MCP server for orchestrating multiple Claude Code instances as a collaborative team
Critical gaps across all dimensions. No tool schemas visible in provided source code, cannot verify input parameter types, constraints, or output structures. Tool registration functions exist (src/tools/index.ts) but actual tool definitions are not shown. Descriptions for tools and parameters are not visible. Without access to the actual tool implementation files (submit-task.ts, list-processes.ts, etc.), schema and description scores default to 0. The server uses a custom logger and orchestrator pattern but provides no evidence of parameter validation, error recovery guidance, or LLM-optimized descriptions. STDIO-only transport caps protocol readiness at 50, further limiting overall score.
No input schemas visible for any tool. Cannot verify parameter types, constraints, enums, or required fields.
No tool descriptions visible in source. Cannot verify LLM-optimized descriptions (target 50-200 chars), WHEN to use each tool, or prerequisites.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 27 | <=2025-11-25 | v2 |
Tool definitions inferred from registration functions only. Actual tool implementation files (src/tools/*.ts) not provided. Cannot verify parameter descriptions, output schemas, error handling, or idempotency.
STDIO-only transport. Not remotely accessible; cannot be used by hosted MCP clients. Hard cap: protocolReadiness ≤ 50.
No evidence of error handling patterns. No recovery guidance, error classification (retryable/user-fixable/fatal), or actionable error messages visible in provided code.