Critical gaps across naming, descriptions, and schemas. Tool names lack consistent verb prefixes (e.g., 'query' vs 'execute_python'). Descriptions are minimal (10-50 chars) and lack context on WHEN to use each tool or what they return. Parameter schemas present but descriptions are sparse or missing. No output schemas documented. Error handling is absent, no guidance on recovery or retryability. Security concerns: execute_python and write_file are high-risk with minimal safeguards visible. No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Composition issues: execute_python and analyze_code overlap; search lacks pagination.

• Expand all descriptions to 100-200 chars. Include: WHAT the tool does, WHEN to call it (vs similar tools), WHAT it returns. Example: 'Execute a SQL SELECT query against the database. Use this to retrieve data; for schema discovery, call get_schema first. Returns rows as JSON array with column names.'
• Document output schemas for all tools. Specify field names, types, and whether results are paginated. Example for search: '{results: [{id, title, url, snippet}], total_count, next_cursor}'
• Add parameter descriptions for all inputs. For 'sql' param: 'SQL SELECT query (no INSERT/UPDATE/DELETE). Max 5000 chars. Timeout 30s.'
• Implement permission gates and dry-run for execute_python and write_file. Add a 'dry_run' boolean param to write_file; require explicit 'confirm=true' for execute_python.
• Add pagination to search tool: accept 'limit' (1-50, default 20) and 'offset' params; return 'total_count' and 'has_more' fields to enable multi-page iteration.

Inferred effective spec: 2026-07-28+.

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