agent-instructions
v0.3.0Write, audit and slim down CLAUDE.md and AGENTS.md: what belongs in them versus docs, skills, hooks and CI, across organisation, repo, path and personal levels; one canonical AGENTS.md for every tool; a measuring script, generator-safe layouts for Laravel Boost and friends with a CI drift check, and a probe-task method to test changes.
By WebgripLicense: MIT0 GitHub starsUpdated 3 days ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 1 skill or MCP entry
- Source updated
- Oct 4, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install agent-instructions for Claude Code
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install agent-instructions@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/webgrip/ai-skillsClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is skills/agent-instructions/.
Plugin files
├── .claude-plugin/plugin.json└── SKILL.md
Included Skills1
Fixes, cleans up, reviews and writes CLAUDE.md and AGENTS.md agent instruction files and agent config. Use when cleaning up or slimming down a CLAUDE.md or AGENTS.md, when it is too long, bloated or ignored by Claude, when Claude keeps doing something the file forbids, when CLAUDE.md and AGENTS.md are duplicated or drift apart, when creating one, when composer install, boost:install or another generator wipes hand-written instructions, when rules in ~/.claude leak into other projects, when reviewing a PR that changes CLAUDE.md, AGENTS.md, .mcp.json or .claude/settings.json, when testing whether an instruction change works, when setting up cloud, CI or headless agents, or when deciding whether a rule belongs in CLAUDE.md, a skill, a hook, CI or docs. Covers .claude/rules, CLAUDE.local.md, .cursor/rules, copilot-instructions.md and Laravel Boost across Claude Code, Codex, Cursor, Copilot, opencode and OpenHands; ships measuring, security and A/B probe scripts.
Plugin manifests1
{
"name": "agent-instructions",
"displayName": "Agent Instructions",
"description": "Write, audit and slim down CLAUDE.md and AGENTS.md: what belongs in them versus docs, skills, hooks and CI, across organisation, repo, path and personal levels; one canonical AGENTS.md for every tool; a measuring script, generator-safe layouts for Laravel Boost and friends with a CI drift check, and a probe-task method to test changes.",
"version": "0.3.0",
"license": "MIT",
"author": {
"name": "Webgrip",
"url": "https://forgejo.webgrip.dev/webgrip"
},
"homepage": "https://forgejo.webgrip.dev/webgrip/ai-skills/src/branch/main/skills/agent-instructions",
"repository": "https://forgejo.webgrip.dev/webgrip/ai-skills",
"keywords": [
"claude-md",
"agents-md",
"context-engineering",
"instruction-files",
"laravel-boost",
"cursor-rules",
"copilot-instructions"
]
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[agent-instructions on Agent Plugins Marketplace](https://pluginsmp.com/plugins/agent-instructions)