A standardized, vCard-like format for publishing and discovering how AI agents can be contacted across different platforms.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-contact-card
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 agent-contact-card 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 Agent Contact Card?
Agent Contact Card provides a structured yet flexible way for AI agents to find and talk to one another. By leveraging a simple Markdown file with YAML frontmatter, this skill enables a discovery protocol similar to a vCard but optimized for the autonomous agent era. It is a critical component for users of Openclaw Skills who want to establish a professional, machine-readable presence on the web.
This skill facilitates seamless agent-to-agent communication by defining clear channels, capabilities, and routing rules. Whether you are hosting an agent for scheduling, support, or data processing, the Agent Contact Card format ensures that other AI entities can interact with your agent using its preferred protocols, such as webhooks, Discord, or email.
Agent Contact Card Use Cases
- Establishing a public discovery point for your personal AI agent at a /.well-known/ URL.
- Allowing your AI agent to automatically find and contact another person's agent to handle scheduling.
- Defining multi-agent environments where different agents handle specialized tasks like technical support or billing.
- Setting up privacy tiers to control which agents can see specific contact channels.
How Agent Contact Card Works
- The system attempts to discover an agent card by fetching the /.well-known/agent-card path on a specific domain.
- It parses the YAML frontmatter to extract structured data, including versioning and specific communication channels.
- The AI reads the Markdown prose within the file to understand natural language routing rules and context.
- Based on the task at hand, the agent selects the most appropriate channel (e.g., a webhook for automated POST requests).
- The agent executes the communication according to the authentication and format requirements specified in the card.
Agent Contact Card Setup
To set up an Agent Contact Card for use with Openclaw Skills, you need to host a markdown file on your domain. Follow these steps:
# Create the necessary directory on your web server
mkdir -p .well-known
# Create the agent-card file
touch .well-known/agent-card
Edit the file to include your agent's configuration. Ensure the server serves this file with appropriate permissions so other Openclaw Skills can access it via HTTPS.
Agent Contact Card Data Schema & Taxonomy
The Agent Contact Card uses a combination of YAML and Markdown to organize data effectively:
| Field | Type | Description |
|---|---|---|
| version | String | The spec version, currently "1". |
| human_contact | String | Contact details for the human owner. |
| channels | Object | Defines methods like email, discord, or webhook. |
| capabilities | Array | Lists what the agent can do (e.g., scheduling). |
| agents | Array | Defines multiple specialized agents for complex setups. |
| last_updated | ISO Date | Timestamp of the last modification. |
Agent Contact Card Advanced Features
- Privacy Tiers: Use random UUIDs in URL patterns to create private or restricted access agent cards.
- Multi-Agent Setups: Define a hierarchy of specialized agents with distinct IDs and handling capabilities.
- Custom Webhook Specs: Specify HTTP methods, custom headers, and expected JSON payloads for deep integrations.
- Natural Language Routing: Use the Markdown body to provide complex logic that LLMs can interpret for sophisticated message routing.
SKILL.md
Loading
Related Openclaw Skills
Sentry Issue Tracker
A specialized tool for fetching and analyzing Sentry error reports and stack traces directly within AI agent workflows.
AIOZ UI V3 Design System Skill
An automated bridge for converting Figma MCP design data into production-ready React components using the AIOZ UI V3 design system.
AIOZ UI V3 Figma-to-Code
A specialized developer tool for mapping Figma design tokens and components directly to the AIOZ UI V3 React framework.
XML Reader for Construction Data
A specialized Python-based parser designed to convert complex construction industry XML files into structured pandas DataFrames.
Joplin CLI
A comprehensive command-line interface for managing Joplin notes, notebooks, and todos with full WebDAV synchronization and Kanban support.
Tado Smart Thermostat Control
A comprehensive integration for managing Tado smart thermostats, zones, and energy usage through automated API commands.
Featured*
