The server implements 5 tools with generally clear naming (verb-noun pattern) and reasonable descriptions. All tools have explicit input schemas with type definitions and parameter descriptions. However, several gaps reduce the score: (1) descriptions are on the shorter side, averaging ~60 chars, below the 194-char baseline for A+ tools; (2) output schemas are not documented, we can infer structure from code but LLMs cannot plan downstream calls without explicit schema documentation; (3) no error handling guidance in tool descriptions; (4) no input validation rules stated (e.g., 'per_page must be 1-100' is in the schema but not surfaced to LLMs in descriptions); (5) prompts are implemented but tool descriptions lack context hints for multi-step workflows. The tool definitions are functional and moderately production-ready but lack the refinement needed for higher scores.
Get full details of a bug report
Get project info (connection verification)
Get the screenshot URL for a bug report
List bug reports with optional filters
Update a bug report's status and optionally its title, description, or severity
Output schemas not documented in tool descriptions. LLMs cannot infer return structure and must guess what fields are available for downstream tool calls. For example, list_bug_reports returns PaginatedResponse<BugReportSummary> but the description does not state this, and the summary structure (which fields?) is invisible to the LLM.
Tool descriptions are brief (50 - 75 chars) and lack context hints. For example, 'Get project info (connection verification)' does not explain WHEN to call this tool or what the LLM should do with the response. Descriptions should be 50 - 200 chars and follow 'WHAT, WHEN, WHAT IT RETURNS' structure.
No error recovery guidance in descriptions. If update_bug_report_status fails because the bug_report_id does not exist, the description does not tell the LLM to call list_bug_reports first to find valid IDs. Actionable error messages and dependency hints are missing.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 57 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 43 | - | v1 |
Input validation rules exist in schemas (e.g., per_page: 1 - 100) but are not repeated in parameter descriptions. LLMs often miss JSON Schema constraints and rely on descriptions. Parameter descriptions should explicitly state constraints: 'Items per page (1 - 100, default 20)'.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint) present. update_bug_report_status is destructive but has no annotation to warn LLMs to confirm before executing. This is a missed opportunity to signal risk.