# Sentry

> Use when the user wants to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data; prefer the current host's Sentry connector when available, otherwise use the bundled read-only API helper with `SENTRY_AUTH_TOKEN`.

- Skill: `hebertzhu/sentry` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add hebertzhu/sentry`
- Raw SKILL.md: https://api.skillmd.com/api/skills/hebertzhu/sentry/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: hebertzhu (https://skillmd.com/u/hebertzhu)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/hebertzhu/sentry

---


# Sentry

## Intent
- Use for read-only Sentry issue inspection, error summarization, event lookups, and lightweight production-health triage.

## Default operating pattern
1. Confirm scope first: organization, project, environment, time range, and whether the user wants issue lists, issue detail, or event detail.
2. Prefer the current host's Sentry connector if it exists; otherwise use `scripts/sentry_api.py` with a locally configured `SENTRY_AUTH_TOKEN`.
3. Read before summarizing: list issues first, then drill into a specific issue or event only when needed.
4. Redact sensitive data and avoid dumping raw stack traces unless the user explicitly asks for deeper debugging detail.
5. Return a clear operational summary: what is happening, how severe it looks, and which issue or event should be investigated next.

## Bundled helper
- `scripts/sentry_api.py` provides deterministic, read-only Sentry API calls for issues and events.

## Pack fit
- Included in: `knowledge-ops`
- Keep the public version read-only, reproducible, and safe for production support workflows.

## Boundary
- Do not request users to paste tokens into chat; ask them to set `SENTRY_AUTH_TOKEN` locally.
- Do not mutate Sentry state or claim root cause unless the issue or event evidence supports it.