it was crypto guides. The opportunists have always been here. AI just made their playbook faster and cheaper.

The scale tells you everything. Draft2Digital now rejects between 40 and 75 percent of all book submissions, and it’s almost entirely AI-generated nonfiction. Thousands arrive every single day. The fiction side is a different story: Austin says authors who’ve tried to write full novels with AI “tend not to do very well in the market. That humanity seems to be missing. It’s kind of one of those things you can’t directly define, but it doesn’t appear that readers are connecting with that content.”

That tracks with everything we’ve been talking about. The notes without the meaning. The cover band without the Friday night memory. Readers notice, even when they can’t articulate exactly what’s off—and Austin predicts they’ll feel betrayed when they find out a book sold as human-authored wasn’t.

Authors who write with opinion, specificity, and a little nerve are going to be fine. Not because AI can’t write. Because AI can’t be you.

So yeah, let the AI agents have the checklists. Writers still get to be weird. And thank all the deities, because the world does not need one more perfectly acceptable book.

What I’m Focused on in 2026

Last weekend I had a website update to handle—SEO cleanup and some structural changes so AI tools could actually find and surface the content correctly (LLM optimization is a whole thing now, and I’ll talk about that next week). My muscle memory said open Make.com. Instead, I opened Claude Code to see if it could handle it.

Spoiler: It did. And then some. Read how:

My own workflow had shifted underneath me without me making some big strategic decision about it. I’ve been reassessing my automations all year, and I keep reaching for agentic AI first—website updates, SEO optimization, content structuring—and reaching for my traditional automation tools second. Make.com, n8n, and Zapier aren’t going anywhere (they’re still the backbone for repeatable, scheduled workflows), but I’m handing the one-shot, messy, or judgment-heavy stuff to agents now.

I’ve also been leaning into two content directions this year outside this newsletter to get some more perspective and options that shape it, and also provide better examples of technical stuff:

CEO interviews — We launched a new series on YouTube/@indieauthormagazine where I sit down with the people running the companies authors depend on. Less “how to use the tool,” more “what’s the vision and why should authors care.” (The Kris Austin interview above came from this series.)

Technical walkthroughs — My channel at YouTube/@chellehoniker is where I break down the actual builds, with deeper automation tutorials living on hub.authorautomations.com.

And one more thing: Indie Author Magazine is now on Substack. If you want industry news, craft deep-dives, and interviews with authors who are doing interesting things, subscribe at indieauthormagazine.substack.com. It’s a different newsletter than this one—less automation nerdery, more publishing strategy and author business. If you’re already subscribed here, you might want both.

What to Do This Week (If You’re Curious About What AI Can Actually Do)

Build an “admin vacuum” for your inbox. Create one rule or filter that catches newsletters, receipts, and platform notifications and sends them to a folder you check once a day. Your brain is not a dumpster. Stop living like it is. (I’ve got a full walkthrough on inbox triage at hub.authorautomations.com—look for the email management workflows.)

Pick one thing from the suck list and hand it to AI this week. Not all of them. One. Maybe it’s reformatting your book description for a new retailer. Maybe it’s pulling your January sales data into a spreadsheet. Maybe it’s updating the website you’ve been ignoring since your last release. Pick the task that’s been sitting on your to-do list making you feel guilty, and let an AI tool take the first pass. You’ll edit the output—that’s fine. You’re not starting from zero anymore.

Create a repurposing pipeline for one piece of content. One newsletter becomes a short post, a longer post, a carousel, and a pin. You edit for voice, but you stop writing five separate things from scratch every week. (The Hub has a content repurposing workflow that does exactly this—connect it to your publishing platform and let it run.)

None of this requires you to have an opinion about AI and writing. This is just getting the business busywork off your plate so you have more time for the work you actually care about—whatever that looks like for you.

If you want help, hit reply and tell me what you’re drowning in right now: inbox, content, research, launches, or customer support. I’ll point you at the simplest automation that gets your time back without turning you into a part-time systems engineer.

—Chelle

Upcoming Free Webinars

All webinars are free and replays are available if you can’t make it live. See the full schedule and register here.

February 10 @ 10am CST — What’s All the Fuss About Claude Code Chelle Honiker (That’s me. I’ll be showing what I’ve been ranting about in this newsletter.)

February 11 @ 11am CST — It’s Not the Price—It’s the Perception: Make Higher Prices a No-Brainer Celeste Barclay

February 18 @ 11am CST — Edit Smarter, Not Harder: How AI Can Reveal What Your Story Is Really Doing Alessandra Torre & JD Lasica

March 4 @ 11am CST — Design a Writing Career That Fits Marla Albertie

March 11 @ 11am CDT — Are You Blocking Your Success? Aryn Van Dyke

April 8 @ 11am CDT — Print Matters: Learn How to Leverage Print to Grow Sales Dave Sheets

April 15 @ 11am CDT — Email Marketing That Turns Readers into Fans (and Buyers) Dale Roberts

This is Author Automations: Advanced, where I share the messy, technical reality of running multiple businesses while building software and occasionally remembering I also write books.

If you're overwhelmed and want to start smaller, the regular Author Automations newsletter is and judges no one. Go to https://authorautomations.com/account and switch off the Advanced Automations notifications and you’ll be all set.

You Went Outside This Weekend. I Talked to Robots

Having said that, I did actually complete three to four months of work in one weekend. I kept a running list, and by Sunday night I was genuinely struggling to remember everything because it got so long. At some point I stopped counting and started just pointing Claude Code at problems like a fire hose.

Funny story: my daughter and her family went on a walk and two blocks later, they were at my house with her husband and Princess Elowyn and Prince Kieran. I was in the middle of using Wispr Flow to dictate instructions, and she asked, “Are you yelling at your AI?” I had to reply yes, but I’m usually nice so when they rise up I’ll be spared.

Anyway. Here’s what got done while I was yelling politely at robots.

The Great Image Migration

I had 32,000 images sitting on an old WordPress server, doing nothing useful except costing me money and whispering “remember me?” every time I logged in. Claude Code moved them to a Photoprism asset manager, labeled them, and made them searchable for the IAM team via ChelleBot. The kind of project that lives on your to-do list for years because you can’t stomach the tedium finally got done while I drank my coffee and pretended I was being productive by watching progress bars.

The Ghost Website Audit

I pointed Claude Code at our Ghost website and asked for a full audit. It found 72 broken images—the slow-drip problem that erodes reader trust and tanks SEO without ever sending you a calendar invite to let you know it’s happening. Claude Code fixed them automatically, flagged the outliers it couldn’t handle, and drafted a complete email to Grace with specific instructions for the manual fixes. I reviewed it, hit send, and felt like a manager for approximately thirty seconds before moving on to the next thing.

Plugin Updates Across 60+ Sites

I own or host more than 60 WordPress sites. (This is what happens when you say “yes” for fifteen years straight.) Each one sends me tagged emails when plugins need updating—emails I had been aggressively ignoring because logging into 60 dashboards sounds like a punishment, not a task. Claude Code reads those emails now, and it updated the plugins as needed. I didn’t log into a single dashboard. Not one.

Cloudflare DNS Housekeeping

I connected Claude Code to Cloudflare and had it audit every domain name record I own. Misconfigurations, outdated entries, the digital equivalent of junk drawers that accumulate when you’re moving fast and telling yourself you’ll fix it later—all cleaned up in one pass. Past Chelle left a mess. Present Chelle had a robot clean it up. Future Chelle is thrilled.

AWS Cost Cleanup

My Amazon AWS account had storage buckets quietly racking up costs with dead files from 2022. Files I forgot existed from projects that ended years ago, just sitting there collecting dust and charging me for the privilege. Claude Code found them and cleared them out, which means my next AWS bill won’t make me mutter curse words under my breath.

A Full Wix-to-WordPress Migration

I migrated an entire website from Wix to WordPress—new design, fully responsive, all the original images moved over. Then I had Claude Code audit the book schemas, check every link, add alt text to images, and verify SEO was solid. One website, completely rebuilt and properly optimized, done in less time than it takes me to decide what to watch on Netflix.

A New Support Portal

I created a dedicated support site at support.indieauthormagazine.com and migrated all our old tickets to the new system. We had a batch of tickets from a technical issue that needed a backend fix. Claude Code fixed the underlying technical problem, then drafted replies to let customers know the issue was resolved, and closed the tickets. The robot fixed the thing AND told people it fixed the thing. I just reviewed and approved.

StorytellerOS Bug Fixes in Real Time

I’ve been using Claude Code to debug and build StorytellerOS.com, and this weekend we cleared the beta user tickets. Here’s where it gets slightly absurd: a user submitted a bug. Claude Code was monitoring my email, identified the issue, fixed it, and sent a notification to the user. Because it’s connected to the server and monitoring logs, it pinged me on Telegram when the user tried the fix and confirmed it worked. I found out a bug was fixed because I got a message saying the user was happy. I was eating lunch. The bug was already handled. I have questions about what my job even is anymore.

Course Migration

I had two courses sitting in LearnDash on an old site, collecting digital cobwebs. Claude Code migrated them to FluentCommunities on IndieAuthorTraining.com—created the courses, built the lessons, moved the videos. The courses that have been on my “I should really move those” list since 2023 are now moved.

