agentleFS
inklateinklate/social-skills/skills/idk/SKILL.md

idk

Skill8 starsChanged 4 months agoMITView on GitHub

Point to the right social skill for what the user is trying to do, and map how the whole social-* family fits together across LinkedIn, X (Twitter), Instagram, Facebook, Threads, TikTok, YouTube, Bluesky, and Mastodon. Use when the user asks "which social skill should I use", "what can you do for my social media", "where do I start", "help me pick a skill", or "what social skills are there". Reads social-context.md to tell whether setup is done, and always points first-time users to the social-context skill before anything else.

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Read social-context.md at the project root (also check .agents/social-context.md).

    line 27
  2. Very specific:

    Check the prerequisite: If social-context.md is missing, recommend social-context first, then the task skill.

    line 39
  3. Very specific:

    Never route past setup: if social-context.md is missing, social-context comes first.

    line 72
  4. Distinctive:

    Always include the trigger phrase, so the user can act without asking how.

    line 74

“Common” is measured: kept by at least 6 in 10 of the 3 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

Map what the user wants onto the one social skill whose procedure fits, and hand them the phrase that triggers it. Route, don't perform — the named skill does the work.

Context

Read social-context.md at the project root (also check .agents/social-context.md). If it is missing, the user hasn't been set up: your first recommendation is always the social-context skill ("set up my social context"), because every other skill reads that file. If it exists, setup is done — route straight to the task skill.

Workflow

  1. Name the job in one line. Restate what the user is trying to produce or decide — one post, a thread, a week of content, a plan, ad copy, a fix, an audit. If it is ambiguous, ask one question; don't guess across two different procedures.
  2. Match the job to exactly one skill using the Skill map below. Match on the procedure the user needs, not the platform — platform is a constraint each skill already handles.
  3. Check the prerequisite. If social-context.md is missing, recommend social-context first, then the task skill. If the task leans on voice and the Voice section is thin, suggest social-voice alongside.
  4. Hand off. Name the skill, say in one line why it fits, and give the trigger phrase the user can say to invoke it. If two skills genuinely both fit (one post vs many), present the pair with the one-line difference and let the user pick.

Skill map

The user wants to…SkillTrigger phrase
Set up who they are / audience / pillars / red linessocial-context"set up my social context"
Teach the agent their writing voicesocial-voice"learn my voice"
Write one post for one platformsocial-post"write a LinkedIn post about…"
Write a thread / multi-post sequencesocial-thread"turn this into an X thread"
Build an Instagram/LinkedIn carouselsocial-carousel"make a carousel about…"
Get scored hook options for a draftsocial-hook"give me hooks for this"
Adapt one message across many platforms at oncesocial-crosspost"post this on LinkedIn, X, and Instagram"
Mine one source into a week of different postssocial-repurpose"turn my blog into a week of posts"
Triage and reply to comments/mentions/DMssocial-reply"help me reply to these comments"
Write paid ad copy variantssocial-ad"write Facebook ad variants"
Plan a 2-week content calendarsocial-calendar"plan my next two weeks"
Validate a draft before postingsocial-check"will this work on Instagram?"
Audit a profile / content history vs goalssocial-audit"audit my LinkedIn"

The two easily-confused pairs:

  • social-post vs social-crosspost — one platform, deeply (two hook variants) vs the same message adapted across several platforms at once.
  • social-crosspost vs social-repurpose — the same message on many platforms now vs many different posts mined from one source over a week.

Quality bar

  • Recommend exactly one skill per job unless two genuinely fit — then name both with the one-line difference.
  • Never route past setup: if social-context.md is missing, social-context comes first.
  • Match on procedure, not platform; every skill handles its own platforms.
  • Always include the trigger phrase, so the user can act without asking how.

Deliverable

A one-line recommendation: the skill to use, why it fits in a phrase, and the exact trigger to say — plus the foundation prerequisite if setup isn't done. Then hand off; the named skill does the work.

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.