skills/acedatacloud/skills/zhihu/Gen Agent Trust Hub

zhihu

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill handles authentication using environment variables (ZHIHU_DEVELOPER_TOKEN and ZHIHU_COOKIES) injected by the connector, which is a secure method for managing secrets without hardcoding.
  • [SAFE]: A strict safety gate is implemented for all write operations (publishing articles, answering questions, or editing content). The scripts require a --confirm flag to be the final argument, ensuring that the AI agent must perform a dry-run and receive explicit user approval before any changes go live.
  • [SAFE]: The blog.py script contains domain-validation logic to ensure that sensitive session cookies are only transmitted to authorized Zhihu domains. This protects against credential leakage during operations like image re-hosting from third-party sources.
  • [SAFE]: The skill uses the Python standard library (urllib) for all network requests, eliminating risks associated with unverified third-party dependencies or dynamic remote code execution.
  • [SAFE]: The instructions explicitly direct the agent to keep credentials secret and forbid echoing or printing them, providing clear operational boundaries for safe execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 08:55 AM
Security Audit — agent-trust-hub — zhihu