The Real Math

I’m sure I’m forgetting things. The weekend blurred together into a stream of “okay, now do this” followed by watching progress bars and reviewing results and occasionally saying “wait, it’s done already?” out loud to no one.

This isn’t about replacing the work. It’s about finally doing the work—the unglamorous, never-urgent, always-important maintenance that keeps a business running smoothly. The stuff that sits on the list for months (or years, no judgment) because there’s always something more pressing, more interesting, or more likely to generate immediate dopamine.

That stuff is done now. And I have feelings about it.

What’s All the Fuss About Claude Code?

You’ve heard people losing their minds over Claude. Maybe you’ve even chatted with it yourself. But now there’s Claude Code, Claude Cowork, and regular Claude Chat — and nobody’s explaining which one actually matters for authors.

Here’s the short version: Claude Chat is the conversation you’re probably already having. Cowork is Anthropic’s attempt at letting Claude touch your files (it’s... fine). Claude Code is the one that just ate four months of my task list in a single weekend.

In this webinar, I’ll break down what each tool actually does, when you’d reach for one over another, and why Code has me rethinking what’s possible for solo operators who don’t have a tech team. We’ll talk real author use cases — not developer hype — and I’ll show you what happened when I pointed Claude Code at my actual backlog and let it run.

Fair warning: I’m going to be honest about the limitations too. These tools aren’t magic, and Cowork in particular has some “bless its heart” moments. But Code? Code is different.

If you’ve been Claude-curious but confused about which flavor to care about, this is the hour that’ll sort it out.

Register here: https://webinars.indieauthortraining.com/talks/whats-all-the-fuss-about-claude-code

This is the reality for indie authors, and it’s annoying.

The platforms that sell your books weren’t built with automation in mind—they gave us dashboards and manual exports instead of APIs and integrations, which means your data stays locked inside their systems until you personally go retrieve it. Some third-party services have emerged to help with this (BookReport, Book Tracker, and similar tools pull KDP data for you), and they’re genuinely useful if their specific features match what you need. But if you want to customize exactly what data you’re tracking and where it goes, building your own automation gives you that flexibility.

That’s where browser automation comes in. These platforms can visit websites, click around, and grab data the way a very caffeinated, very patient assistant would—except they never need coffee breaks and they don’t judge you for checking your sales rank at 2 AM.

These tools are genuinely useful. They’re also operating in a space that deserves a few minutes of thought before you dive in. Let’s talk about it, and then I’ll show you how to actually set one up.

The Quick Ethics Bit

I’m not going to bury you in legal citations (you’re welcome), but here’s the landscape in plain English.

Scraping publicly available data—the stuff anyone can see without logging in—is generally considered fair game. Courts have consistently ruled that visiting a public webpage with a bot isn’t meaningfully different from visiting it with a browser. Your Amazon product page, your public reviews, your book’s ranking—all visible to anyone with an internet connection.

The murkier territory involves logging into accounts. Your KDP dashboard, your detailed analytics, your private reports—those sit behind authentication. Automating access to that data violates platform terms of service, and platforms can enforce those terms by doing things you really don’t want, like suspending your account.

For authors, the practical risk calculation looks like this: scraping your public Amazon product page is low-risk. Automating logins to your KDP dashboard is higher-risk, and if your income depends on uninterrupted Amazon access, that risk might not be worth taking.

The automation I’m going to show you stays on the safe side of that line. We’re grabbing publicly visible data from your Amazon book page—sales rank, review stats, category rankings. Data anyone can see by visiting the URL, no login required.

Meet Browse AI

Browse AI is a no-code browser automation platform that lets you point at data on a webpage and tell a robot to go grab it.

The free plan gives you 50 credits per month, which is enough to test things and run small extractions. Paid plans start around $19/month if you need more volume.

Here’s what we’re building: a robot that visits your Amazon book page, extracts your category rankings and review stats, and can push that data wherever you need it—Google Sheets, Airtable, Make.com, you name it.

Building Your Book Data Robot

Step 1: Grab Your Book URL

You need your Amazon product page URL. Get the cleanest link you can without all the keywords and marketing add-ons. I used an older title that’s about to have a book two, so I want to see what this does to this book’s ranking.

https://www.amazon.com/dp/B0F6NN5WF1/

The ASIN (that alphanumeric code after /dp/) is unique to your book. Grab yours from your Amazon product page or your KDP dashboard.

Step 2: Create a New Robot

Head to browse.ai and sign up if you haven’t already. From your dashboard, click to create a new robot.

You’ll land on the “New Data Extraction Robot” screen. Paste your Amazon book URL in the Origin URL field. Leave the “This website needs logging in” checkbox unchecked—we’re scraping public data, no authentication needed.

Hit “Start Training Robot.”

Step 3: Tell the Robot What to Grab

Browse AI loads your actual book page and lets you point and click on the data you want. When you’re just getting started, choose “text extraction” and then choose what fields on the page you want.

I grabbed two things:

Sales Rank: Click on your Best Sellers Rank on the product page. Browse AI captures it as text. For my book, that pulled in “#5,622 in Witch Romances #6,966 in Witch & Wizard Mysteries #12,336 in Paranormal Witches & Wizards Romance”—all my category rankings in one grab.

Customer Reviews: Click on the review summary area. This captures the star rating, total ratings count, and the percentage breakdown by star level. Mine showed “Customer reviews 4.5 out of 5 stars 4.5 out of 5 2 global ratings 5 star 50% 4 star 50% 3 star 0% 2 star 0% 1 star 0%.”

You can grab whatever data matters to you—price, title, description, whatever’s visible on the page. Just click on it, give it a name, and the robot remembers.

Step 4: Review and Approve

Once you’ve selected your data points, Browse AI runs a test extraction. You’ll see exactly what the robot captured in a clean table format.

Check that everything looks right. The robot shows you the “Final Screen”—a screenshot of your book page with the steps it took listed on the side: Navigate to origin URL, Capture visible text (Rank), Capture visible text (bestsellers), Capture visible text (customer reviews).

If something’s off, you can retrain. If it looks good, click “Yes, looks good” and approve your robot.

Step 5: Connect Your Integrations

This is where it gets fun. Click on the Integrate tab and you’ll see your options: Google Sheets, Zapier, Webhooks, Rest API, Airtable, Pabbly Connect, Make.com, Workflows, Integrately, and AWS.

For most authors, Google Sheets is the easy button. Connect your Google account, pick a spreadsheet (or create one), and map your captured fields to columns. Now every time the robot runs, your data lands in a spreadsheet automatically.

If you’re already living in Airtable or running automations through Make.com, those integrations work the same way—connect, map, done.

Step 6: Set Up Monitoring

If you want to take a look regularly, click the Monitor tab and set a schedule. Daily, weekly, whatever makes sense for how often you want to track your rankings.

You can also set up alerts to get notified when your rank crosses a certain threshold or when new reviews appear. Very satisfying when you’re running a promo and want to watch the numbers move.

What You Can Actually Do With This

A spreadsheet of rankings is nice. A spreadsheet that feeds into other automations is better.

Track your launch. Set monitoring to daily (or hourly during launch week) and watch your category rank climb. Export to a chart. Feel the dopamine.

Spot trends. Over time, you’ll see patterns—how your rank responds to promotions, how reviews affect sales rank, when your category gets competitive.

Monitor your backlist. Set up a robot for each book. One dashboard showing all your titles, updating automatically. No more clicking through five different product pages.

Compare to comps. Nothing stops you from running this same robot setup on competitor books in your genre. Create a Browse AI bot for each comp title you want to track, point them at the same data (rank, reviews, pricing), and feed everything into the same spreadsheet or Airtable base. Now you’ve got a competitive intelligence dashboard that updates itself. You can see when a competitor runs a sale, when their rank spikes (probably a promo worth investigating), and how your performance stacks up over time.

Monitor your categories. You can also build bots that scrape Amazon category pages directly—not just individual books, but the bestseller lists themselves. Track which books are climbing, which are falling, and what patterns emerge in your corner of the market. This is the kind of market research that used to require hours of manual clicking, and now it just shows up in a spreadsheet every morning.

Feed your marketing brain. This is where the real magic happens. Connect Browse AI to Make.com and you can trigger actions based on what the data shows. Hit top 1,000 in a category? Auto-post a celebration to social media. New review appears? Send yourself a Slack notification so you can respond quickly.

But here’s where it gets genuinely exciting: you can add an AI node to your Make.com scenario. Once your rank and review data flows into Make, route it through an Anthropic or OpenAI module and ask Claude or ChatGPT to analyze what’s happening. Your prompt might be something like: “Here’s my current category rank, my review stats, and my pricing. Here’s the same data for my top three competitors. What patterns do you see, and what three actions would you recommend to improve my visibility this week?”

Now you’re not just collecting data—you’re getting personalized business intelligence delivered automatically. The AI can spot things you might miss when you’re too close to your own numbers, and it can suggest experiments worth trying based on what’s working for similar books. You can schedule this to run weekly and land in your inbox every Monday morning with fresh strategic suggestions.

