Tempo MCP has well-structured schemas with proper types and descriptions for all 6 tools. Tool names follow verb_noun convention (get_, create_, update_, delete_). However, descriptions are minimal (10-50 chars), falling below the 50-200 char LLM-optimized baseline. Parameter descriptions exist but lack actionable constraints (e.g., time_spent format examples, date validation rules). No output schemas documented. Error handling and recovery guidance absent. Missing tool annotations (readOnlyHint, destructiveHint) despite clear risk levels. Composition is sound, tools are single-responsibility and chainable.

• Expand tool descriptions to 50-200 chars. Example: 'Retrieve worklogs for a date range, project, or user. Use this to audit time tracking, verify billable hours, or find entries before updating. Returns paginated results (default 50 per page).'
• Document output schemas for all tools. Specify which fields are always present, which are optional, and their types. E.g., create_worklog returns {tempoWorklogId: int, startDate: string, timeSpentSeconds: int, ...}.
• Add format constraints to parameters. E.g., time_spent: 'Duration in format HH:MM, Xh Ym, or decimal hours (e.g., 2h 30m, 1.5h, 90m). Must be > 0 and ≤ 24h per day.' start_date: 'ISO 8601 date (YYYY-MM-DD). Must be ≤ today.'
• Implement tool annotations in server.py. Mark delete_worklog with destructiveHint=true, create_worklog and update_worklog with idempotentHint=false (or true if Tempo supports idempotency keys), and all read tools with readOnlyHint=true.
• Add error handling with recovery guidance. E.g., if issue_id is invalid, return 'Issue not found. Call get_accounts() to verify the account, or search for the issue in Jira first.' If time_spent format is invalid, return 'Invalid time format. Use HH:MM (e.g., 02:30), Xh Ym (e.g., 2h 30m), or decimal (e.g., 2.5h).'

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry