Snail Mail

A slow-channel inbox for leaving your operator important messages. Use when something notable, abnormal, or decision-requiring happens and the operator should see it — but not urgently enough to interrupt. Also use when the operator asks to see their inbox, mark messages read, or archive items.

This skill is a simple, local inbox implemented in Node and appears to do only what it says: write/read messages at {workspace}/inbox/messages.json and render them for different channels. Before installing, consider: (1) choose a safe OPENCLAW_WORKSPACE location (don't point it at a directory with sensitive secrets) because messages are stored in plaintext JSON; (2) OPENCLAW_CHANNEL controls render formatting—set it appropriately; (3) the skill will potentially run during automated heartbeats and surface urgent unread items, so be comfortable with the agent proactively notifying you; (4) the metadata lists the env vars as required though the script has sensible fallbacks—this is a minor inconsistency but not malicious; and (5) if you want absolute assurance, inspect the included scripts/inbox.js file yourself (it contains no network calls or obfuscated code). Overall the package is coherent and low-risk for typical use.

Type: OpenClaw Skill Name: snail-mail Version: 1.0.2 The OpenClaw AgentSkills bundle for 'snail-mail' is benign. It provides a local, slow-channel inbox system for agent-to-operator messages, storing data in a JSON file within the `OPENCLAW_WORKSPACE`. The `SKILL.md` instructions guide the agent on when and how to use the inbox, and the `scripts/inbox.js` code implements standard file operations (read, write, atomic save) and message management. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection attempts designed to subvert the agent's behavior beyond the skill's stated purpose. The script uses only built-in Node.js modules and includes HTML escaping for rendered output, demonstrating good security practices.