Critical gaps across all dimensions. The source code shows only a contract validation script (check-live-contract.mjs) that tests tool existence and annotations, but does NOT reveal actual tool definitions, schemas, or parameter documentation. Tool names are inferred from the expectedTools array. No input schemas, parameter descriptions, or output documentation are visible in the provided source. The server appears to be a remote HTTP service (https://mcp.memoket.ai/mcp), but the actual tool implementation is not included in the repository snapshot. Per HARD SCORING RULES: tools without visible definitions are capped at 50; tools without visible schemas score 0 on schema; tools without visible descriptions score 0 on description.

• Provide complete tool definitions in the repository, including input schemas with parameter types, descriptions, and constraints. Use JSON Schema format with explicit type declarations for all parameters.
• Add 50-200 character descriptions for each tool explaining WHAT it does, WHEN to use it, and what it returns. Current contract validation only checks annotation presence, not description quality.
• Document output schemas for all tools. Specify returned fields, types, pagination support (limit/offset/cursor), and any IDs needed for downstream tool chaining.
• Add parameter descriptions for all inputs. For example, if get_transcripts accepts a conversation_id, document its format, whether it's required, and what happens if not found.
• Implement error handling patterns: categorize errors as retryable vs fatal, include the invalid value in error messages, and provide recovery hints (e.g., 'Conversation not found. Try list_conversations() first.').

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

• Relies on Stateful initialize / Mcp-Session-Id (removed; protocol is stateless) - make each request self-contained