This is a configuration wrapper around @apify/actors-mcp-server, not a standalone MCP server implementation. The .mcp.json file shows tool invocation via npx but provides NO visible tool definitions, schemas, parameter descriptions, or error handling. All 11 tools are declared by name only in the --tools argument. No input schemas, output schemas, parameter types, or descriptions are visible in the provided source. The governance hooks are shell scripts that post-process Claude tool calls, they do not define MCP tools. Without access to the actual @apify/actors-mcp-server source code, tool quality cannot be assessed. The average of 11 tools capped at 50 is 50, but given the complete absence of visible definitions, descriptions, and schemas in THIS repository, the score reflects what IS present: configuration only.

• Obtain and review the actual @apify/actors-mcp-server source code to extract tool definitions, input schemas, and output schemas. This evaluation is limited to configuration only.
• Expand each tool description from one-liner to 50-200 characters following the pattern: 'WHAT does it do? WHEN to use it? Any prerequisites?' Example: 'Execute Apify actors to scrape web data. Use when you need to extract structured content from websites. Requires APIFY_TOKEN environment variable.'
• Document every parameter with type, description, constraints, and examples. Use enums for known values (e.g. status: [pending, running, completed]). Specify ranges for numeric params (e.g. limit: 1-100).
• Add error handling that categorizes failures as retryable, user-fixable, or fatal. Return the invalid value and constraint violated. Example: 'Invalid actor_id: got "xyz", must be alphanumeric, 1-50 chars.'
• Document output schemas for each tool showing field names, types, and what data they contain. Include pagination info (total_count, next_cursor) for list tools. Ensure downstream tools can chain on returned IDs.