pythia-review
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions.
- Ingestion points: PL/SQL source code retrieved via
pythia src, linter results frompythia lint, and project-specific configuration files likeconventions.json. - Boundary markers: The skill instructions do not specify any delimiters or safety warnings to separate external code content from agent instructions.
- Capability inventory: The agent uses the
pythiaCLI tool to fetch source code, check errors, and analyze impact, which involves executing shell-based commands. - Sanitization: There is no mention of sanitizing or filtering the source code content before it is read and analyzed by the agent.
Audit Metadata