The combination of automated data collection plus AI analysis turns a simple rank tracker into something that actually thinks alongside you. And because you built it yourself, you can customize the prompts, add new data sources, and tweak the analysis to match exactly how you think about your business.

The Bigger Picture

Browser automation tools like Browse AI represent a practical middle path. You can’t get official API access to your book data, but you can extract publicly visible information in ways that are straightforward and low-risk. And unlike off-the-shelf tracking services, you control exactly what gets captured and where it goes.

Use it thoughtfully. Stick to public data. Build in ways that won’t break your business if the tool stops working tomorrow. And remember that the goal isn’t automation for its own sake—it’s automation that helps you make better decisions and write more books.

Your sales rank is sitting there on Amazon right now, being useful to nobody. A robot can go fetch it for you every day and put it somewhere you’ll actually look at it.

That’s pretty cool.

Have you played with Browse AI or built something clever with browser automation? Hit reply and tell me about it—I collect these stories.

Years ago, I was the executive director of a lively nonprofit in Austin. Everyone wanted to take me to coffee and pick my brain (which sounds considerably more gruesome when you think about it for two seconds). You can’t say no to all of them because networking matters, but you also can’t spend two hours with everyone who has “just a few questions” that somehow multiply like gremlins fed after midnight.

I didn’t have an assistant to call me with a fake emergency and rescue me from conversations that should have ended twenty minutes ago. So I engineered my own escape hatch.

I stuck an NFC tag—a little sticker smaller than a quarter—in the outside pocket of my purse. When a coffee meeting hit the forty-five-minute mark and showed no signs of wrapping up, I’d casually reach into my bag (ostensibly looking for a tissue or checking my phone) and tap my phone against it. Instantly, my phone would erupt with an obnoxious alarm and display an urgent-looking notification: “CALL IMMEDIATELY - TIME SENSITIVE.” I’d look apologetic, excuse myself, and escape with my afternoon intact.

Was it a little theatrical? Absolutely. Did it save my sanity? More times than I can count.

The beauty of NFC tags is that they’re programmable little escape pods, productivity boosters, and automation triggers that cost next to nothing. And once you understand how they work, you’ll start seeing opportunities everywhere. First dates that aren’t going well. HOA board meetings that have devolved into arguments about fence heights. That networking event where someone has cornered you to explain their cryptocurrency strategy for the fourth time.

But beyond social self-defense, NFC tags are genuinely useful automation tools that can streamline your daily life in ways that feel almost magical once you set them up.

So What Exactly Is an NFC Tag?

NFC stands for Near Field Communication. It’s the same technology that lets you tap your phone to pay for things, tap your transit card on the subway, or badge into your office or hotel room. The “near field” part means it only works when devices are practically touching—we’re talking a couple of centimeters, not across the room.

An NFC tag is a tiny, unpowered chip embedded in a sticker, card, keychain, or other small object. It contains a minuscule amount of data and an antenna. When you tap your phone to it, your phone powers the chip with its electromagnetic field (this is why the tag doesn’t need batteries) and reads whatever information is stored on it.

Here’s the part that matters for automation: you can program NFC tags to trigger specific actions on your phone. Tap the tag, phone does the thing. No unlocking required, no opening apps, no hunting through menus. Just tap and go.

iPhone vs. Android: The Setup Difference

Both major phone ecosystems support NFC tags, but they handle them a bit differently.

iPhone (iOS 13 and later): Apple uses the Shortcuts app for NFC automation. You create an automation triggered by scanning a specific NFC tag, then tell it what actions to perform. The tag itself doesn’t store the automation—your phone recognizes that particular tag and runs whatever automation you’ve assigned to it.

iPhones can scan NFC tags in the background (no need to open an app first) on iPhone XS and later models. Older iPhones require you to open the Shortcuts app or use the NFC scanner in Control Center.

Android: Android devices have supported NFC for years, and the implementation varies by manufacturer. You can use apps like Tasker, MacroDroid, or NFC Tools to program your tags. Android phones generally need to be unlocked to trigger NFC automations, though some phones support it from the lock screen.

For most authors, the iPhone Shortcuts approach will be the most straightforward, so that’s what I’ll focus on here. Android users—your mileage may vary depending on your specific phone and automation app of choice, but the concepts translate.

The Shortcuts App: Two Tabs That Do Very Different Things

If you’ve opened the Shortcuts app and felt immediately confused, you’re not alone. Apple managed to make this both powerful and bewildering, which is their special talent. (Powerful and Bewildering sounds like a good book title… or maybe a band name, hold on… adding that to a spreadsheet to write that book or start a band.)

The app has two main tabs you need to understand: Shortcuts and Automation. They sound similar but work completely differently.

The Shortcuts Tab: This is where you build reusable actions—like little programs you can run whenever you want. A shortcut might be “text my partner that I’m on my way home” or “turn on my writing playlist and open Scrivener.” You run these shortcuts manually by tapping them, asking Siri, or adding them to your home screen.

Think of shortcuts as recipes you’ve written down. They sit there waiting until you decide to cook.

The Automation Tab: This is where the magic happens for NFC tags. Automations are shortcuts that run automatically when something specific happens—a trigger. Triggers can be things like: arriving at a location, a specific time of day, opening a particular app, or (here’s our star) scanning an NFC tag.

Think of automations as recipes that start cooking themselves the moment you walk into the kitchen.

When you tap an NFC tag, you’re providing a trigger. Your phone recognizes that specific tag, checks if you have an automation assigned to it, and runs whatever actions you’ve set up.

The screenshots I’ve included show both tabs. The Shortcuts tab displays your library of manual shortcuts. The Automation tab shows your list of “when X happens, do Y” rules—including NFC triggers.

Real NFC Tags in My Real Life

Let me walk you through some NFC automations I actually use, because the best way to understand this tech is to see it in context.

The Morning Boot-Up Sequence

I have an NFC tag stuck to my nightstand. When my alarm goes off and I’m tempted to doom-scroll instead of starting my day like a functional human, I tap the tag.

Here’s what happens:

1. The Calm app opens to my preferred meditation

2. A log entry gets created in my Notion journal noting that I meditated (because if it’s not tracked, did it even happen?)

3. After meditation, my Notion journal opens to my morning prompts page

This last part is key—I have a template with prompts I answer each morning. Things like intentions for the day, what I’m grateful for, what’s weighing on me. The NFC tag creates a smooth runway from “barely conscious” to “actually journaling” without requiring any decision-making from my groggy brain.

The automation handles the cognitive load so I don’t have to. I just tap and follow the breadcrumbs.

The Walking Workout Tracker

I stuck an NFC tag by my front door—the inside, where I’d naturally tap it heading out and coming back in.

When I tap it on my way out, it starts logging a walking workout in the Health app. GPS tracking kicks in, timer starts, the whole thing.

When I tap it again coming home, it ends the workout and does two things: logs the walk to my Notion journal (date, duration, distance) and sends the data to MyFitnessPal so my exercise and nutrition tracking stay in sync.

I used to forget to start my workout tracker until I was three blocks from home. Or I’d finish a walk and forget to log it anywhere. Now my front door handles the remembering for me.

The Refrigerator Grocery Trigger

There’s a tag stuck to the side of my refrigerator, right at eye level. When I tap it, my grocery list app opens directly to the input field.

This sounds so simple it’s almost embarrassing, but it solved a real problem. I’d notice we were out of eggs while cooking breakfast, think “I should add that to the list,” then promptly forget by the time my hands were free. Now I tap the tag while the thought is fresh, speak or type the item, done.

The key is reducing friction. The fewer steps between “I need eggs” and “eggs are on the list,” the more likely the eggs actually make it to the list.

The Coffee Bean Autoorder

I have an NFC tag stuck to my coffee maker. When I tap it, a shortcut runs that opens the Lavazza subscription site with my usual order pre-loaded. One more tap to confirm, and beans are on their way. I used to have this on autoship, but as much travel as I do, I ended up with an entire freezer full of bricks of beans in the garage freezer and some side-eyes from neighbors once when I opened it to get a popsicle for Princess Elowyn.

You could set this up with any regularly ordered item—printer paper, pet food, vitamins, whatever you tend to run out of at inconvenient times. The tag becomes a physical “reorder” button placed right where you’ll notice you need it.

The Starbucks Car Tag

This one brings me a possibly unreasonable amount of joy. I have an NFC tag in my car, stuck between two air vents. Tap, and my phone places my exact Starbucks mobile order: hot grande quad-shot pistachio oatmilk latte, upside down, ristretto. (Yes, that’s a real order. Yes, the baristas probably have opinions about me. No, I don’t care because it’s delicious and I order it in the app so they don’t have to key it in. I’m thoughtful that way.)

By the time I pull off exit 200 on I-35, I sail through the drive-thru, with enough caffeine to tackle the Austin traffic.

The automation handles the order details I always want. If I ever need to modify it, I can just open the app normally. But 95% of the time, my order is exactly the same, and the tag makes it effortless.

