Two tools with basic definitions but significant gaps. List-Workflows-Tool has empty input schema ({}), violating schema requirements. Run-Workflow-By-Id-Tool has a proper schema but descriptions are minimal (under 100 chars). Neither tool has parameter descriptions beyond the schema. No output schemas documented. No error handling guidance. No tool annotations (readOnlyHint/destructiveHint). Descriptions lack context on when to use each tool or what to do on failure.

• Add explicit output schema documentation to both tools. For List-Workflows-Tool, document that it returns [{"id": int, "name": str, ...}]. For Run-Workflow-By-Id-Tool, document the structure of the result dict (e.g., {"status": str, "output": str, "execution_id": str}).
• Add parameter descriptions in tool docstrings. For Run-Workflow-By-Id-Tool, clarify: 'input (str, optional): The text input to pass to the workflow. If omitted, the workflow's default input_text is used.'
• Add tool annotations via FastMCP. Use @mcp.tool(..., description=...) with readOnlyHint=True for List-Workflows-Tool and destructiveHint=True for Run-Workflow-By-Id-Tool to signal operation safety.
• Improve error messages. Instead of returning empty results, return structured errors: {"error": "Failed to fetch workflows", "reason": "API returned invalid response", "next_step": "Check FLYDER_API_KEY and FLYDER_EMAIL environment variables."}
• Expand tool descriptions to 100-200 chars. Example: 'List-Workflows-Tool: Retrieves all workflows available to the authenticated user. Returns workflow names and IDs needed to run a specific workflow. Call this first to discover available workflows.'

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

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