Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Security hardening release: removed shell-command execution path from poll daemon, redacted API key by default in onboarding output, and updated docs to safer defaults.
What This Skill Does
Provisions a dedicated email inbox for an AI agent via thrd.email, keeping it separate from any personal inbox. Supports inbound polling, idempotent sending, reply-all threading, and tiered trust controls including Proof of Reasoning for cold outbound and human verification via X.
Isolates agent email activity from personal inboxes while adding built-in spam controls, tiered trust verification, and idempotent sending that raw SMTP or general-purpose email APIs lack.
When to use it
- Giving an AI agent its own inbox without exposing personal email
- Polling for inbound email replies in an automated workflow
- Sending cold outbound emails with spam-prevention challenges
- Verifying agent ownership by linking a responsible human on X
- Monitoring monthly email usage and delivery status
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Set up an email inbox for my agent and send a reply to an inbound message
- 1Run scripts/onboard.py to provision a new inbox and receive an API key
- 2Set THRD_API_KEY as an environment variable in the runtime secret manager
- 3Sync the latest API contract with scripts/openapi_sync.py
- 4Poll for inbound emails via GET /v1/events and ACK each event
- 5Reply to the target thread using the reply endpoint with reply-all behavior
Agent sends a reply from its dedicated inbox, with delivery status trackable via the API
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.