feishu-messaging
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill provides code templates for sending messages to Feishu that directly interpolate content into JSON strings without demonstrating sanitization or boundary markers. This creates an indirect injection surface if an agent processes untrusted external data and sends it via these scripts.
- Ingestion points: Message content fields in Python scripts within SKILL.md.
- Boundary markers: None present in the code examples.
- Capability inventory: Network access via the lark-oapi library to Feishu API endpoints.
- Sanitization: Absent; the templates use raw string interpolation for the 'content' field.
- [SAFE]: The skill utilizes the official lark-oapi library to interact with Feishu, which is a well-known enterprise communication service.
- [SAFE]: Sensitive API credentials (App ID and App Secret) are handled correctly using standard placeholders rather than hardcoded secrets.
Audit Metadata