a mcp server to get sentry issue by id or url for LLM
Two tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun pattern (get_*), which is good. Descriptions are present but generic (avg ~80 chars, below 194-char baseline). Parameters lack detailed constraints and format guidance. No error handling guidance. Output is raw JSON strings with no structured schema documentation. Schemas use Zod but lack explicit type annotations in descriptions (e.g., 'mode' enum not explained in terms of token impact). No pagination support despite potentially large result sets.
get issue events by inputting sentry organization id or slug and sentry project name or slug
get issue event by inputting sentry issue event url or sentry issue event id
Output schema not documented. Tools return raw JSON strings with no structured schema definition. LLMs cannot plan downstream operations or extract fields reliably.
Parameter descriptions lack actionable constraints. 'mode' param described as 'control LLM token usage' but no guidance on when to use 'tiny' vs 'huge'. 'url_or_id' accepts both URL and ID but format/regex not specified.
No pagination support. get_project_events returns all events with no limit, offset, or cursor. Large result sets will exhaust context window and degrade LLM reasoning.
Error handling returns bare text strings ('Invalid Event ID', 'Failed to Get Event') with no recovery guidance. LLM cannot determine if error is retryable or what to do next.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 48 | 2026-07-28+ | v2 |
Credentials (SENTRY_ORG, SENTRY_PROJ, SENTRY_USER_TOKEN) injected via environment variables, which is correct. However, no documentation of required scopes or permissions for audit/compliance.