agentleFS
syangasyanga/astack/.agents/skills/author-skill/SKILL.md

author-skill

Skill0 starsChanged 57 days agoMITView on GitHub

Create, import, or modify skills in the astack repository. Use for changes to skill instructions, invocation, supporting resources, or callers.

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

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

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 4

    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 the existing skill and its callers before adding another skill.

    line 9
  2. Very specific:

    Put shared skills in skills/ and repository-local skills in .agents/skills/.

    line 10
  3. Distinctive:

    Read writing-for-agents and its skill-mechanics guidance.

    line 12
  4. Distinctive:

    After writing or modifying the skill, run two independent subagent reviews:

    line 18
  5. Distinctive:

    If subagents are unavailable, perform both reviews yourself and report that they were not independent.

    line 31
  6. Distinctive:

    Start upstream imports from the original.

    line 42
  7. Very specific:

    When adapting an upstream skill, record its repository, path, commit, and any source-to-destination mappings in upstream/manifest.json.

    line 48

“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

Author an astack skill

  1. Identify the requested behavior and the smallest change that achieves it. Read the existing skill and its callers before adding another skill. Put shared skills in skills/ and repository-local skills in .agents/skills/. For upstream material, follow upstream imports.
  2. Read writing-for-agents and its skill-mechanics guidance.
  3. Edit the source, reusing existing guidance through pointers. When moving guidance, update its callers and preserve the decisions it controls. Apply technical-writing and unslop to the diff.
  4. After writing or modifying the skill, run two independent subagent reviews:

    • Writing: apply the linked writing guidance. Check for redundancy across skills, unnecessary rules, excessive prose, and misplaced guidance.
    • Behavior: check the request against the skill and its callers. Trace changed invocation, stopping conditions, invalidated acceptance results, and external writes through linked workflows. For changed behavior, exercise a representative task and a case where the new behavior must not run.

    Give both reviewers the request, diff, repository rules, and referenced files without suggested conclusions. Reviewers report concrete findings without editing or delegating. Assess the findings, fix demonstrated problems, and repeat affected reviews until clean or blocked. Report unresolved blockers. If subagents are unavailable, perform both reviews yourself and report that they were not independent.

  5. For approved changes, follow open-pr to submit a PR unless the request is limited to a proposal or local edits. Explain the behavior change, verification, and untested branches. Keep private transcript details out of the PR. Report publication blockers with the local diff. Merge only when authorized. If the PR merges, ask whether to reinstall unless the user already authorized it.

Upstream imports

Start upstream imports from the original. Preserve clear wording, structure, and examples. Make targeted changes for compatibility or a specific demonstrated problem, and compare each change against the original for clarity and meaning. For imported examples that claim to exclude invalid states, check a counterexample under the stated compiler or runtime assumptions.

When adapting an upstream skill, record its repository, path, commit, and any source-to-destination mappings in upstream/manifest.json. Keep its license in the skill directory. Limit manifest notes to rationale the diff cannot explain. To review upstream changes, clone the source and diff from the pinned commit.

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.