More Ideas to Get You Thinking

Beyond my personal setups, here are some NFC tag possibilities that might fit your author life:

Writing mode activation: Tap a tag at your writing desk to enable Do Not Disturb, open your writing app, start a focus playlist, and maybe log your start time for tracking.

Conference networking: Program a tag with your contact info (vCard) that people can tap to instantly save your details. Cheaper and more memorable than business cards, and you’ll never run out.

Sprint timer: Tap to start a 25-minute Pomodoro timer. Tap again when the alarm goes off to log your sprint and take a break reminder.

Car mode: Tap when you get in the car to start your podcast app, enable driving focus mode, and pull up your navigation app.

End of workday: Tap to log your stop time, close work apps, disable work notifications, and maybe text your partner that you’re wrapping up.

Getting Started: What You Need

The tags themselves: NFC tags are shockingly cheap—you can get a pack of 10 stickers for under $5 on Amazon. Look for NTAG215 or NTAG216 tags, which have enough memory for most uses and work well with both iPhone and Android.

The setup process (iPhone):

1. Open the Shortcuts app

2. Tap the Automation tab at the bottom

3. Tap the + in the top right corner

4. Select “NFC” as your trigger

5. Tap “Scan” and hold your phone to the NFC tag you want to use

6. Name your tag something memorable (you’ll thank yourself later when you have twelve tags and can’t remember which is which)

7. Add the actions you want to happen when you scan that tag

8. Save your automation

That’s genuinely it. The first one might take you ten minutes while you figure out the interface. After that, you’ll be cranking them out in seconds.

When things don’t work (because they won’t always):

Your phone needs to be awake to scan NFC tags. It doesn’t need to be unlocked, but the screen needs to be on. If you’re tapping a tag and nothing’s happening, wake your phone first.

Some phone cases—especially thick ones or those with metal components—can interfere with NFC scanning. If you’re having trouble, try removing the case temporarily to see if that’s the culprit.

The NFC reader on iPhones lives near the top of the phone (around the camera area). Hold that part of your phone close to the tag, not the bottom or middle.

If an automation suddenly stops working, check that the Shortcuts app still has permission to run automations. iOS updates occasionally reset these settings, which is annoying but fixable in your phone’s Privacy settings.

And if your carefully programmed tag launches the wrong automation—congrats, you’ve discovered that NFC tags can’t actually be “erased” the way you might expect. What’s happening is you accidentally assigned a new automation to that tag, overwriting the old one. Each tag has a unique ID, and your phone remembers what automation goes with what ID. To fix it, just edit the automation in the Shortcuts app; no need to throw out the tag.

Placement tips:

• Stick tags where you’ll naturally reach for them—the goal is zero extra effort

• The adhesive is pretty good but not indestructible—consider using a clear case or holder for tags you’ll tap frequently

• Tags work through thin materials (phone cases, fabric, paper) but not through metal. Don’t stick them on metal surfaces or the signal won’t get through

• Keep tags away from credit cards and key fobs, which can interfere with the signal

• If you’re putting a tag somewhere public (like your car), remember that anyone with an NFC-enabled phone could potentially scan it. Don’t program a public-facing tag to do anything sensitive

• Colored or printed NFC tags exist if aesthetics matter—they work the same as plain white stickers

• Some people write directly on tags with a Sharpie to remember what they do. Low-tech solution to a high-tech problem, but it works

The Bigger Picture

NFC tags are automation training wheels. They teach you to think in triggers and actions, which is the foundation of every automation platform—Zapier, Make.com, n8n, all of it. “When this happens, do that.” The tag is just a very physical, very tangible version of that logic.

Once you start thinking this way, you’ll see automation opportunities everywhere. Every repetitive task becomes a candidate for “how could I trigger this automatically?” Every frequent action becomes “where could I put a tag to make this easier?”

And here’s the sneaky benefit: these little wins build momentum. Setting up an NFC tag and watching it work feels like a small superpower. That feeling makes you more willing to tackle bigger automations, more complex workflows, more ambitious systems.

You start with a tag that opens your grocery list. You end up with an entire automated content calendar.

Your Assignment (Should You Choose to Accept It)

Pick one friction point in your daily routine. Something small, something you do often, something that always requires a few too many taps to accomplish.

Order some NFC tags (or use one from that pack you bought three years ago and never figured out). Set up one automation. Just one.

Stick the tag somewhere logical. Use it for a week.

Then reply to this email and tell me what you built. I love seeing what people create once they realize these little stickers are actually tiny automation buttons disguised as arts and crafts supplies.

And if you ever find yourself trapped in a meeting that should have ended twenty minutes ago—well, now you know there’s a solution that fits in your purse pocket.

Happy automating, Chelle

I spoke at the Women in Publishing Summit this week, and about three minutes into the Q&A, someone asked The Question. You know the one. It comes up at nearly every conference, every panel, every workshop I teach. Sometimes it’s phrased politely, sometimes it’s loaded with suspicion, but it always boils down to the same thing:

“What’s your stance on using AI ethically?”

I took a breath. Not because I don’t have an answer—I have a lot of thoughts on this—but because the question deserves more than a soundbite. It deserves honesty, nuance, and the acknowledgment that reasonable people can disagree.

So here’s what I said, more or less. And here’s what I want you to hear.

I’m Not Here to Persuade You

Let me be clear about something right up front: it’s not my job to convince you to use AI. That’s never been what Author Automations is about.

My job is to educate. To show you what’s possible, explain how things work, and give you enough information to make your own informed decision. Whether you decide AI is a useful tool for your author business or you decide it’s not for you—both of those are valid choices. I’ll respect either one.

What I won’t do is pretend there aren’t real concerns. There are. And we, as humans and entrepreneurs and members of creative communities, need to pay attention to them.

Environmental impact. Training large language models requires massive computational resources, which means massive energy consumption. Data centers aren’t running on good vibes and optimism. The environmental footprint of AI is real, and it’s worth understanding what you’re contributing to when you use these tools. Some companies are more transparent about this than others. Some are investing heavily in renewable energy; some are not. This matters.

Ethical considerations. How were these models trained? What data did they learn from? Were creators compensated—or even informed—when their work was used? These aren’t hypothetical philosophy questions. They’re active legal battles and ongoing industry debates. The answers aren’t settled, and they vary significantly between different AI providers and models.

Labor displacement. Are these tools replacing jobs? In some cases, yes. In some cases, they’re augmenting jobs or shifting what work looks like. The impact isn’t uniform across industries or roles, and pretending otherwise doesn’t help anyone.

I’m not going to stand here (or sit here, typing in my pajamas with my third cup of coffee) and tell you these concerns don’t matter. They do. And you get to weigh them against the potential benefits and decide what’s right for you, your business, and your values.

Eyes on Your Own Paper

Now here’s where I’m going to get a little spicy.

I’ve watched colleagues—people I respect, people with real platforms and influence—use their social media presence to incite mobs against authors they suspect of using generative AI. One-star review brigades. Public callouts. Accusations based on vibes and writing style analysis. See that em-dash up there? Down there? I’ve prayed at the altars of the em-dash since the second grade, friends.

This mob mentality horrifies me.

Not because I think authors owe the world a detailed inventory of their creative process. They don’t. There’s a difference between secret—which carries the whiff of shame, like you’re hiding something wrong—and private, which just means how I run my business isn’t yours to audit. I don’t ask other authors which dictation software they use, whether they hire ghostwriters, or how much their developmental editor rewrote. Those are private business decisions, not public confessions.

What horrifies me is that we’ve created an environment where suspicion alone is enough to tank someone’s career. Where the court of public opinion moves faster than anyone can defend themselves. Where the energy we could be spending on our own work is instead being poured into policing other people’s creative processes.

Here’s a radical thought: the readers get to decide.

Readers are smart. They know what they like. They can tell when something feels off, when a book doesn’t resonate, when the voice feels hollow. They vote with their wallets and their reviews and their recommendations to friends. That’s how it’s always worked. That’s how it should work.

If a book written with AI assistance delights readers, who exactly are we protecting by destroying its rating? If a book written entirely by human hands doesn’t connect with its audience, no amount of “authentically human” marketing is going to save it.

The market will sort this out. It always does. And in the meantime, I’d rather see authors focused on writing great books than building dossiers on their competitors.

Keep your eyes on your own paper. Your business, your craft, your readers. That’s where your energy belongs.

AI Is Always Your Choice

In this newsletter, in the courses I teach, in any software I build—AI is optional. Always.

I want to make this crystal clear because I know some of you are here specifically for the non-AI automation content. Good news: most of what I teach doesn’t require AI at all. The vast majority of automations in Make.com, Zapier, and n8n work beautifully without a single AI module in sight.

You can automate your newsletter subscriber tagging without AI. You can sync your sales data across platforms without AI. You can create task management workflows and calendar integrations and file organization systems—all without touching anything that resembles artificial intelligence.

That said, there are decided advantages when you do incorporate AI into certain workflows. Speed, flexibility, the ability to handle unstructured data, personalization at scale. I’ll always be honest about when AI makes a workflow dramatically better and when it’s just adding complexity for the sake of being fancy.

But the choice is yours. Every single time.

