sast-pathtraversal
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted external codebase files, which constitutes a surface where malicious instructions embedded in the analyzed code could potentially influence the agent's analysis or output.
- Ingestion points: Processes files from an external codebase (Phase 1 and 2) and reads
sast/architecture.md. - Boundary markers: The instructions for subagents lack explicit boundary markers or "ignore embedded instructions" warnings for the interpolated code snippets.
- Capability inventory: The skill performs recursive file reads across the codebase and writes reports and intermediate files to the
sast/directory. - Sanitization: There is no evidence of sanitization, escaping, or schema validation for the source code content before it is processed by the agent or its subagents.
Audit Metadata