The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the agenttrace binary from the system PATH or via go run if the user is within the source repository. This is used to analyze session logs, generate reports, and enforce CI health gates.
[INDIRECT_PROMPT_INJECTION]: The skill processes external session logs and export files which could contain untrusted data. Ingestion points: Reads files from paths provided by the user or discovered via --doctor in SKILL.md. Boundary markers: Not explicitly defined in the prompt interpolation. Capability inventory: Executes shell commands (agenttrace) to process these files. Sanitization: Relies on the agenttrace tool for processing; no additional sanitization is defined in the skill instructions.

agenttrace-session-audit