agentleFS
Sign inStart free
Amine-MhiriAmine-Mhiri/creatorscompanion-plugin/skills/setup/SKILL.md

setup

Skill0 starsChanged 6 days agoMITView on GitHub

Set up the user's CreatorsCompanion account from the chat in about a minute, so drafts sound like them. Checks what is already set up, asks at most 3 questions in one message (niche and who they help, content language, a few of their own posts), saves the profile and voice examples, then offers a first draft. Use right after install, when the user says get started or set me up, or when drafts come back with a tip about a thin profile. Setup itself is free; it never spends credits without a yes.

191 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use add-voice-transcription, add-discord and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • juancruzmunozalbeloSkill

    add-voice-transcription

    Add voice message transcription to NanoClaw using OpenAI's Whisper API. Automatically transcribes WhatsApp voice notes so the agent can read and respond to them.

    Also in 5 projects that use this

  • MicrosoftSkill

    add-discord

    Add Discord bot channel integration via Chat SDK.

    Also in 11 projects that use this

  • MicrosoftSkill

    add-slack

    Add Slack channel integration via Chat SDK.

    Also in 11 projects that use this

  • juancruzmunozalbeloSkill

    add-telegram-swarm

    Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".

    Also in 7 projects that use this

  • +8 more from microsoft

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 3 are distinctive, and 7 are very specific to this project.

  • 0

    Common

    Most of the 171 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Call get_started. It returns, in one call: the credits and what is left of today's limit, the profile fields present and missing, the …

    line 14
  2. Very specific:

    Ask in a single message, and only for what get_started says is missing.

    line 24
  3. Very specific:

    Do not ask for anything else, such as their name or audience, unless the user offers it.

    line 35
  4. Very specific:

    Profile: call update_creator_profile with the fields the answer gives: bio (one or two lines on who they help and how, in their words), …

    line 40
  5. Very specific:

    Voice from pasted posts: call update_voice_examples with the format the posts were written for (tiktok for Reels and TikTok scripts, …

    line 44
  6. Very specific:

    If get_started showed that the user already saved their own examples for that format, say these will replace example 1 onward and ask first.

    line 47
  7. Very specific:

    Voice from a handle: call list_my_posts with the platform (instagram, tiktok, youtube, linkedin or x) and the handle.

    line 50
  8. Distinctive:

    Ask the questions in one message, never one at a time.

    line 78
  9. Distinctive:

    Use the user's own words for the bio and niche.

    line 79
  10. Distinctive:

    If a save fails, say what did not save and offer to try again.

    line 80

“Common” is measured: kept by at least 6 in 10 of the 171 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Set up CreatorsCompanion

Get the user from a fresh account to a first draft in their own voice, with as few messages as possible. Every draft is written from the profile and voice examples saved here, so a minute spent on setup is what makes the drafts sound like them.

1. See what is set up (free)

Call get_started. It returns, in one call: the credits and what is left of today's limit, the profile fields present and missing, the creator's own voice examples per format, the Creator Brain status, library and draft counts, and next_steps.

  • Everything is set (no missing fields, own voice examples saved): say so in one line, show the suggested next_steps, and skip to step 4.
  • Something is missing: go to step 2 and ask only about what is missing.

2. Ask once, at most 3 questions

Ask in a single message, and only for what get_started says is missing. At most these three:

  1. Niche and who they help: "What do you post about, and who do you help (and with what result)?"
  2. Content language: "Which language do you post in?" Skip it when the profile already has one or the user is clearly writing in it; then confirm it in one line instead of asking.
  3. Their own posts: "Paste 1 to 3 posts or scripts you wrote, the ones that sound most like you."
<!-- full-server -->

Or offer the shortcut: "Or give me your Instagram, TikTok, YouTube, LinkedIn or X handle and I can pull your recent posts."

<!-- /full-server -->

Do not ask for anything else, such as their name or audience, unless the user offers it; ask nothing that get_started already answered. If the user answers only part, save that part and continue.

3. Save what they told you (free)

  1. Profile: call update_creator_profile with the fields the answer gives: bio (one or two lines on who they help and how, in their words), niche, audience, language (a code or name, such as en, fr or es), and name or social_handles if they mentioned them. Only the fields you send change. Read back the fields it changed and the ones still missing.
  2. Voice from pasted posts: call update_voice_examples with the format the posts were written for (tiktok for Reels and TikTok scripts, linkedin, twitter for X, youtube or newsletter) and the posts as examples, best first, one post per item. Calls to action they use go in ctas. If get_started showed that the user already saved their own examples for that format, say these will replace example 1 onward and ask first. Posts for different formats need one call per format.
<!-- full-server -->
  1. Voice from a handle: call list_my_posts with the platform (instagram, tiktok, youtube, linkedin or x) and the handle; it is free. Show the posts briefly and let the user pick up to 5. Then tell the user, before anything runs: "Importing them costs 100 credits and replaces the voice examples saved for that format (example 1 onward)." Only after a yes, call import_voice_from_posts with the same platform and handle and posts as a list of url and text items copied from list_my_posts, best first. If the user says no, offer to paste the posts instead, which is free.
<!-- /full-server -->

Then confirm in two or three lines what was saved: the profile fields, and how many voice examples for which format.

4. Offer a first draft

Offer one concrete first draft from their niche, for example: "Want me to write your first Reel? Give me a topic, or I can use: <one idea from their niche>." When they pick a topic, follow the write-in-my-voice skill: check the cost with estimate_cost, tell them, and call generate_content only after they say yes.

Mention once, without pushing, that a Creator Brain can come later: they paste their offers, story and proof with add_brain_document, and build_creator_brain (it uses credits, shown before it runs) turns them into a brief every draft follows.

Rules

  • Never spend credits without a yes. Profile and voice saves are free. generate_content and build_creator_brain use credits, and so does any other paid tool: state the cost and wait for the user to agree before calling it.
  • Ask the questions in one message, never one at a time.
  • Use the user's own words for the bio and niche; do not polish them into marketing copy.
  • If a save fails, say what did not save and offer to try again. Do not send the user to the website for something a tool here can do.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free