mirror of
https://github.com/getsentry/skills.git
synced 2026-09-14 20:00:03 +08:00
David Cramer
9c0e34277f
* fix(skills): Standardize allowed-tools to comma-separated format Claude Code docs specify comma-separated format for allowed-tools. Update all skill frontmatters, documentation, and the scan_skill.py parser to use and support comma-separated tool lists. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(skills): Remove model attribute from skill documentation Skills should not hardcode model overrides — let users control which model runs their skills. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
60 lines
2.2 KiB
Markdown
60 lines
2.2 KiB
Markdown
# Sentry Agent Skills
|
|
|
|
A collection of agent skills for use by Sentry employees, primarily designed for [Claude Code](https://claude.ai/claude-code).
|
|
|
|
This repository is structured as a Claude Code plugin (see `plugins/sentry-skills/`), but the skills themselves follow the open [Agent Skills specification](https://agentskills.io) format to maintain compatibility with other tools that adopt the standard.
|
|
|
|
## Structure
|
|
|
|
```
|
|
plugins/sentry-skills/skills/<skill-name>/SKILL.md
|
|
```
|
|
|
|
Each skill is a directory containing a `SKILL.md` file with YAML frontmatter (`name`, `description`) and markdown instructions.
|
|
|
|
## Creating a Skill
|
|
|
|
1. Create `plugins/sentry-skills/skills/<skill-name>/SKILL.md`
|
|
2. Add YAML frontmatter (see below)
|
|
3. Write clear instructions in markdown
|
|
4. Update `README.md` to include the new skill in the Available Skills table
|
|
5. Update the skills allowlist in `plugins/sentry-skills/skills/claude-settings-audit/SKILL.md`
|
|
6. Add the skill to `.claude/settings.json` in the `permissions.allow` array as `Skill(sentry-skills:<skill-name>)`
|
|
|
|
### Frontmatter
|
|
|
|
**Important:** The YAML frontmatter must be at the very beginning of the file. Do not place comments or any other content before it—parsers expect `---` as the first line.
|
|
|
|
**Required:**
|
|
- `name` - kebab-case, 1-64 chars
|
|
- `description` - up to 1024 chars, include trigger keywords
|
|
|
|
**Optional:**
|
|
- `allowed-tools` - comma-separated list of permitted tools
|
|
- `license` - license name or path (for attribution, place a LICENSE file in the skill directory)
|
|
- `compatibility` - environment requirements (max 500 chars)
|
|
|
|
```yaml
|
|
---
|
|
name: example-skill
|
|
description: What this skill does and when to use it. Include trigger keywords.
|
|
allowed-tools: Read, Grep, Glob, Bash
|
|
license: LICENSE
|
|
---
|
|
|
|
<!-- Attribution comments go AFTER the frontmatter -->
|
|
|
|
# Example Skill
|
|
|
|
Instructions for the agent.
|
|
```
|
|
|
|
## Skill Design Guidelines
|
|
|
|
When writing skills that include Python scripts, always instruct the agent to use `uv run <script>` instead of `python <script>` or `python3 <script>`.
|
|
|
|
## References
|
|
|
|
- [Agent Skills Spec](https://agentskills.io/specification)
|
|
- [Sentry Engineering Practices](https://develop.sentry.dev/engineering-practices/)
|