coze-api
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE_AND_EXFILTRATION]: Facilitates network communication with api.coze.cn. This is the primary function of the skill and targets a well-known service domain.\n- [INDIRECT_PROMPT_INJECTION]: The skill implements a client that ingests user-defined text to be sent to an LLM provider. \n
- Ingestion points: User input provided to chat, stream_chat, and chat_with_polling functions in scripts/coze_client.py and SKILL.md.\n
- Boundary markers: Not present; the client passes raw strings to the API.\n
- Capability inventory: Network access to api.coze.cn via requests library in scripts/coze_client.py.\n
- Sanitization: Not present; typical for a client library where the LLM provider handles content filtering.
Audit Metadata