doc-management

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill interacts with official Feishu/Lark Open API endpoints (open.feishu.cn) to manage document data. It retrieves authentication tokens from environment variables (LARK_TENANT_TOKEN), which is a secure and standard practice for API integrations. References to these well-known services are considered safe and functional.
  • [PROMPT_INJECTION]: The skill is subject to a low risk of indirect prompt injection because it reads raw document content from external sources. * Ingestion points: The read_doc_content function in run.py. * Boundary markers: No delimiters are explicitly defined in the output content. * Capability inventory: The skill can create documents but does not execute local commands. * Sanitization: No content filtering is performed on the document data before it is returned to the agent.
  • [REMOTE_CODE_EXECUTION]: The testing utility tests/test_doc_management.py dynamically loads the run.py script using importlib. This is an expected pattern for local unit testing and does not constitute a remote code execution vulnerability as it targets a local, known file path.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 12:35 AM