Sentry Skills

Install

claude plugin marketplace add getsentry/skills
claude plugin install sentry-skills@sentry-skills

This is third-party code. Review the repository files before installing.

What it does

The working skills of Sentry engineers, published as a Claude Code plugin and a skills.sh package. General skills fit any project: finding bugs and vulnerabilities in branch changes, security review, GitHub Actions workflow exploit audits, code simplification, iterating a PR to green CI with review feedback addressed, writing PR descriptions, maintaining a concise AGENTS.md, auditing Claude Code permissions and scanning third-party skills for dangerous content. Some skills are tied to internal Sentry processes: brand guidelines, the blog, error queue triage and SR&ED tax reports.

Who it is for. For developers who want review and PR skills proven in a large engineering team.

Example request

Find bugs and security issues in this branch's changes against main

Limitations

The README calls these skills for Sentry employees, so some rules and branch and commit conventions reflect internal standards. Some skills require gh, Linear or Notion. Claude Code needs a restart after install.

How to disable. Remove the plugin via /plugin or claude plugin uninstall sentry-skills@sentry-skills.

Security check

• Skills run git and gh and create branches, commits and PRs
• They read the whole project code during review

README in short

The README points to a separate repository for Sentry setup skills and covers installing the plugin via the Claude Code marketplace, updating it, and installing via npx skills for Cursor, Cline, GitHub Copilot and other agents. It lists all skills and two sub-agents, local development steps and the repository layout. Apache 2.0 licensed.

SKILL.md

---
name: agents-md
description: Creates and maintains concise AGENTS.md and CLAUDE.md project instruction files. Use when asked to create AGENTS.md, update AGENTS.md, maintain agent docs, set up CLAUDE.md, document repository agent conventions, or keep coding-agent instructions minimal and reference-backed.
---

# Maintaining AGENTS.md

Goal: concise, actionable agent instructions. Target under 60 lines; never exceed 100.

## Workflow

1. Inspect before writing:
   - package manager: lock files and manifests
   - commands: `package.json`, `Makefile`, task runners, CI workflows
   - docs/specs/policies: `README.md`, `CONTRIBUTING.md`, `docs/`, `specs/`, `policies/`, `SECURITY.md`, `.github/`
   - conventions: current code patterns, test layout, generated files, legacy areas to avoid
2. Choose scope:
   - root `AGENTS.md`: repo-wide defaults
   - nested `AGENTS.md`: only when a subtree has different commands or rules
   - closest instruction file wins; keep narrower files shorter than root files
3. Write the smallest useful file.
4. Verify exact paths and commands exist.

## File Setup

- Create `AGENTS.md` at the repository root.
- If a Claude-compatible entrypoint is required, symlink `CLAUDE.md` to `AGENTS.md`.
- Do not maintain divergent `AGENTS.md` and `CLAUDE.md` copies.

Related