Wechat Forward

按需将当前对话内容转发到用户自己的微信。仅在用户明确要求"把这个发微信"时才执行， 将最近的消息内容通过 wxclawbot CLI 推送到用户微信，用于手机端查看。 Trigger ONLY when explicit: 发微信, 推送到微信, 发到微信, forward to wechat, 发给我微信, 帮...

This skill appears coherent for forwarding chat content to your own WeChat using the local wxclawbot CLI. Before installing/use: (1) ensure wxclawbot is trustworthy and installed from a known source; (2) confirm the accounts directory (~/.openclaw/openclaw-weixin/accounts/) contains only your own credentials; (3) be cautious when asking the agent to "send file" — that will read local file paths you specify and forward them to your phone; (4) verify the agent actually respects the listed trigger phrases and does not auto-trigger; (5) review wxclawbot logs if you need auditability. If any of these items are unclear or you do not trust wxclawbot, do not enable the skill.

Type: OpenClaw Skill Name: wechat-forward Version: 1.0.1 The skill 'wechat-forward' facilitates forwarding conversation history and files to a user's WeChat account using the 'wxclawbot' CLI. While the instructions in SKILL.md emphasize explicit user consent and 'self-only' data flow, the implementation pattern of passing user-generated text directly into shell commands (e.g., wxclawbot send --text "...") presents a significant risk of shell injection. Additionally, the skill requires access to sensitive credentials stored in '~/.openclaw/openclaw-weixin/accounts/', which, combined with the ability to exfiltrate data to an external IM service, warrants a suspicious classification despite the lack of clear malicious intent.