probability-inference
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run a local Python script
scripts/probability_inference_diagnostics.pyfor data processing. This is a functional component of the workflow and presents no autonomous execution risk.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a CSV file, which is a potential entry point for malicious instructions.\n - Ingestion points: The
probability_inference_diagnostics.pyscript readsinput.csv.\n - Boundary markers: No delimiters are specified to isolate data from instructions.\n
- Capability inventory: The skill can execute local scripts and generate report files.\n
- Sanitization: Input data is cast to numeric types, mitigating risks from text-based injection payloads.
Audit Metadata