cognicode-quality-investigator
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions were audited for malicious patterns, including prompt injection, data exfiltration, and unauthorized command execution; no suspicious activities or security risks were identified. The skill incorporates best practices for security auditing, such as manual verification and limiting tool use to authorized environments.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted source code from repositories or pull requests (ingestion points). It includes explicit boundary markers and safety precautions that direct the agent not to execute unknown scripts (boundary markers). The skill's capability inventory includes file reading and using the CogniCode MCP/CLI analysis tools. Sanitization is achieved by requiring the agent to verify all findings against multiple sources and maintain a clear distinction between static analysis and runtime behavior.
- [EXTERNAL_DOWNLOADS]: The documentation includes a command for installing the skill from a repository associated with the author (Rubentxu). This installation step is a standard deployment procedure for agent skills and is not part of the skill's runtime execution logic.
Audit Metadata