What problem does it solve?
This Skill helps users rapidly identify why a protocol proxy is unhealthy, slow, failing to forward requests, or unable to use MCP tools by running structured diagnostics and summarizing findings into clear next actions.
Core Features & Use Cases
- Health checks: Collects runtime, proxy, provider, and API key health data, then highlights abnormal items and severity for quick triage.
- Log and request analysis: Pulls system logs and recent request failures to detect error patterns, frequency spikes, and affected proxies/providers/models.
- Targeted proxy diagnostics: Diagnoses a specific proxy’s connectivity, provider availability, key health, request chain behavior, routing strategy, and (optionally) end-to-end client connectivity.
- MCP troubleshooting: Checks MCP server connectivity, validates common misconfigurations (package name, stdio vs HTTP, port conflicts), and reports tool availability.
- Sub-agent/delegation debugging: Investigates failed delegated tasks by reviewing task details, exec-policy behavior, sandbox and permission issues, and configuration settings.
Quick Start
Ask the AI to run a system health check and, if problems are found, automatically drill down into the most likely failing proxy, provider key, or MCP service.