analytics-toolkit
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a local Python script
scripts/check-output.pyintended for quality assurance. The agent is instructed to execute this script to verify that its generated output contains the required sections. - [INDIRECT_PROMPT_INJECTION]: The
scripts/check-output.pyscript serves as an ingestion point for data, as it reads the content of an agent-generated deliverable to perform validation. However, this poses no risk as the script only performs static string matching for headers and does not execute the file content or transmit data externally. - Ingestion points:
scripts/check-output.pyreads a file provided via command line argument. - Boundary markers: None (not applicable for simple string matching).
- Capability inventory: File read via
pathlib.Path.read_text. No network or write capabilities. - Sanitization: None required as content is not executed or rendered.
Audit Metadata