setup-notifications-via-wecom
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user-provided data and transmit it as a message to an external WeCom webhook. * Ingestion points: User prompts and text files provided to the agent for notification delivery. * Boundary markers: Instructions require a human confirmation gate for recipients classified as 'others'. * Capability inventory: The
scripts/send_wecom.pyscript utilizesurllib.requestto perform network POST operations. * Sanitization: Message content is JSON-encoded before transmission. - [DATA_EXFILTRATION]: The skill enables network communication from the local host to a user-defined WeCom webhook URL. While the webhook is intended for legitimate notifications, this network capability can be misused to transmit sensitive data if the agent is compromised or misled. The skill mitigates this by requiring explicit target classification and integrity verification of the sender script.
Audit Metadata