skills/cklxx/elephant.ai/feishu-task/Gen Agent Trust Hub

feishu-task

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructions for managing Feishu tasks by executing a local script (skills/feishu-cli/run.py) through the bash tool. This is a standard integration method for extending agent capabilities within a repository.
  • [SAFE]: All authentication requirements are clearly documented as needing user OAuth tokens, avoiding the use of hardcoded credentials. Placeholders such as ou_xxx and t_xxx are utilized for IDs.
  • [SAFE]: Indirect Prompt Injection Surface Analysis: Ingestion points: The skill handles task summaries and descriptions which are external data from Feishu. Boundary markers: No explicit boundary markers are used in the documentation examples. Capability inventory: The skill is restricted to task management operations (CRUD) via a specific Python CLI. Sanitization: While not explicitly defined in the markdown, the functionality is restricted to the specific task management scope of the underlying tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 02:52 PM
Security Audit — agent-trust-hub — feishu-task