The Agent Skills Directory

Command Execution: The skill facilitates the execution of diagnostic and maintenance commands using uv run and docker compose. These operations are used to verify model connections, build sandboxes, and retrieve system logs, which is consistent with its role as a troubleshooting tool.
Log and Trace Ingestion: The skill instructs the agent to read and interpret system logs and LangSmith traces to diagnose failures.
Ingestion points: System logs are accessed via uv run mda logs and docker compose, while traces are accessed through LangSmith integration (list_traces, read_trace).
Boundary markers: No specific boundary markers are defined for the ingested log or trace data.
Capability inventory: The skill possesses the capability to execute shell commands (via uv and docker) and read files.
Sanitization: The instructions explicitly advise against printing credentials or keys, noting that logs typically contain exception classes rather than sensitive provider messages.
Credential Handling: The skill reinforces security best practices by explicitly instructing the agent never to print API keys, tokens, or environment variables in its output, even when reporting missing settings.

design-agent-troubleshoot