dingtalk-oa

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the 'dws' binary using Python's 'subprocess.run' module in 'scripts/oa_batch_approve.py' and 'scripts/oa_pending_review.py'. The implementation securely passes arguments as a list, which effectively mitigates the risk of shell injection attacks.
  • [PROMPT_INJECTION]: The skill retrieves and displays data from external DingTalk OA approval forms in 'scripts/oa_pending_review.py'. This data ingestion point serves as a potential surface for indirect prompt injection. However, the skill manages this risk by requiring human-in-the-loop confirmation for sensitive actions such as 'approve' or 'reject', as outlined in the 'Dangerous Operations' section of the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 11:20 AM
Security Audit — agent-trust-hub — dingtalk-oa