repo-recon
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze untrusted content from a repository, creating a vulnerability surface.\n- Ingestion points: The agent is directed to read README, ARCHITECTURE, and manifest files (SKILL.md Step 1).\n- Boundary markers: The instructions lack markers or warnings to disregard instructions found within the repository files.\n- Capability inventory: The agent is encouraged to execute build, test, and run commands discovered in the repo (SKILL.md Step 2).\n- Sanitization: No validation of discovered commands is required before execution.\n- [COMMAND_EXECUTION]: The skill directs the agent to verify repository commands by running them (e.g., npm test, Makefile targets). This could lead to execution of malicious scripts defined in an untrusted codebase.
Audit Metadata