A curated, machine-readable directory of MCP-compatible APIs designed specifically for AI agents and developers.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agentapi
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 agentapi 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 AgentAPI?
AgentAPI is a specialized directory built to bridge the gap between AI agents and the services they need to function effectively. It offers a searchable database of over 50 APIs that are structured to be machine-readable and fully compatible with the Model Context Protocol (MCP). By leveraging this resource within the Openclaw Skills ecosystem, developers can effortlessly find tools for web search, LLM inference, messaging, and data storage.
Beyond simple discovery, AgentAPI introduces support for the x402 protocol, a decentralized pay-per-use system. This allows agents to access premium services like Gemini Chat and Embeddings by paying small amounts of USDC on the Base chain, bypassing the need for traditional monthly subscriptions for every individual service an agent might touch.
AgentAPI Use Cases
- Locating MCP-compatible search engines like Brave, Exa, or Tavily for RAG-enhanced agents.
- Finding communication APIs for Resend, Twilio, or Slack to enable agent-to-human notifications.
- Discovering vector databases and key-value stores like Pinecone or Upstash for agent memory.
- Identifying fast LLM inference providers like Groq or Gemini for low-latency responses.
- Researching pay-per-use API options to reduce fixed overhead costs in agentic workflows.
How AgentAPI Works
- Query the AgentAPI database using keyword searches or category filters through standard REST endpoints.
- Filter results specifically for MCP-compatible tools to ensure easy integration with agent frameworks.
- Review the machine-readable JSON metadata for each API, including pricing, rate limits, and authentication types.
- For x402-enabled services, follow the payment-required flow to pay for individual requests via USDC on the Base network.
- Implement the chosen API into your agent's toolset using the provided documentation links and specifications.
AgentAPI Setup
The directory functionality is free and accessible via simple CLI commands. To search for APIs or browse the directory, use the following bash commands:
# Browse the full directory of APIs
curl https://agentapihub.com/api/v1/apis
# Search for a specific capability (e.g., email) with MCP support
curl "https://agentapihub.com/api/v1/apis?q=send+email&mcp=true"
# Filter the directory by a specific category like 'ai'
curl "https://agentapihub.com/api/v1/apis?category=ai"
AgentAPI Data Schema & Taxonomy
Each API entry in the directory returns a structured JSON object. The schema is designed for easy parsing by AI agents within Openclaw Skills:
| Field | Description |
|---|---|
id |
Unique string identifier for the API (e.g., 'resend') |
name |
The official name of the service |
category |
The functional area (search, ai, communication, etc.) |
auth |
Required authentication method (e.g., api_key) |
mcpCompatible |
Boolean flag indicating Model Context Protocol support |
x402Enabled |
Boolean flag indicating if pay-per-use is supported |
pricingDetails |
Human-readable summary of costs and free tiers |
AgentAPI Advanced Features
- Comprehensive MCP support filtering to find agents that work out-of-the-box with Openclaw Skills.
- x402 Pay-Per-Use protocol integration for Gemini and other high-demand AI models.
- Real-time directory searching by capability, allowing agents to find their own tools dynamically.
- Detailed metadata for every provider including rate limits and authentication requirements.
- Support for the Base chain for low-cost, decentralized API payments.
SKILL.md
Loading
Related Openclaw Skills
AgentAPI: AI-Ready API Directory
AgentAPI is a curated, machine-readable directory of APIs specifically designed for AI agents, featuring native MCP compatibility and x402 crypto-billing.
Arena Research
An agentic research tool for exploring human-curated collections and conceptual connection graphs on Are.na.
UxrObserver
An automated UX research ethnographer that passively observes and analyzes user interactions within the OpenClaw environment.
Clawsight UXR Observer
An embedded ethnographic research tool for Openclaw Skills that captures user interactions, surveys, and verbatim feedback to generate daily UX insight reports.
GizmoLab Tools & UI
A comprehensive suite of blockchain developer utilities and React-based Web3 UI components for Ethereum and Solana.
QA Check
A comprehensive quality assurance suite designed to validate web projects and builds before they go live to production.
METADATA
Featured*
