Loading ToolBench
MCP server for profiling and analyzing distributed traces from LLM applications using OpenTelemetry, identifying bottlenecks and exporting trace data
This server has critical definition failures. The single tool 'execute_skill_action' has no visible input schema, no parameter descriptions, and a generic 1-line description (18 chars, below the 20-char minimum). The mcp_server.py file does not implement the MCP protocol, it merely prints a JSON status message and exits. There is no tool registration, no schema validation, no parameter handling, and no actual MCP server loop. The tool definition is inferred from example_usage.py, not explicitly registered. This is not a functional MCP server.
Execute distributed trace profiling and analysis skill action
No input schema visible for execute_skill_action. Tool accepts parameters (trace_id, spans_list) but no JSON Schema definition exists in the code.
Tool description is 18 characters ('Execute distributed trace profiling and analysis skill action'), below 20-char minimum. Description lacks WHAT, WHEN, and prerequisites.
No parameter descriptions. 'trace_id' and 'spans_list' parameters have no documentation explaining format, constraints, or purpose. LLMs cannot infer parameter meaning from names alone.
Tool definition is inferred from example_usage.py, not explicitly registered in mcp_server.py. This server does not implement MCP tool registration.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 15 | 2026-07-28+ | v2 |
No output schema documented. The client.profile_trace() method returns a dict with fields (total_duration_ms, bottleneck_span, otel_export_status), but no schema is declared for the tool response. LLMs cannot plan downstream calls without knowing output structure.