Do Your Homework (But Skip the Hysteria)

If you’re trying to figure out where you stand on AI, here’s my advice: research, learn, understand, then decide.

Read the Terms of Service. Actually read them. I know, I know—they’re long and boring and written by lawyers who get paid by the word. But this is where you’ll find out what happens to your data, whether your inputs are used for training, and what rights you’re granting when you use a tool. Different platforms have wildly different policies. Assuming they’re all the same is a mistake.

For example, some AI tools explicitly state they won’t train on your inputs. Others reserve that right unless you opt out. Some have enterprise tiers with different data handling than their free versions. Some have changed their policies multiple times in the past year alone. The only way to know what you’re agreeing to is to actually look. I keep a running document of the tools I use and their relevant policy details—boring but necessary.

Subscribe to newsletters that educate. (Hi, hello, you’re already doing this one.) Find sources that explain how things work without pushing you toward a predetermined conclusion. Be suspicious of anyone who’s absolutely certain they have all the answers—the landscape is changing too fast for that kind of confidence.

Avoid the hype vortex. There are people who want you to believe AI will solve every problem and revolutionize your entire existence. There are people who want you to believe AI is a moral catastrophe that will destroy creativity forever. Both camps are selling something, and neither is giving you the full picture. The truth is messier and more boring: it’s a tool. It does some things well and some things poorly. The end.

Make your decisions from a place of information, not fear. Not hype. Not whatever your Facebook group is panicking about this week.

Humans in the Loop, Always

I’m always going to be an advocate for humans doing human jobs. That’s not a contradiction to using AI—it’s how I think AI should be used.

Here’s a real example from my own business. We used to have a team member spending significant hours every week creating social media posts. Researching content, writing captions, formatting for different platforms, scheduling—the whole production. It was tedious, time-consuming work.

We automated the creation process. AI helps draft posts, pulls from our content library, formats appropriately for each platform. The workflow handles the mechanical parts that used to eat up hours.

Did we eliminate that team member’s job? Nope. We shifted their hours to engagement. Now instead of writing posts, they’re answering questions from real humans. They’re proactively chatting with our community. They’re building relationships and solving problems and doing the work that actually requires a human being.

The posts still get reviewed by a human before they go out, by the way. We’re not just firing content into the void and hoping for the best. The automation creates the draft, suggests optimal posting times, handles the formatting—but a real person makes the final call. That’s the loop. That’s where the human judgment lives.

That’s the goal. Automate the mechanical so humans can do the meaningful.

Nobody wants AI slop—bots talking to bots, engagement pods where no actual person is reading anything, comments generated by machines responding to posts generated by other machines. That’s not a community; that’s a very sad theater production with no audience.

The human stays in the loop. The human does the human work. The robot handles the robot work. Everybody’s happier.

Why I’m Not Worried About AI Taking My Job

People ask me this too, usually with genuine concern in their voices. Aren’t you worried AI will make what you do obsolete?

Honestly? No.

Can AI build automations? Sort of, sometimes, with a lot of hand-holding and cleanup. Can AI write tutorials? Technically, though they tend toward the bland and generic. Can AI teach workshops and answer questions and help authors figure out what they actually need?

Can AI drink four cups of coffee before noon and be strategically snarky about technology and make Taylor Swift references that actually land?

No. No, it cannot.

The things that make my work mine—the voice, the perspective, the ability to read a room and adjust, the relationships built over years of showing up—those aren’t replicable. And the same is true for you. Your voice. Your stories. Your particular way of seeing the world and translating it into words.

AI is a tool. I use it. In some cases, I specifically don’t use it because of my own ethical stances on certain applications. But at the end of the day, it exists so I can get back to the thing I actually care about: storytelling.

Let the bots do the boring, so I can do the brilliant. That sounds like an excellent tagline, if I do say so myself.

The Bottom Line

Use AI or don’t. That’s your call, and I’ll support you either way.

But whatever you decide, make it an informed decision. Read the fine print. Understand what you’re trading and what you’re gaining. Stay curious, stay critical, and for the love of all that is holy, stop one-star reviewing people based on suspicion.

We’re all just trying to tell stories and build sustainable businesses. Let’s act like it.

Got thoughts on this? Disagree with me? Agree enthusiastically? Hit reply—I read every response, and this is exactly the kind of conversation I want to be having.

Chelle

I don’t do resolutions anymore.

I know, I know—it’s the first Saturday of a brand new year and you’re probably expecting something about fresh starts and ambitious goal-setting and making 2026 your best year yet. That’s not what you’re getting today. Pour yourself some coffee (or tea, I don’t judge) and settle in, because we’re going to talk about jars and rocks instead.

You’ve probably heard the metaphor. A professor stands in front of a class with an empty jar and a pile of rocks. She fills the jar with the big rocks first, then adds smaller pebbles that settle into the gaps, then sand that fills the tiny spaces, and finally water that seeps into whatever’s left. The lesson: if you don’t put the big rocks in first, they’ll never fit. Start with what matters most.

Here’s the thing nobody mentions about that metaphor. You have to decide what your big rocks actually are. And that’s harder than any resolution I’ve ever tried to keep.

My Big Rocks (Not Goals, Values)

A few years ago, I stopped thinking about what I wanted to accomplish and started thinking about how I wanted to feel. Not in a woo-woo manifestation way—in a practical, “what does a good life actually look like for me” way. I landed on six core values that I return to every year. Not to measure. Not to grade myself. Just to check in and ask: am I still protecting these rocks?

Financially peaceful. This isn’t a number. It’s not a net worth goal or a revenue target. Financial peace means I’m not stressed about the things I choose to spend money on. It means buying the good coffee without guilt and booking the trip without a spreadsheet negotiation in my head. It means enough margin that an unexpected expense is annoying, not catastrophic. The goal isn’t wealth—it’s the absence of money anxiety.

Creatively free. This is why I do everything I do with automations. I want to read and write. That’s it. That’s the dream. Every workflow I build, every system I create, every team member I bring on—it’s all in service of protecting my creative time. If I’m spending my days doing things that a well-designed automation could handle, I’ve failed this value. The robots work so I can write.

Light in responsibilities. Once upon a time, I was a chronic hustler. Volleyball team mom. The unofficial web developer for every nonprofit in a 300-mile radius. If a hand needed raising, mine shot up first. I thought being helpful meant saying yes to everything.

I don’t do that anymore.

Now I’m incredibly picky about my time, my attention, and my commitments. If I say yes, it means I genuinely care—not that I felt obligated or guilty or like I should. This value also extends to stuff. A few years ago, I downsized everything expecting a big move overseas. The pandemic killed those plans, but the lesson stuck: I need so much less than I thought. Every thing I own is something I’m responsible for maintaining, storing, and eventually figuring out who inherits. That mental weight adds up. Now when the urge to buy hits, I pause. Do I want to be responsible for this object? Usually the answer is no.

Politically active. I’m not going to overshare here, but this value is non-negotiable for me. It means working locally with candidates I believe in. It means showing up to vote—every single time, no matter how small the election seems. School board. City council. All of it. Freedom isn’t free, and the price I’ve decided to pay is being an engaged member of the electorate. That’s my civic big rock, and it stays in the jar.

Image confident. Something magical happened when I turned 40: I joined the “we do not care” club and never looked back. A number on a scale or a size on a tag doesn’t define me. Image confidence doesn’t mean looking a certain way—it means that wherever I am and whatever I’m doing, I feel good. I’m the only audience that matters for this one. Some days that means a full face of makeup; some days it means yoga pants and a messy bun. The confidence comes from internal permission, not external validation.

Ready for adventure. Oh, I’ve taken this one to heart. Eight to ten countries every year. Travel and coffee are my lifeblood—the two non-negotiables that make everything else feel worth doing. This year looks different; I’m staying closer to home for some family obligations. But the go-bag is always packed (literally, it lives by my door), and the passport is current. Adventure doesn’t always mean international flights. Sometimes it means saying yes to the weird local thing or taking the scenic route or trying the restaurant that looks a little questionable but smells amazing.

The Worst Best Year

2025 was one of the best and worst years of my life. I’m not going to unpack all of it here, but if you’ve been through a year like that—where incredible highs and devastating lows happened almost simultaneously—you know what I mean. You come out the other side different. Not broken, but... recalibrated.

What I learned (or maybe re-learned) is that rituals matter more than resolutions. I can’t control outcomes, but I can control what I do every day. The morning coffee routine. The weekly check-in with my team. The daily writing time that’s protected like the sacred thing it is. Rituals are promises I can actually keep because they’re small enough to show up for.

And processes matter more than goals. “Write a book” is a goal. “Write 500 words every morning before checking email” is a process. Goals live in the future and taunt you with how far away they are. Processes live in today and give you something to do right now. I’ll take the process every time.

What This Year Looks Like

For 2026, I’m thinking about how to better support my team. New apps, new workflows, new automations—tools that make their work easier so they can make my creative freedom possible. It’s a virtuous cycle when it works: I build systems that free up their time, they handle the business operations that would eat my writing hours, and everybody wins.

