agentleFS
astronomerastronomer/agents/AGENTS.md

agents

AGENTS.md459 starsChanged 36 days agoApache-2.0View on GitHub

After adding skills or making changes, reinstall the plugin: Skills are markdown files with YAML frontmatter in skills/<name>/SKILL.md: Plugin configuration is split across two files: Use ${CLAUDE_PLUGIN_ROOT} to reference files within the plugin (required because plugins are copied to a cache location when installed). Important: Hooks in SKILL.md frontmatter can use relative paths from the skill's directory (e.g., ./scripts/bar.py). Use ${CLAUDE_PLUGIN_ROOT} in plugin.json to reference the plugin root. This plugin uses ~/.astro/agents/ for user configuration (warehouse credentials, etc.).

The file

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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    After adding skills or making changes, reinstall the plugin:

    line 44
  2. Very specific:

    Use ${CLAUDE_PLUGIN_ROOT} to reference files within the plugin (required because plugins are copied to a cache location when installed).

    line 75

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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
<!-- START doctoc generated TOC please keep comment here to allow auto update --><!-- DON'T EDIT THIS SECTION, INSTEAD RE-RUN doctoc TO UPDATE -->

Table of Contents generated with DocToc

  • Claude Code Plugin Development
    • Plugin Structure
    • Installing the Plugin
    • Skills
    • Configuration
    • Key Files
    • Config Location
<!-- END doctoc generated TOC please keep comment here to allow auto update -->

Claude Code Plugin Development

Plugin Structure

project-root/
├── .claude-plugin/
│   ├── marketplace.json        # Marketplace catalog (strict: false)
│   └── plugin.json             # Plugin manifest with hooks
└── skills/                     # Skills (auto-discovered)
    └── skill-name/
        ├── SKILL.md            # Skill with YAML frontmatter
        └── hooks/              # Hook scripts (co-located with skill)
            └── *.sh

Installing the Plugin

# Add the marketplace (from repo root)
claude plugin marketplace add astronomer/agents

# Install the plugin
claude plugin install astronomer-data@astronomer

# Or test locally (session only)
claude --plugin-dir .

After adding skills or making changes, reinstall the plugin:

claude plugin uninstall astronomer-data@astronomer && claude plugin marketplace update && claude plugin install astronomer-data@astronomer

Skills

Skills are markdown files with YAML frontmatter in skills/<name>/SKILL.md:

---
name: skill-name
description: When to use this skill (Claude uses this to decide when to invoke it)
---

# Skill content here...
  • Skills are auto-discovered from the skills/ directory
  • Claude invokes skills automatically based on the description matching user requests
  • Users can also invoke directly with /plugin-name:skill-name (e.g., /astronomer-data:authoring-dags)

Configuration

Plugin configuration is split across two files:

  • hooks: Inlined in .claude-plugin/plugin.json, scripts co-located in skills/<name>/hooks/
  • mcpServers: Inlined in .claude-plugin/plugin.json
  • skills: Auto-discovered from skills/ directory
  • marketplace metadata: Defined in .claude-plugin/marketplace.json (strict: false)

Use ${CLAUDE_PLUGIN_ROOT} to reference files within the plugin (required because plugins are copied to a cache location when installed).

Important: Hooks in SKILL.md frontmatter can use relative paths from the skill's directory (e.g., ./scripts/bar.py). Use ${CLAUDE_PLUGIN_ROOT} in plugin.json to reference the plugin root.

Key Files

  • .claude-plugin/plugin.json - Plugin manifest with hooks and mcpServers
  • .claude-plugin/marketplace.json - Marketplace catalog (metadata only, strict: false)
  • skills/*/SKILL.md - Individual skills (auto-discovered)
  • skills/*/hooks/*.sh - Hook scripts (co-located with skills, referenced via relative paths from SKILL.md or ${CLAUDE_PLUGIN_ROOT}/skills/<name>/hooks/... from plugin.json)

Config Location

This plugin uses ~/.astro/agents/ for user configuration (warehouse credentials, etc.).

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.