codebase-analysis
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill is configured to execute a local Python module (
skills.codebase_analysis.analyze) to manage the analysis workflow. This is intended functionality for the skill but involves executing local script code.\n- [PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection (Category 8) due to its role in processing untrusted codebase data. Evidence Chain: 1. Ingestion points: Ingests entire repositories via an 'Explore agent' and script. 2. Boundary markers: No explicit boundary markers or 'ignore' instructions for the agent are defined in the provided files. 3. Capability inventory: The skill uses automated exploration and script execution. 4. Sanitization: No sanitization of code excerpts or file content is documented before the data is processed by the agent.
Audit Metadata