context-map
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data via the task description template variable. Ingestion point: SKILL.md. Boundary markers: Absent. Capability inventory: Codebase searching and writing to the .aix/ directory. Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill instructs the agent to use standard shell commands for workspace management. Evidence: Use of mkdir -p .aix in SKILL.md to ensure the artifact directory exists.
Audit Metadata