gsd-map-codebase
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes existing codebase content which is considered untrusted data. This creates a surface for indirect prompt injection where malicious instructions hidden in code comments or files could influence the agent's behavior.
- Ingestion points: The skill ingests codebase data using the
Read,Glob, andGreptools to provide context to mapper agents. - Boundary markers: The instructions lack explicit boundary markers or directives for the sub-agents to treat codebase content strictly as data rather than instructions.
- Capability inventory: The skill and its sub-agents have access to
Bash,Write, andAgenttools, which provides a significant capability surface if an injection were successful. - Sanitization: No sanitization or filtering logic is defined for the codebase content before it is processed by the LLM.
Audit Metadata