The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the coval CLI and a local Python script (send-test-span.py) to verify connectivity and trace ingestion. These are legitimate diagnostic actions performed using vendor-provided tools.
[CREDENTIALS_UNSAFE]: Security is enhanced by specific instructions that prohibit the agent from asking for raw API keys and mandate the redaction of sensitive provider information from metadata responses.
[DATA_EXFILTRATION]: No patterns of data exfiltration or unauthorized network access were found. All diagnostic network operations are directed toward official tracing infrastructure as part of the primary troubleshooting function.

debug-traces