sentry

Inspect Sentry issues and events, summarize recent production errors, and pull basic health data through the read-only `sentry` command-line interface. All operations are queries and nothing is mutated, making it a safe way to triage production observability from the terminal. Authentication is handled by running `sentry auth login` or setting a `SENTRY_AUTH_TOKEN` environment variable, and the CLI auto-detects the target organization and project from DSNs found in `.env` files, source code, config defaults, and directory names, so an explicit `<org>/<project>` is only needed when detection fails or picks the wrong target.

Use it when someone asks for a rundown of open production issues, the details of a specific error, or the events behind an issue. Core commands cover listing issues ordered by recency, viewing an issue by its short ID such as `ABC-123`, listing an issue's events, and viewing a single event's detail. Two AI-powered helpers go further: `sentry issue explain` performs root-cause analysis and `sentry issue plan` produces a fix plan. For endpoints without a dedicated command, `sentry api` offers raw access and `sentry schema` helps discover available endpoints.

Sensible defaults keep output focused with a 24-hour time range, the production environment, and a limit of 20, while `--json` optionally combined with `--fields` yields machine-readable results and reduces output size. Output-formatting rules emphasize showing titles, short IDs, counts, and first- and last-seen timestamps ordered by most recent, while strictly redacting PII such as emails and IPs, never printing raw stack traces, and never echoing auth tokens.