agentleFS
taivoptaivop/marketplace/AGENTS.md

marketplace

AGENTS.md17 starsChanged 56 days agoMITView on GitHub

This repository is a dual-distribution marketplace for reusable agent plugins: Packages include thinking and writing skills and Estonian search tools. The public overview is taivo.ai/projects/skills/. When adding, removing, renaming, or materially changing a package, keep that overview's descriptions, examples, and links current. Skill files stay here. For each package, SKILL.md frontmatter is authoritative. Expected frontmatter fields: Generated files: Do not hand-edit generated files. Run generators instead. A PR should fail if:

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 32,671 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository7 other files it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

10 rules. About 0 are common, 6 are distinctive, and 4 are 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.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    When adding, removing, renaming, or materially changing a package, keep that overview's descriptions, examples, and links current.

    line 10
  2. Distinctive:

    Do not hand-edit generated files.

    line 51
  3. Very specific:

    Create the package directory at skills/<slug>.

    line 55
  4. Very specific:

    Add SKILL.md with valid frontmatter.

    line 56
  5. Very specific:

    Add any package-specific resources (agents/, scripts/, references/, assets/, docs).

    line 57
  6. Distinctive:

    Ensure clean generation check:

    line 64
  7. Distinctive:

    Commit package changes together with regenerated manifests.

    line 73
  8. Distinctive:

    Keep package names and plugin names kebab-case.

    line 91
  9. Very specific:

    Keep version increments explicit in metadata.distribution.plugin_version.

    line 92
  10. Distinctive:

    Keep package-specific operational rules inside that package's own docs (README.md, AGENTS.md).

    line 93

≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md 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
AGENTS.md

Marketplace Agent Guide

This repository is a dual-distribution marketplace for reusable agent plugins:

  1. Anthropic Claude Code marketplace plugins.
  2. Codex installable skills.

Packages include thinking and writing skills and Estonian search tools. The public overview is taivo.ai/projects/skills/. When adding, removing, renaming, or materially changing a package, keep that overview's descriptions, examples, and links current. Skill files stay here.

Core Layout

.
├── .claude-plugin/marketplace.json
├── skills/
│   └── <plugin-package>/
│       ├── SKILL.md
│       ├── .claude-plugin/plugin.json
│       └── ...
├── scripts/
│   ├── generate_plugin_manifests.py
│   ├── generate_marketplace.py
│   ├── validate_codex_skills.py
│   └── validate_anthropic_plugins.py
└── .github/workflows/validate-distribution.yml

Source Of Truth

For each package, SKILL.md frontmatter is authoritative.

Expected frontmatter fields:

  1. Required:
    • name
    • description
  2. Distribution metadata (recommended and used by generators):
    • metadata.distribution.publish_anthropic: true/false
    • metadata.distribution.plugin_name: Anthropic plugin name
    • metadata.distribution.plugin_version: SemVer (X.Y.Z)
    • metadata.distribution.plugin_author: display author name

Generated files:

  1. <package>/.claude-plugin/plugin.json
  2. /.claude-plugin/marketplace.json

Do not hand-edit generated files. Run generators instead.

Add A New Plugin Package

  1. Create the package directory at skills/<slug>.
  2. Add SKILL.md with valid frontmatter.
  3. Add any package-specific resources (agents/, scripts/, references/, assets/, docs).
  4. Set metadata.distribution.* values in SKILL.md.
  5. Run generation + validation:
    • python3 scripts/generate_plugin_manifests.py
    • python3 scripts/generate_marketplace.py
    • python3 scripts/validate_codex_skills.py
    • python3 scripts/validate_anthropic_plugins.py
  6. Ensure clean generation check:
    • python3 scripts/generate_plugin_manifests.py --check
    • python3 scripts/generate_marketplace.py --check

Update Existing Plugin Package

  1. Edit package content and/or SKILL.md.
  2. Regenerate manifests.
  3. Re-run validators.
  4. Commit package changes together with regenerated manifests.

CI Expectations

CI runs:

  1. Codex structure/frontmatter validation.
  2. Anthropic manifest/marketplace validation.
  3. Generation drift check (git diff --exit-code after generation).

A PR should fail if:

  1. SKILL.md metadata is invalid.
  2. Generated files are stale.
  3. Marketplace or plugin JSON is malformed.

Working Conventions

  1. Keep package names and plugin names kebab-case.
  2. Keep version increments explicit in metadata.distribution.plugin_version.
  3. Keep package-specific operational rules inside that package's own docs (README.md, AGENTS.md).

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.