I’m also thinking about you—what you need to make your author business feel less like a second job and more like the sustainable creative life you actually want. That’s what this newsletter is for. That’s what Author Automations exists to do. Not to add more to your plate, but to help you figure out which rocks belong in your jar and which ones you can leave on the ground.

So no, I’m not making resolutions. I’m not declaring 2026 the year of anything, although I do have a word that I’ve adopted as a North Star for the year, and my dear friend Bradley Charbonneau and I talked about it on his YouTube channel over a firepit in my backyard last month.

I’m just going to keep checking in with my values, protecting my big rocks, and trusting the process.

That’s enough. That’s always been enough.

What about you? What are the big rocks you’re protecting this year? Hit reply—I actually read these, and I’d love to know what matters most to you heading into a new year.

Chelle

One Google Sheet for my series bible. Another for tracking releases. A third for character details (that I never updated). A fourth for event schedules. And somewhere in the chaos, a lonely CSV file holding pen name information that I’d exported from... somewhere? I honestly couldn’t tell you.

Sound familiar? Most authors I know are running their entire careers on a patchwork of spreadsheets, Notion pages, Scrivener projects, and sticky notes that have achieved sentience. And look, it works. Kind of. Until you’re hunting for that one character’s eye color across forty-seven documents at 11 PM the night before your deadline.

When I built StorytellerOS, I needed a backbone that could handle the beautiful chaos of author life. Something that could connect pen names to series to books to chapters to characters to worldbuilding—and let me actually find things when I needed them. That’s when I went all-in on Airtable, and I haven’t looked back.

Let me show you why.

The Spreadsheet Problem (And Why Your Current System Is Lying to You)

Spreadsheets are fantastic. I’m not here to trash-talk Google Sheets or Excel. They’ve served us well for decades, and for simple lists, they’re absolutely the right tool.

But here’s the thing about spreadsheets: they’re flat. Every piece of information lives in its own little cell, completely unaware of what’s happening in the cell next door. Your spreadsheet doesn’t know that “Aurora Quinn” in your pen names sheet is the same “Aurora Quinn” listed as the author in your releases sheet. It just sees text. Dumb, disconnected text.

This means when you want to see all the books by Aurora Quinn, you have to manually filter. When you want to update her bio, you have to hunt down every place you’ve typed her name and update each one. When you want to know which series she writes and what characters appear across those series... well, you better have a good memory and a lot of coffee.

I spent years doing exactly this. Duplicating information across sheets. Copying and pasting pen name details into multiple places. Trying to maintain consistency across documents that had no idea they were supposed to be talking to each other.

It was exhausting. And (here’s the kicker) it was totally unnecessary.

Enter the Relational Database (But Make It Pretty)

Airtable looks like a spreadsheet. That’s part of its genius. You open it up and see familiar rows and columns, a comfortable grid that doesn’t immediately scream “I’m going to require a computer science degree.” But underneath that friendly interface is something entirely different: a relational database.

Here’s what that means in human terms.

In Airtable, you don’t just type information into cells—you create relationships between pieces of information. Your pen name isn’t just text; it’s a record that can be linked to other records. Your series connects to your pen name. Your books connect to your series. Your chapters connect to your books. Your characters can appear across multiple books and series.

Everything is connected. Everything knows about everything else.

Let me give you a concrete example from my own setup.

I have a table called “Author Pen Names” where each pen name lives as its own record with all the relevant details—bio, genre focus, brand voice, social handles, the works. Then I have a “Series” table, and each series record has a field that links to the pen name table. I don’t type the pen name again; I select it from a dropdown that pulls from the pen name table.

This means if I ever update Aurora Quinn’s bio in the pen names table, that change ripples everywhere. Every series, every book, every place that references her—all automatically current. No hunting. No copying. No forgetting to update that one obscure spreadsheet you made three years ago.

And the queries? Oh, the queries.

I can ask my Airtable: “Show me all the incomplete chapters in the Port Tempest series that haven’t been sent to my editor yet.” Boom. An answer in seconds. Or: “Which characters appear in more than three books?” Done. “What events in my timeline happen before the protagonist is born?” Easy.

Try doing that with a spreadsheet. Actually, don’t. Life’s too short.

What I Actually Store (And What I Don’t)

What I Actually Store (And What I Don’t)

Here’s my personal Airtable setup for StorytellerOS, and honestly, it’s become the command center for everything in my author business. And I do mean everything.

The Creative Side:

• Pen names: Full profiles including brand voice, genre, target audience, social media handles, brand colors, and every detail I might need when switching hats

• Series information: Series-level data that connects to pen names, including genre, trope bundles, series arc summaries, and reading order

• Title information: Every book with its metadata, publication dates, ISBNs, links, cover art, blurbs—all connected to the relevant series

• Chapter information: Scene-by-scene breakdowns with word counts, status, beat sheet alignment, and draft tracking

• Characters: Full character profiles with physical descriptions, backstory, personality traits, relationships to other characters, and (crucially) which books they appear in

• Lore and worldbuilding: Everything from magic systems to family trees to historical timelines to made-up swear words (Port Tempest has a lot of those)

• Events and timeline: Chronological events that can span multiple books, with automatic sorting and filtering

The Business Side:

• Social media content: Every post across eleven channels (yes, eleven) with scheduling dates, content type, which book or series it promotes, and—here’s the good part—performance metrics. Engagement, reach, clicks, all tracked so I can actually see what’s working instead of guessing.

• Book sales: Data synced from KDP, Draft2Digital, and Direct2Readers.com. Sales by title, by platform, by time period. When everything lives in one place, you can finally answer questions like “which series performs best on which platform” without exporting seventeen CSVs and crying into your coffee.

• Email marketing: A full backup of my subscriber list from FluentCRM, plus every campaign I’ve sent with open rates and click rates. I can see which emails drove sales, which subject lines bombed, and how my list health changes over time.

• Task management: Because of course I have a task manager in here. Tasks linked to specific books, series, or business functions. Due dates, priorities, status tracking—all connected to the projects they serve.

• Finance tracking: Income and expenses tied to specific books, series, pen names, or business activities. Royalty tracking, advertising spend, production costs. The data I need for tax season without the annual panic.

One database. Everything connected. Nothing lost in a forgotten spreadsheet tab.

What doesn’t live in Airtable:

The actual manuscript. I don’t write my books in Airtable (though technically you could—there’s a long text field that handles quite a bit). For the actual prose, I use Scrivener, but keep it all in Github. Airtable is the brain, but the writing happens elsewhere.

This division makes sense to me. Airtable excels at structured data—the about of your book. Character heights, eye colors, publication dates, chapter statuses. It’s less ideal for flowing prose and the creative chaos of drafting. Know your tools, use them for what they’re built for.

The Magic of Linked Records

Let me nerd out for a second about linked records, because this is where Airtable goes from “nice spreadsheet alternative” to “how did I function without this.”

In my Characters table, I have a field called “Appears In” that links to my Books table. When I’m creating a new character profile, I can select which books feature this character. But here’s where it gets good: Airtable automatically creates a reciprocal field in my Books table showing which characters appear in that book.

I didn’t have to set that up twice. I linked in one direction, and Airtable said, “Oh, you probably want to see this from the other side too, right?” Yes. Yes, I do.

This means I can be working in my Books table and instantly see every character in that book without switching views. Or I can be in my Characters table and see their full appearance history across the series. The information flows both ways because the relationship exists both ways.

Now multiply this by every relationship in your author business.

Pen names ↔ Series ↔ Books ↔ Chapters ↔ Characters ↔ Locations ↔ Events

Every piece of information is one click away from every other piece of information it relates to. That scene where your protagonist meets their love interest for the first time? You can see the chapter details, the character profiles for both characters, the location description, where it falls in the series timeline, and which book it belongs to—all without leaving the record.

When I’m chatting with AI about my stories (more on this in a minute), this interconnected structure means I can pull any detail instantly. No more scrolling through documents hoping I described the tavern consistently. It’s all right there, linked and findable.

Automations: The Robots Living Inside Your Database

Here’s where Airtable starts to feel a little bit like magic.

Airtable has a built-in automation engine. You can set up triggers and actions that happen automatically based on conditions you define. And unlike external automation tools (which I love, don’t get me wrong), these automations live inside your database with direct access to all your data.

A few examples from my setup:

When a chapter status changes to “Complete”: Airtable automatically updates the word count roll-up for that book and sends me a little congratulatory notification. (Listen, writing is hard. I need the dopamine hits.)

When a new book record is created: Airtable auto-populates certain fields based on the linked series—genre, pen name, brand voice guidelines—so I don’t have to manually enter information that already exists elsewhere.

When a publication date is 30 days out: Airtable creates a reminder in my marketing checklist and sends an email to my assistant.

These aren’t complicated automations. Airtable’s automation builder is genuinely user-friendly, with a visual interface that shows you exactly what’s going to happen at each step. You pick a trigger, you pick actions, you tell it which fields to use, and you’re done.

I’ll go deeper into Airtable automations next week because there’s so much more to explore here. For now, just know that this capability exists, it’s powerful, and it doesn’t require you to learn code or connect external tools.

