codebase-analysis
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill is designed to ingest and analyze untrusted data from external codebases.
- Ingestion points: The skill uses Grep, Glob, and Read tools to process codebase files.
- Boundary markers: There are no explicit delimiters or instructions to ignore embedded prompts within analyzed code.
- Capability inventory: The skill has the ability to Write and Edit files and use the Task tool to launch specialist agents.
- Sanitization: No sanitization or validation of analyzed content is mentioned before it is processed or documented.
- Mitigation: The mandatory 'Review Phase' requires the agent to present findings and wait for user confirmation before proceeding, which serves as a primary defense against automated exploitation of injected instructions.
Audit Metadata