Skills DirectoryCurated open-source Skills from GitHub

sentrycommit d4a9df5SKILL.md · Monitoring & Ops

Sentry (Read-only)

Uses the sentry CLI to read Sentry issues and events, summarize recent production errors and pull basic health data, read-only.

Requires Sentry credentials (sentry auth login or SENTRY_AUTH_TOKEN) and an installed sentry CLI.

npx skills add openai/skills --skill sentry
Permissions
  • read-repo
  • write-files
  • shell
  • network
Compatible with
Codex
GitHub
28.4k
Source
Official · OpenAI
Status
Scan passed
License
Apache-2.0

Core rule

Perform read-only queries only, never echo auth tokens, and never ask the user to paste a token into chat.

Execution Phases

  1. Phase 01Check auth

    If not authenticated, ask the user to run sentry auth login or set SENTRY_AUTH_TOKEN locally; give install steps if the CLI is missing.

  2. Phase 02Query issues

    By default list 20 unresolved production issues from the last 24h with --json, letting the CLI auto-detect org and project.

  3. Phase 03Drill into details

    View issue details and events by short ID, use issue explain or issue plan for analysis, and sentry api for other endpoints.

  4. Phase 04Format the output

    List title, short ID, status, counts and more by recency, redacting emails and IPs, omitting raw stack traces, and stating when nothing is found.

Permissions (summarised from SKILL.md)

CapabilityNeededDetails
File writesNot requiredA read-only skill; it does not create or modify files.
Command executionLimitedRuns only sentry CLI query commands such as issue list, issue view, event view, schema and GET api calls.
Network outboundLimitedAccesses only the Sentry API through the sentry CLI.

SKILL.md frontmatter (YAML)

SKILL.md frontmatter (YAML)commit d4a9df5
---
name: "sentry"
description: "Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry CLI; perform read-only queries using the `sentry` command."
---

Related skills

A rule scan cannot catch every risk. Read this skill's SKILL.md and check its permissions before installing.