Whatsapp Auto Reply

Generates polite, context-aware reply suggestions for WhatsApp messages to assist in personal chats, customer support, and AI response planning.

Do not install or enable this skill until the developer clarifies and fixes the inconsistencies. Specific actions to take before proceeding: 1) Ask the publisher to reconcile SKILL.md, README, manifest, and code — the skill either should only generate reply suggestions (no network send) or must explicitly declare it will send messages and require WHATSAPP_API_KEY. 2) Verify the external endpoint (https://api.whatsapp-service/send): who operates it, is it trustworthy, and is TLS/certificate properly configured. 3) Require the skill to declare WHATSAPP_API_KEY in registry metadata and explain minimal required permissions for that key. 4) If you must test, run in a network-isolated sandbox with a throwaway API key and monitor outbound requests. 5) Prefer skills from known sources or with transparent manifests; rotate and limit any keys that were used with this skill. If the developer cannot justify the external send behavior and missing credential declaration, treat the skill as untrusted.

Type: OpenClaw Skill Name: whatsapp-auto-reply Version: 0.1.0 The skill exhibits a critical discrepancy between its documentation and implementation: SKILL.md explicitly instructs the AI agent that the skill 'does NOT access WhatsApp directly' and only generates suggestions, whereas skill.py performs a POST request to an external, non-standard endpoint (https://api.whatsapp-service/send). This deception could lead an AI agent to transmit sensitive user data to an unverified third-party service under the false impression that the processing is local or non-networked.