npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

agent-rules-init

v0.7.0

Published

Generates CLAUDE.md, AGENTS.md, copilot-instructions.md and prompt files from what your repo actually is.

Vulnerabilities

Powered byPowered by Snyk
  • 0High
  • 0Medium
  • 0Low

Links

Git

Maintainers

rafa4787rafa4787

Keywords

cliclaudeclaude-codecodexcopilotai-agentsagent-rules

Readme

agent-rules-init

This is the publishable CLI package from the agent-rules-init repository. The root README is the canonical source for the complete feature, configuration, security and contribution documentation.

The operating model is intentionally simple: generate staging files, review them, then activate them with npx agent-rules-init --apply. Existing active files are backed up before replacement.

Generate repository-specific instructions for Claude Code, Codex, GitHub Copilot, Cursor, Gemini CLI and Windsurf from the manifests and commands that already exist in your project.

npx agent-rules-init

The CLI detects JavaScript/TypeScript, Python, Java, PHP, Ruby, Go, Rust, C#/.NET, Kotlin, Swift, Dart/Flutter, C/C++, Elixir, Scala and R projects. It also understands mixed repositories and npm, pnpm, Yarn and Bun workspaces, generating package-scoped AGENTS.generated.md files.

All output is written with a .generated suffix and existing files are never overwritten during generation. Review them and run --apply to activate with safe backups.

Use --lang es or --lang en to select the output language. See the full documentation for supported frameworks, generated files and contribution instructions.

Automation is supported through --dry-run, --force, --apply, --check, --json and --non-interactive. Repository defaults, enrichment assistant/model and per-project overrides can be stored in .agent-rules-init.yml.

Use --apply after review to activate staging files; replaced finals are backed up under .agent-rules-init/backups/.

Complete generations store a git-ignored hash receipt so --check can verify deterministic or enriched active files without rerunning an assistant. Activated final files take precedence over staging during checks.

AI enrichment runs assistants with read-only restrictions and falls back to deterministic output when invocation or validation fails. Review generated files before activation, especially when the repository did not come from a trusted source.

Verified enrichment is reused when repository inputs, assistant/model and accepted staging hashes are unchanged. Use --no-enrich-cache for a deliberate fresh run, and bound latency with --enrich-timeout <seconds> plus --enrich-retries <0..2>.

Semantic validation rejects prompt-injection language, new sections, destructive instructions, commands absent from repository facts and evidence paths outside the repository. Metrics report cache hits, changed lines and security rejections in addition to batch, retry, input-size, fallback and duration data.

Generated documents share an evidence-backed model but are not duplicates: Claude receives broad repository context, AGENTS receives operational commands and scope, and Copilot receives concise implementation conventions. Observed architecture and local conventions include their source files so specific claims can be audited.

License

MIT