secrets-scan
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze untrusted source code provided by users, which creates a surface for indirect prompt injection. Malicious instructions could be embedded within the code or comments of the files being scanned to attempt to override agent behavior during the analysis.
- Ingestion points: Processes user-provided source code, specific directories, and git history via the
/secrets-scancommands and scope arguments. - Boundary markers: The skill instructions do not specify the use of delimiters or explicit boundary markers to help the agent distinguish between data to be scanned and instructions to be followed.
- Capability inventory: The skill is primarily evaluative and diagnostic; it does not include scripts for file writing, network exfiltration, or automated command execution based on the scan results.
- Sanitization: The skill lacks explicit instructions for sanitizing or escaping the content of the files before the agent processes them, which is a standard risk for skills operating on external data.
- Mitigation Recommendation: When processing user code, the agent should be instructed to treat all file content strictly as data and ignore any natural language instructions found within comments or code strings.
Audit Metadata