policy-as-code-linter
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external YAML files, which serves as a surface for indirect prompt injection. \n
- Ingestion points: The
policy_as_code_linter.pyscript reads user-provided YAML files via the--inputargument (line 160). \n - Boundary markers: There are no explicit markers or instructions provided to the agent to distinguish between the policy data and its own system instructions. \n
- Capability inventory: The skill can read local files and write JSON reports to local paths, but it does not have capabilities for network operations or arbitrary command execution. \n
- Sanitization: The script performs schema and type validation but does not sanitize the string content of the YAML fields (like descriptions) against adversarial natural language instructions.
Audit Metadata