The New AI Features (They’re Actually Useful)

Airtable recently rolled out what they’re calling Omni AI, and I’ll admit I was skeptical. AI features bolted onto existing products often feel like afterthoughts—impressive-sounding but practically useless.

This is different.

The AI in Airtable can help you build tables from plain English descriptions. I can say “I need a table to track my characters with fields for name, physical description, personality traits, backstory, and which books they appear in” and Airtable will generate a reasonable starting structure. Does it nail everything perfectly? Not always. But it gets me 80% of the way there in seconds, and I can adjust from there.

Chatting Via Claude or ChatGPT

More importantly (for my daily use), I can have conversations with my data. Because I’ve setup MCP server access in Claude and ChatGPT, I can query across my entire base—every table, every linked record—and pull information I need. “Which characters in the Jamison Pack series have a connection to the Scottish family?” The AI can trace those relationships and give me an answer.

This changes everything for series continuity. Instead of manually hunting through character sheets, I can ask questions in plain English and get answers drawn from the structured data I’ve already built. The more complete your Airtable setup, the smarter these conversations become.

Starting Your Own Author Command Center

If you’re sold on Airtable (and honestly, how could you not be after this love letter I’ve just written?), here’s how I’d suggest approaching it.

Start with one table. Seriously. Don’t try to build the whole universe on day one. Pick your biggest pain point—maybe it’s keeping track of characters, maybe it’s series metadata—and build just that table first. Get comfortable with how records and fields work.

Add links when they feel natural. Once your first table is humming along, you’ll start noticing connections. “I keep typing this series name over and over” is a sign you need a Series table that links to your Books table. Let the structure emerge from your actual workflow.

Use views liberally. Airtable views are just different ways of looking at the same data. You can have a “By Series” view and a “By Publication Date” view and a “Needs Editing” filtered view—all pointing at the same underlying information. Create views that match how you actually think about your work.

Don’t over-complicate. You don’t need every field on day one. You don’t need every automation. Start simple, use it for a while, and add complexity when you feel the need. The best system is the one you’ll actually use.

Ready to stop wrestling with disconnected spreadsheets? Paid subscribers can download my Author Command Center starter template—a ready-to-use Airtable base with tables for pen names, series, titles, chapters, characters, worldbuilding, events, and lore, all pre-linked and waiting for your data. Copy it to your own workspace and start building your author brain today.

The Backbone That Holds It All Together

I chose Airtable as the backbone for StorytellerOS because an author’s career isn’t a flat list—it’s a web of interconnected pieces. Pen names connect to series. Series connect to books. Books connect to characters and chapters and covers and keywords and publication dates and marketing campaigns. Everything touches everything else.

A flat spreadsheet can’t hold that complexity without becoming a maintenance nightmare. Airtable can.

The relational structure means your data stays consistent. The automation engine means your workflows can run themselves. The AI features mean you can actually have conversations with the universe you’ve built, pulling details when you need them without drowning in tabs and documents.

Is there a learning curve? A small one. But if you can use a spreadsheet, you can use Airtable. And once you experience the difference between disconnected cells and truly linked data, you’ll wonder how you ever managed without it.

Next week, I’m diving deep into Airtable automations—the specific recipes that can save you hours every week and keep your author business running even when you’re buried in drafting. Get ready to make your database work as hard as you do.

Until then, if you’re already using Airtable for your author business, hit reply and tell me what you’re tracking. I’m always looking for clever setups I haven’t thought of yet.

Read more

I’m personally holding out hope that I can train a bot to handle this for me, but until then, I’ve got something actually useful: a way to capture tasks the moment they hit your brain, whether you’re elbow-deep in baking pies or standing in line at the post office.

Android Folks —I’ll do a comparable tutorial for you next. This week’s for the Apple Fangirls (and boys).

Your iPhone Has a Secret Weapon (And It Lives on Your Wrist)

Let me introduce you to Shortcuts—or reintroduce you, because chances are you’ve ignored them since they appeared on your phone. Shortcuts are Apple’s built-in automation tool, and they’re criminally underused. Think of them as tiny automation recipes that live on your phone and Apple Watch, ready to handle repetitive tasks with a single tap.

The specific scenario I’m walking through today came from a conversation with a colleague who was tired of fumbling with her phone every time she needed to capture a task. She wanted something faster—tap her watch, speak the task, done. Turns out, you can do exactly that with Shortcuts and Notion.

(If you’re wondering why Notion specifically, I wrote and recorded whole guide with video about how I use it to run my author business and why it replaced my messy juggle of Todoist, Evernote, and three other apps. The short version: it’s my single source of TASK truth for everything.)

What Shortcuts Actually Do

Shortcuts let you string together actions from different apps to create custom workflows. You can trigger them with a tap, a voice command, or even automatically based on time or location. The real magic happens when you sync them to your Apple Watch—suddenly you’ve got powerful automation sitting on your wrist.

Here’s what makes them great for authors:

• No third-party app downloads required (they’re built into iOS)

• Free to use (no subscription or premium tier)

• Work offline (your phone handles everything locally)

• Sync across all your Apple devices automatically

Setting Up Your Notion Task Shortcut

The goal here is simple: say “Hey Siri, Don’t Forget” to your watch, dictate what you need to do, and have it land in your Notion database instantly. Here’s how to build it.

Step 1: Create a New Shortcut

Open the Shortcuts app on your iPhone. Tap the plus icon in the upper right corner to start a new shortcut.

Step 2: Search for Notion

In the search actions bar, type “Notion.” (If you haven’t downloaded Notion or connected it yet, do that first—the shortcut won’t work without it.)

Step 3: Choose “Create Document Without Opening”

This is the part that makes it seamless. Your task gets created in the background without forcing you through extra screens.

Step 4: Set Up the Text Input

Now you’re at a screen for “Create a page with text.” When you get to the text field, choose “Ask Every Time” from the dropdown. This tells the shortcut to wait for you to dictate what the task actually is.

For the body field, just type something simple (even a single word works). This prevents you from having to deal with a two-step process every time.

Step 5: Choose Your Database

Select which Notion database you want tasks to land in. If you’re using my Notion setup, this will be your “Tasks Current” database or “Tasks PARA” depending on how you’ve organized things. (If you want to see how I’ve structured my Notion workspace, I’ve got a full video here)

Click done when you’re finished.

Step 6: Rename Your Shortcut

This is critical. Go back and rename your shortcut to something easy for Siri to recognize. I use “Don’t Forget” because it’s natural to say and Siri never misunderstands it. Whatever you choose, make it short and distinct.

Step 7: Turn It On for Your Apple Watch

Open the details page for your shortcut (it’s the little ‘i’ like informatio) and make sure it’s toggled on for your Apple Watch. Without this step, your watch won’t know the shortcut exists.

Now you can raise your watch and say, “Hey Siri, Don’t Forget,” then dictate your task. It gets added to your Notion database automatically while you move on with your life.

Empty Your Brain - It’s Psychologically Programmed to Relax When You Get it Down Somewhere.

Your brain is terrible at holding onto tasks. It’s not designed for that. What it’s really good at is spinning in anxious loops, constantly reminding you “don’t forget the thing, don’t forget the thing, don’t forget the thing” until you want to scream.

The moment you capture that task somewhere external—your watch, your phone, a database—your brain exhales. It stops the mental overdrive because it knows the information is safe. You’re not going to forget anymore. The loop breaks.

You need a capture system fast enough to match the speed of your thoughts. This setup takes five minutes to build and gives your brain permission to let go of the mental clutter.

Plus, it works for any database-style app that integrates with Shortcuts, not just Notion. Adapt it for Airtable, Todoist, Things, or whatever task manager you’re using. The tool doesn’t matter—what matters is getting thoughts out of your head before they start looping.

Watch the Walkthrough

I’ve recorded a quick video showing you the exact steps:

If you’ve been sleeping on Shortcuts, this is your wake-up call. They’re not just for tech nerds—they’re for anyone who wants to reclaim tiny pockets of time and mental energy.

And if you figure out how to train a bot to do my holiday shopping, please share. I’m running out of time.

Stay organized (or at least try to), Chelle

P.S. If you’ve built any useful Shortcuts for your author business, hit reply and tell me about them. I’m always looking for new ways to automate the boring stuff so I can focus on the work that actually matters.

Six months ago, Claude was a really smart conversation buddy. Today, it’s quietly running half my business. That shift didn’t happen because Claude suddenly got wiser. It happened because Claude can finally do things instead of simply explaining them.

Model Context Protocol (MCP) is the reason. And it’s reshaping how I work every single day.

Life Before MCP: The Copy-Paste Tango

Every author knows this routine. You’re planning a book launch and suddenly juggling five tabs like you’re auditioning for Productivity Cirque du Soleil.

First, you check your calendar to see if you’re free the third week of March. Then you open your book tracking spreadsheet to update the launch date. Then you draft an announcement email—but you need the book description from that spreadsheet, so you copy it, paste it into Claude to help you write the email, copy Claude’s draft, paste it into Gmail. Then you remember you need to alert your launch team, so you open Slack and type out another message.

Four tools. Multiple browser tabs. Constant context-switching. And Claude can only help with the thinking parts—you’re still doing all the manual work of moving information between systems.

