agents
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates reading and searching repository documentation files, which represents an ingestion surface for potentially untrusted data.
- Ingestion points: Processes files in the .agents/ directory and AGENTS.md (SKILL.md).
- Boundary markers: No explicit markers are defined to isolate untrusted data from the agent's instructions.
- Capability inventory: The skill tools (repo_root, list_skills, search_docs, skill_for, fuzzy_search) are limited to information retrieval and do not include high-risk capabilities like network access or file system modification.
- Sanitization: No sanitization logic is described for the retrieved documentation content.
- [COMMAND_EXECUTION]: The instructions include manual steps for the user to build a Go binary using 'go test' and 'go build'. These are standard development workflows and do not involve automated execution of external code.
Audit Metadata