AgentMeet is a peer-to-peer protocol enabling AI agents to coordinate meetings over email without a centralized server or account.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agentmeet
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agentmeet using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is AgentMeet?
AgentMeet is an open-source protocol designed specifically for AI agents to handle the friction of scheduling. By leveraging existing email infrastructure and calendar integrations, it allows agents to communicate availability and confirm bookings autonomously. This tool is a prime example of how Openclaw Skills can extend the capabilities of AI assistants into real-world coordination tasks.
The system is entirely peer-to-peer, meaning there is no middleman server or third-party domain required, ensuring privacy and direct interaction between the parties involved. It transforms the standard email inbox into a coordination layer for autonomous agents.
AgentMeet Use Cases
- Coordinating cross-company meetings between two AI executive assistants without manual intervention.
- Automating lead qualification where an agent sends available slots to a prospect via email.
- Managing internal team syncs by finding intersections of availability among multiple participants.
- Setting up technical interviews by having an agent propose slots based on recruiter and engineer calendars.
How AgentMeet Works
- The AI agent analyzes the user's Google Calendar and identifies available time slots based on defined preferences.
- It generates and sends an HTML email containing a specific protocol payload and clickable mailto links for each slot.
- The recipient selects a slot by clicking a link, which opens a pre-filled reply containing a machine-readable JSON payload.
- The sender's agent monitors the inbox, parses the reply, and automatically creates calendar events for both the sender and the recipient once a slot is confirmed.
AgentMeet Setup
To get started with this project among your Openclaw Skills, run the following command in your development directory:
cd ~/Dropbox/Dev/agentmeet && bun install
You will also need to configure OAuth access for Google Calendar and Gmail to allow the agent to read busy periods and send/receive protocol emails.
AgentMeet Data Schema & Taxonomy
The skill uses a standardized protocol embedded in email bodies as HTML comments. Data is organized as follows:
| Message Type | Description | Payload Content |
|---|---|---|
| INVITE | Initial proposal | List of ISO8601 start/end slots |
| SELECT | Recipient choice | The specific request ID and chosen slots |
| CONFIRM | Final confirmation | Confirmation of the calendar event creation |
| COUNTER | Negotiation | Alternative slots proposed by the recipient |
Configurations are handled via a YAML structure defining working hours, buffers, and timezones.
AgentMeet Advanced Features
- Multi-party scheduling support for finding the intersection of availability across several recipients.
- Viral adoption links included in every invite to facilitate easier setup for new users.
- Custom preference overrides for specific invites, including buffer minutes and max slots to share.
- Support for both Gmail API and standard MCP for broader integration within the Openclaw Skills ecosystem.
SKILL.md
Loading
Related Openclaw Skills
China Stock Analysis
A specialized AI agent skill designed to analyze Chinese and international stock markets to provide technical insights and investment recommendations.
Requirement Clarity Protocol
A structured communication protocol to disambiguate natural language requests and ensure alignment between users, main agents, and downstream workers.
SKU Management Reference
A comprehensive technical reference for designing product hierarchies, naming conventions, and inventory classification logic.
Evidence Report Skill
A standardized reporting protocol for AI agents to verify rumors through traceable evidence chains and multi-source cross-validation.
Mixpanel Analytics
A robust integration for programmatically querying Mixpanel product analytics, user profiles, and conversion funnels via the Data Export API.
If You Vanished Video Generator
An AI-powered video generation skill for creating vertical absence shorts that depict quiet homes and bittersweet solitude through automated captioning and cinematic visuals.
Featured*
