Postwall

Secure email gateway for AI agents - human-in-the-loop approval for reading and sending emails. Get your API key at https://postwallapp.com

This skill appears coherent, but before installing: 1) Verify the npm package and GitHub repo (author/maintainers and package publisher) match the official PostWall project to avoid a typosquat. 2) Only provide a scoped API key with minimal privileges (and rotate it if needed). 3) Be aware the PostWall service and its CLI will see email content and can change mailbox state (mark-read, submit drafts), so only use with trusted accounts. 4) Test in a non-production mailbox first to confirm behavior (what is marked read, how drafts are handled). 5) Monitor usage and revoke the API key if you see unexpected activity.

Type: OpenClaw Skill Name: postwall Version: 1.7.0 The skill bundle describes a 'secure email gateway' with human-in-the-loop approval for email operations, which is a security-enhancing feature. All described commands for the `postwall` CLI tool are directly related to managing emails (checking, listing, reading, drafting, updating, status checks). Crucially, sending emails requires explicit human approval, mitigating a major risk. The `SKILL.md` contains no prompt injection attempts, obfuscation, or instructions for data exfiltration, persistence, or unauthorized execution. The `npm install` instruction for the `postwall` CLI is a standard dependency installation method.