paper-assembly
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- COMMAND_EXECUTION (SAFE): The skill executes a local Python script
assembly_checker.pyto verify artifact completeness and pipeline state. This is a standard and documented operational procedure. \n- PROMPT_INJECTION (LOW): The skill is susceptible to indirect prompt injection (Category 8). \n - Ingestion points: The skill reads from the project directory (
paper/) and a state file (checkpoint.json). \n - Boundary markers: No explicit delimiters or safety instructions are provided to separate project data from agent orchestration commands. \n
- Capability inventory: The skill coordinates several high-privilege tasks, including code execution, experiment design, and paper compilation. \n
- Sanitization: There is no evidence of sanitization or validation for the content of the artifacts or the metadata in the checkpoint files.
Audit Metadata