deep-codebase-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and analysis of external, untrusted codebases, which creates a potential surface for indirect prompt injection.
- Ingestion points: The agent is instructed to perform "Scanning and Indexing" and "Static Analysis" of the codebase, including reading source code, documentation, and commit history, as outlined in the
SKILL.mdworkflow. - Boundary markers: The instructions lack specific boundary markers or "ignore embedded instructions" warnings for the agent when it deep-dives into third-party source code.
- Capability inventory: The skill provides a script (
scripts/analyze_structure.py) to map the directory structure and utilizes the agent's native file-reading capabilities to analyze project files. - Sanitization: There is no evidence of sanitization or filtering of the content being analyzed to prevent malicious instructions hidden in code comments or documentation from influencing the agent's behavior.
Audit Metadata