agentleFS
ramkesh430ramkesh430/facebook-agent-skill/skills/fb-group/SKILL.md

fb-group

Skill1 starsChanged 16 days agoMITView on GitHub

Run or post in Facebook Groups - start a Group, write its rules, welcome posts and weekly prompts, and write posts for other people's Groups that add value instead of reading as spam. Use when the user mentions Facebook Groups, community, "start a group", "post in a group", "group rules", "grow my group", or wants local word-of-mouth.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the rules first: Ask the user to paste them.

    line 44
  2. Distinctive:

    Read the last 20 posts: what gets replies here?

    line 46
  3. Distinctive:

    Write a post that is useful on its own: an answer, a local tip, a specific question, a lesson with a number.

    line 47
  4. Distinctive:

    Never post the same text in multiple Groups.

    line 51
  5. Distinctive:

    No engagement bait, no "comment INTERESTED", no fake questions to plant a recommendation.

    line 56
  6. Distinctive:

    Never write posts pretending to be a customer, or sock-puppet replies.

    line 58
  7. Very specific:

    Run everything through /fb-human and lint with ../fb-post/post.py.

    line 59

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

fb-group

Groups are where Facebook still behaves like a community: people read, reply and recommend. They are also where self-promotion gets people removed. Two modes.

Mode 1: the user's own Group

Setup (write all of it, the user creates it):

  • Name: the member's identity or problem + place, not the brand. "Kathmandu Home Bakers" beats "SweetBakes Community".
  • Description: who it's for, what happens here, what doesn't.
  • 3 membership questions: one qualifying, one about their goal, one for consent to rules.
  • 5-8 rules: short, specific, enforceable. Include the promo rule.
  • Welcome post, pinned: how to introduce yourself, with the admin's own intro as the example.

Weekly rhythm (prompts the admin posts):

Mon  question about their week / goal
Wed  a useful post from the admin (tip, template, local info)
Fri  wins thread: members share one result
Sat  member spotlight or Q&A

Prompts must be answerable in one sentence. "What's one thing that..." beats "Thoughts?". Admin replies to every early comment in the first hour.

Mode 2: posting in someone else's Group

  1. Read the rules first. Ask the user to paste them. If promo is banned, don't write promo in disguise.
  2. Read the last 20 posts: what gets replies here?
  3. Write a post that is useful on its own: an answer, a local tip, a specific question, a lesson with a number. No link unless the rules allow. The user's business appears only as context ("I run a bakery in Patan, so here's what we found...").
  4. Never post the same text in multiple Groups. Facebook treats it as spam and members notice. Write a different version per Group, or skip it.

Rules

  • No engagement bait, no "comment INTERESTED", no fake questions to plant a recommendation.
  • Never write posts pretending to be a customer, or sock-puppet replies.
  • Run everything through /fb-human and lint with ../fb-post/post.py.
  • The user posts. This skill writes.

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.