skills/qwibitai/nanoclaw/add-wechat/Gen Agent Trust Hub

add-wechat

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the wechat-ilink-client package via the standard NPM registry to enable API communication.
  • [EXTERNAL_DOWNLOADS]: It retrieves the necessary adapter source code (wechat.ts) from the repository's channels branch during installation.
  • [COMMAND_EXECUTION]: System-level commands including systemctl (Linux) or launchctl (macOS) are used to restart the agent service during installation and removal.
  • [COMMAND_EXECUTION]: The skill uses sed to programmatically modify the .env configuration file to enable or disable the WeChat channel.
  • [COMMAND_EXECUTION]: A utility script (wire-dm.ts) executes SQL operations via better-sqlite3 to configure messaging group routing in the local database.
  • [PROMPT_INJECTION]: The skill establishes a data ingestion point for messages from external WeChat users, creating an indirect prompt injection surface.
  • Ingestion points: Incoming messages received through the iLink Bot API and processed by the wechat.ts adapter.
  • Boundary markers: None defined within the skill files.
  • Capability inventory: The agent can read message content and send responses to messaging groups.
  • Sanitization: Not explicitly performed within the skill; relies on the platform's core message handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 09:50 PM