crewai
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION] (SAFE): The skill's instructions are focused on domain-specific expertise and do not contain patterns designed to bypass safety filters or override system instructions.
- [DATA_EXFILTRATION] (SAFE): There are no indicators of unauthorized file access, credential harvesting, or network requests to non-whitelisted domains.
- [REMOTE_CODE_EXECUTION] (SAFE): While the skill contains Python code snippets for the CrewAI framework, it does not involve downloading scripts from the internet or executing untrusted code.
- [COMMAND_EXECUTION] (SAFE): No dangerous shell commands, privilege escalation attempts, or persistence mechanisms were detected in the provided files.
- [INDIRECT_PROMPT_INJECTION] (SAFE): Although the skill processes user-defined configurations, it lacks the execution capabilities (like file writing or network calls) necessary to weaponize indirect injections.
Audit Metadata