integration-checker
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted source code and project documentation from the local repository to perform static analysis. This creates an attack surface where instructions embedded in analyzed files (like CLAUDE.md or source comments) could potentially manipulate the agent's logic. \n
- Ingestion points: Phase 0 (PRIME) reads the project's CLAUDE.md file. Phase 1 and Phase 2 scan and read source files across the repository using Read, Grep, and Glob tools. \n
- Boundary markers: The instructions lack explicit delimiters or instructions to treat ingested code and metadata purely as data. \n
- Capability inventory: The skill utilizes Bash, Read, Grep, and Glob, providing it with extensive access to the repository's file system. \n
- Sanitization: The skill does not implement any sanitization or filtering of the ingested content before it is processed by the agent's reasoning engine.
Audit Metadata