This is a migration checker tool, not a production MCP server. The 8 tools are test fixtures across multiple languages (TypeScript, Python, Go, Rust) demonstrating migration readiness patterns. Most tool definitions lack proper schemas and descriptions. search_documents, list_notes, read_note have minimal descriptions (10-50 chars, below the 50-200 char baseline). Schemas are inferred from fixture code rather than explicitly registered. No error handling guidance, no output schema documentation, no parameter validation hints. The tools are intentionally varied to test the checker's detection logic, not to exemplify quality.

• Expand tool descriptions to 50-200 chars. Example: 'search_documents' → 'Full-text search across the tenant's document set. Returns up to 20 matching documents with titles and IDs. Use this to discover documents before calling read_note.'
• Document input schemas explicitly in tool registration. Specify types (string, number), constraints (enum, min/max), and required vs optional for all parameters.
• Document output schemas. Example: 'search_documents returns {results: [{id, title, snippet}], total_count, next_cursor}'.
• Resolve duplicate 'list_notes' by renaming one variant (e.g., 'list_notes_in_folder' vs 'list_notes_flat') to make the distinction clear to LLMs.
• Add error handling guidance to descriptions. Example: 'read_note: If path not found, call search_documents() to discover valid paths.'

Inferred effective spec: 2026-07-28+.

• Relies on Roots (deprecated) - pass files/dirs via tool parameters, resource URIs, or configuration