parallel-exploring
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze arbitrary files within a codebase, creating a surface for indirect prompt injection where malicious instructions hidden in code or comments could influence agent behavior.
- Ingestion points: The skill reads files from directories like
src/,lib/, and configuration files such as.env. - Boundary markers: None are defined to distinguish between untrusted file content and the subagent's core instructions.
- Capability inventory: Subagents are read-only but can be launched concurrently to perform broad filesystem exploration.
- Sanitization: No sanitization or filtering of codebase content is mentioned before the synthesis step.
Audit Metadata