agentleFS
ntaffziintaffzii/Skill-Agents/skills/in-progress/SKILL.md

in-progress

Skill4 starsChanged 2 months agoMITView on GitHub

Manage draft skills that are not ready to ship yet. Use when the user asks to prototype, sketch, test, evaluate, or promote an experimental skill from in-progress into a stable skill bucket.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill for draft skills that need iteration before becoming public or team-ready.

    line 8
  2. Distinctive:

    Keep experiments small enough to promote or retire quickly.

    line 17
  3. Distinctive:

    Do not add large references until the core workflow works.

    line 18
  4. Distinctive:

    Do not promote a skill without realistic trigger examples.

    line 19
  5. Distinctive:

    Prefer workflow steps over broad advice.

    line 30
  6. Distinctive:

    Add stop conditions when misuse is likely.

    line 31
  7. Distinctive:

    Test with realistic prompts

    line 33
  8. Distinctive:

    Keep in progress when examples still fail or scope is unclear.

    line 40
  9. Distinctive:

    Update README references if the skill is public.

    line 45
  10. Distinctive:

    Update workflows or registries only when the skill is part of a repeatable job.

    line 46

≈ “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

In Progress Skill Incubator

Use this skill for draft skills that need iteration before becoming public or team-ready.

Purpose

skills/in-progress/ is a staging area. A draft belongs here when the idea is useful but the trigger, workflow, scope, or validation is still uncertain.

Rules

  • Draft skills should not be listed as stable skills in the root README.
  • Keep experiments small enough to promote or retire quickly.
  • Do not add large references until the core workflow works.
  • Do not promote a skill without realistic trigger examples.

Workflow

  1. Define the experiment
    • State the repeated task the draft skill should improve.
    • Name the expected trigger phrases.
    • Identify what the agent should do differently after loading it.
  2. Draft narrowly
    • Give the skill one main job.
    • Prefer workflow steps over broad advice.
    • Add stop conditions when misuse is likely.
  3. Test with realistic prompts
    • Try at least three prompts that should trigger the skill.
    • Try one prompt that should not trigger it.
    • Note where the instructions are too vague or too broad.
  4. Decide status
    • Promote to a domain bucket when the workflow is stable and useful.
    • Keep in progress when examples still fail or scope is unclear.
    • Move to deprecated/ when the idea is no longer worth pursuing.
  5. Promote cleanly
    • Move the skill folder into the right bucket.
    • Update README references if the skill is public.
    • Update workflows or registries only when the skill is part of a repeatable job.

Promotion Checklist

  • name is lowercase and hyphenated.
  • description clearly says when to use the skill.
  • The body has a workflow, rules, and verification or output guidance.
  • The skill has one main job.
  • It does not duplicate another stable skill.
  • Any long examples or references are moved into references/.

Output Format

End with:

  • Draft status
  • What changed
  • What still needs testing
  • Promote, keep, or retire recommendation

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.