sast-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to analyze arbitrary files within a codebase, creating a surface for indirect prompt injection where malicious code could influence agent behavior.\n
- Ingestion points: Reads project manifests (package.json, requirements.txt, etc.) and source code files to map architecture.\n
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potential commands embedded in the analyzed codebase.\n
- Capability inventory: The agent is granted file-read access for analysis and file-write access to generate reports in the
sast/directory.\n - Sanitization: The skill lacks instructions for sanitizing or validating codebase content before incorporating it into the
sast/architecture.mdfile.\n- [CREDENTIALS_UNSAFE]: The skill explicitly directs the agent to identify and inventory sensitive information such as connection strings, API key patterns, tokens, and credentials. While intended for an internal report, this process involves the agent accessing and handling raw secrets during the reconnaissance phase.
Audit Metadata