getsentry__sentry-for-ai/AGENTS.md

Agent Instructions

Project Overview

Claude Code plugin providing Sentry integration: MCP server, slash commands, skills, and sub-agents.

Commit Attribution

AI commits MUST include:

Co-Authored-By: (the agent model's name and attribution byline)

Plugin Structure

.claude-plugin/     # Plugin metadata
commands/           # Slash commands (/seer, /getIssues)
skills/             # Setup and review skills
agents/             # Sub-agents (issue-summarizer)

Skills

Sentry Code Review

Analyze and resolve Sentry bot comments on GitHub PRs. See skills/sentry-code-review/SKILL.md

Sentry Setup Skills

• AI Monitoring - Instrument OpenAI/Anthropic/Vercel AI/LangChain/Google GenAI. See skills/sentry-setup-ai-monitoring/SKILL.md
• iOS (Swift) - Setup Sentry in iOS/Swift apps with error monitoring, tracing, session replay, logging, and profiling. See skills/sentry-ios-swift-setup/SKILL.md
• Logging - Configure structured logging (JS/Python/Ruby). See skills/sentry-setup-logging/SKILL.md
• Metrics - Setup counters/gauges/distributions. See skills/sentry-setup-metrics/SKILL.md
• Tracing - Enable performance monitoring and distributed tracing. See skills/sentry-setup-tracing/SKILL.md

Commands

Command	Description
/seer <query>	Natural language Sentry environment queries
/getIssues [project]	Fetch 10 most recent Sentry issues

MCP Server

Sentry MCP server configured at https://mcp.sentry.dev/mcp. See MCP-SETUP.md for advanced configuration.

Key Conventions

• All setup skills must detect platform/SDK before suggesting configuration — never assume
• Sentry code review skill only processes comments from sentry[bot], ignores other bots
• Issue summarizer agent runs on sonnet model and processes issues in parallel
• GitHub CLI (gh) required for PR-related skills