chatkit
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFE
Full Analysis
- [CREDENTIALS_UNSAFE] (SAFE): The documentation correctly uses placeholders like 'sk-xxx' and 'YOUR_GEMINI_API_KEY' for API keys, demonstrating safe documentation practices without exposing real secrets.
- [EXTERNAL_DOWNLOADS] (SAFE): Instructions for installing the 'openai-agents' and 'openai-chatkit' frameworks via 'uv' are legitimate and necessary for the intended use of the library.
- [COMMAND_EXECUTION] (SAFE): Command-line examples for server management and environment verification use standard, non-malicious utilities such as 'curl', 'env', and 'psql'.
- [DATA_EXFILTRATION] (SAFE): No patterns of unauthorized data collection or external transmission were identified; all network examples focus on local development endpoints.
- [PROMPT_INJECTION] (SAFE): The provided instructions are purely technical configuration guides and do not contain any patterns designed to subvert AI safety protocols or override agent instructions.
Audit Metadata