openclaw-wechat-integration
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPERSISTENCEPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to install a third-party plugin @canghe/openclaw-wechat using the openclaw command-line tool.
- [PERSISTENCE]: The skill provides a systemd service configuration (openclaw-gateway.service) to ensure the WeChat gateway process starts automatically and restarts on failure.
- [PRIVILEGE_ESCALATION]: Instructions include the use of sudo for administrative tasks such as configuring the firewall (ufw) and managing system services via systemctl.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from WeChat messages. 1. Ingestion points: Incoming messages via webhooks configured at channels.wechat.webhookHost and channels.wechat.webhookPath. 2. Boundary markers: None identified in the provided configuration or sample handler code. 3. Capability inventory: The skill allows the agent to send text and image messages back to WeChat users and groups. 4. Sanitization: No explicit sanitization or validation of the incoming message content is described.
- [COMMAND_EXECUTION]: The skill utilizes several shell commands for configuration, plugin management, and troubleshooting, including openclaw config, curl, and telnet.
Audit Metadata