bento-cso
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is primarily instructional and provides a methodology for security reviews. It does not execute arbitrary code, perform network operations, or access sensitive system credentials.\n- [INDIRECT_PROMPT_INJECTION]: As a security auditing tool, the skill naturally processes untrusted data (the code being reviewed). It instructs the agent to follow instructions from repository-level files such as AGENTS.md or .agents/bento/bento-cso.md, which could potentially be used by an attacker to influence the audit's outcome.\n
- Ingestion points: Processes code changes and repository-specific instruction files (AGENTS.md, CLAUDE.md, .agents/bento/bento-cso.md).\n
- Boundary markers: None specified for separating audit instructions from the code being analyzed.\n
- Capability inventory: The skill instructions state that the audit itself edits nothing and only returns findings; no file-write, network, or subprocess tool calls are defined.\n
- Sanitization: No specific sanitization process for the code under review is mentioned.
Audit Metadata