adapter-sentry

Files (1)

• SKILL.md

SKILL.md (first lines)

# Adapter: Sentry

## Focus

- Single, correct initialization per runtime (client/server/edge).
- Environment-aware event sending (no dev noise in production dashboards).
- Signal-to-noise: filter known-expected errors before they reach Sentry.

## Starter Checklist

- [ ] Sentry initializes from one dedicated config file per runtime, not ad hoc feature-module
      calls.
- [ ] Event sending is disabled or filtered in local development.
- [ ] Backend logger routes error/fatal events to Sentry centrally, not via per-service capture
      calls.

## Implementation Pattern

Evidence: 2-repo field extraction — see `.ai/MERGE_REPORT.md`.

- Initialize Sentry via dedicated environment-specific entrypoints (client/server/edge config
  files), never from a random feature module.
- Disable or drop event sending in local development by default; only send from staging/
  production.
- Filter known-expected noisy errors (e.g. expected auth/token refresh failures) before they
  reach Sentry, so alerts stay high-signal.
- Route error/fatal-level logs through the shared logger's Sentry capture path, not duplicated
  `Sentry.captureException` calls scattered across services.

If the current project already has an established pattern for any of these, follow the existing
pattern instead of this default.

From JahacksV1/day-one-agent-kit, MIT.

More from JahacksV1/day-one-agent-kit