debug-sentry-monitor

# Sentry Monitor Automated Sentry issue triage, root cause analysis, fix, architecture audit, and monitoring enhancement workflow. Works with **any project** — auto-detects configuration from the codebase. Uses the `sentry` MCP server for all Sentry API operations. ## Critical Rules > **NEVER resolve an issue without a verified fix.** > Resolving means "this will not happen again." If you cannot prove that, leave it unresolved. > **NEVER apply a band-aid fix.** > Wrapping code in try/catch, adding `?.` chains, or guarding with `Array.isArray()` are symptom suppressors. > Only use defensive coding *after* fixing the root cause, to harden against truly unpredictable external input. > **Understand the WHY before touching any code.** > **Research before fixing non-trivial bugs.** > Use `firecrawl_search` + `firecrawl_scrape` to find best practices for the specific error pattern before implementing a fix. --- ## Step 0: Auto-Detect Project Configuration Before making any Sentry MCP calls, discover the project's Sentry setup. ### 0a. Find Organization and Project First, try to detect from local config files. Search for these (in order): 1. `.sentryclirc` — contains `[defaults]` with `org` and `project` 2. `sentry.properties` — contains `defaults.org` and `defaults.project` 3. `.env`, `.env.local`, `.env.production` — look for `SENTRY_ORG`, `SENTRY_PROJECT`, `SENTRY_DSN`, `SENTRY_AUTH_TOKEN` 4. `sentry.client.config.ts`, `sentry.client.config.js` — Next.js Sentry config ...