Or maybe you’re sending ARCs to your reviewer list. You have their names and email addresses in a spreadsheet. You want personalized emails for each person. So you copy the first name. Paste it into Claude. Get the email draft. Copy the draft. Paste it into Gmail. Add the email address manually. Attach the file. Send. Repeat for the next reviewer. And the next. And the next.

Forty-five minutes later, you’ve sent emails to twenty people and your brain is mush from the repetition.

Or you’re trying to organize files after your last launch. Cover variations, formatted manuscripts, marketing graphics, and sales reports are scattered across your Google Drive. You want to archive everything properly but you’re not sure what should go where.

Every task requires you to be the middleman—shuttling information back and forth between Claude and your actual tools. You’re the human API, the biological bridge between the AI’s brain and your business systems.

It’s exhausting. And it means Claude can only help you think about work, not actually do work.

Enter MCP: The USB-C Moment for AI

Model Context Protocol is like USB-C for AI. Just like USB-C gave us one universal port that works with everything from your phone to your monitor, MCP gives AI a standardized way to connect to your tools.

The AI doesn’t need to know the specifics of how Gmail works versus Outlook, or how Google Sheets differs from Airtable. It just uses MCP to say “I need this data” or “take this action,” and the service handles it.

This is the turning point. It shifts AI from “smart assistant” into “operations partner.”

Life After MCP: Claude as Your Operations Partner

With MCP, the copy-paste tango disappears.

You tell Claude “update the blurb for my latest book in my master spreadsheet,” and Claude finds the record, updates it, and saves it. You never open the spreadsheet.

You say “what’s on my calendar next Tuesday?” and Claude checks your actual calendar and tells you. No screenshots needed.

You ask “organize my Drive files from last month’s launch into the archive folder,” and Claude does it. You don’t touch Drive.

Claude stops being something you consult and becomes something that executes. You’re having a conversation, and the work is getting done in the background.

That’s the fundamental shift. You stop being the middleman between Claude and your tools.

Two Ways to Connect: Native Connectors and Zapier MCP

Claude offers two ways to add these connections, and I use both depending on what I need.

Native Claude Connectors are built directly into Claude.ai and Claude Desktop. These are rock-solid integrations for the most common tools: Google Drive, Gmail, and Google Calendar. They’re straightforward to set up, and they just work.

I use these for the tools I access constantly. My calendar lives in Claude now—I can ask about scheduling conflicts, find open time slots, or see what’s coming up without ever opening Google Calendar.

Zapier MCP is where things get wild. Zapier built an MCP server that connects Claude to 8,000+ apps through a single connection point. Instead of waiting for Anthropic to build an integration for every tool you use, you connect Claude to Zapier once, and then you have access to everything Zapier supports.

I use Zapier MCP for the tools that are critical to my business but don’t have native Claude integrations yet—like Airtable (where I track all my books, content calendar, and business metrics), Dropbox (where I store manuscript files and resources), and the full version of Slack (Zapier’s integration has more capabilities than the native one).

You’ll notice that Google Drive is connected through both the native connection and Zapier. That’s because I have multiple Google accounts. The native connection is my personal Google and the Zapier is my business.

Setting Up Native Connectors: The 5-Minute Version

Let’s start with the easy stuff. Getting Claude connected to Gmail, Calendar, and Google Drive takes about five minutes.

Step 1: Open Claude Connectors

In Claude.ai, click your profile in the bottom left, then select “Connectors” from the menu. (In Claude Desktop, it’s under Settings > Connectors.)

You’ll see a list of available integrations. The ones with the little plug icon are native connectors.

Step 2: Connect Google Drive

Click “Connect” next to Google Drive. A new window opens asking you to sign in to your Google account and grant permissions.

Claude will ask for permission to:

• See and download your Google Drive files

• See information about your Drive files

• See your Drive folders and their contents

Grant the permissions. The window closes, and you’re back in Claude with a green “Connected” status next to Google Drive.

Step 3: Connect Gmail

Same process. Click “Connect” next to Gmail. Sign in. Grant permissions.

Claude will ask for permission to:

• Read your emails

• Send emails on your behalf

• Manage drafts

Grant the permissions. You’re connected.

Step 4: Connect Google Calendar

One more time. Click “Connect” next to Google Calendar. Sign in. Grant permissions.

Claude will ask for permission to:

• View your calendars

• View events on your calendars

• Create, edit, and delete events

Grant the permissions. Done.

Try it out: Ask Claude “what emails do I have from the last two days that need responses?” or “what’s on my calendar for Friday?” and watch it pull the actual information.

Setting Up Zapier MCP: The Power User Version

Zapier MCP takes a bit more setup, but it’s worth it for the flexibility.

Step 1: Create Your Zapier MCP Server

Go to

https://mcp.zapier.com

and sign in with your Zapier account.

Click “New MCP Server.” A dialog opens.

In the “MCP Client” dropdown, select “Claude” (or whatever AI tool you’re using—Zapier MCP works with ChatGPT, Cursor, and others too).

Give your server a name. I called mine “Author Ops” because it handles my author business operations.

Click “Create MCP Server.”

Step 2: Add Your Apps

You’ll see a “Configure” tab with a big empty space that says “No tools added yet.”

Click “Add Tool.” A search bar appears with Zapier’s entire app library.

Search for “Airtable.” Select it. You’ll see a list of available actions: Create Record, Update Record, Find Record, etc.

Select the actions you want Claude to be able to use. I select all tools:

• Find Record (so Claude can look up book details, content calendar entries, etc.)

• Update Record (so Claude can update fields when I ask)

• Create Record (so Claude can add new entries)

Click “Add Tool” and repeat for your other apps. I added:

• Dropbox: Upload File, Find File, Create Folder, Move File

• Slack: Send Channel Message, Send Direct Message, Find Message

Each time you add a tool, Zapier will ask you to authenticate with that service. Click “Connect,” sign in, grant permissions. Standard OAuth flow.

Step 3: Connect to Claude

Once you’ve added all your tools, click the “Connect” tab at the top.

You’ll see instructions specific to Claude. It gives you a URL that looks like: https://mcp.zapier.com/mcp/YOUR-SECRET-KEY

Copy that URL. (Don’t share it with anyone—it’s like a password.)

Open Claude.ai. Go to Connectors. Scroll down and you’ll see a section called “Custom MCP Servers” or “Add MCP Server.”

Click “Add Server.” Paste your Zapier MCP URL. Give it a name (like “Zapier MCP”).

Click “Connect.”

Claude will test the connection. If it works, you’ll see a green “Connected” status and a list of all the tools you added in Zapier.

Step 4: Test It

Ask Claude something like: “Find my book titled ‘Shadow’s Edge’ in my Airtable base and tell me the current blurb.”

Claude will use the Zapier MCP connection to search your Airtable, find the record, and show you the blurb. Magic.

Four Practical Examples of What This Actually Looks Like

Example 1: Content Calendar Management

Before MCP, managing my content calendar meant opening Airtable, scrolling through records, copying information into Claude for help with planning, then manually updating fields based on Claude’s suggestions.

Now I say: “Show me what’s scheduled for Author Automations in December” and Claude pulls the data from Airtable. Then I say “move the MCP article to December 15th and mark it as ‘In Progress’” and Claude updates the record. I never open Airtable.

Example 2: Launch Coordination

I’m launching a new course next year. I asked Claude: “Check my calendar for the week of January 20th, pull the launch checklist from my Dropbox folder called ‘Courses,’ and create a Slack reminder in the #launches channel three days before.”

Claude checked my calendar, found the Dropbox file, read the checklist, and scheduled the Slack message. Three tools, one conversation, zero tab-switching.

Example 3: Inbox Zero (Finally)

Every morning I tell Claude: “Show me unread emails from the last 24 hours, categorize them by priority, and draft responses for the top three.”

Claude scans my Gmail, uses context from my past emails to determine what’s urgent, and writes draft responses. I review, make tweaks if needed, and Claude sends them. I’ve cut my inbox processing time by 60%.

Example 4: File Organization

After every webinar, I have a mess of recordings, slides, transcripts, and attendee lists scattered across my Drive and Dropbox. I used to spend 20 minutes organizing everything.

Now I say: “Find all files from today’s webinar, move the recording and slides to the ‘Webinars 2025’ folder in Drive, move the transcript to ‘Transcripts’ in Dropbox, and add the attendee list to my Airtable CRM.”

Claude handles it.

The Bigger Picture

I started the Author Automations newsletter to document what was working in my own systems—the automations in Make.com and Zapier that kept my creative business from catching fire at 2 a.m. Somewhere along the way, it became a resource for people who want to build smarter systems, not just dream about them.

MCP represents the next evolution of that. It connects AI to action, conversation to execution. As more tools adopt it, the possibilities expand exponentially.

MCP is one of the most practical implementations of AI in day-to-day business use available right now. The protocol is gaining momentum because it solves a real problem: how do we give AI access to real-world systems without building a thousand custom integrations?

Chelle

P.S. – If you hit any snags setting up MCP connections, drop a comment. I’ll either answer directly or fold it into the next deep dive.