← 返回 Skills 市场
gizmo-dev

AgentAPI

作者 Gizmolab · GitHub ↗ · v1.0.8
cross-platform ✓ 安全检测通过
1131
总下载
2
收藏
5
当前安装
9
版本数
在 OpenClaw 中安装
/install agentapi
功能描述
Browse and search the AgentAPI directory - a curated database of APIs designed for AI agents. Find MCP-compatible APIs for search, AI, communication, databas...
使用说明 (SKILL.md)

AgentAPI

A curated directory of APIs designed for AI agents. Machine-readable. MCP-compatible. Agent-accessible.

Website: https://agentapihub.com
Docs: https://api.agentapihub.com/api/docs

Free Features

The directory and search functionality is completely free:

# Browse all APIs (FREE)
curl https://agentapihub.com/api/v1/apis

# Search by capability (FREE)
curl "https://agentapihub.com/api/v1/apis?q=send+email&mcp=true"

# Filter by category (FREE)
curl "https://agentapihub.com/api/v1/apis?category=ai"

x402 Pay-Per-Use APIs

Some APIs are available through x402 — a pay-per-use protocol using USDC on Base chain.

Available x402 APIs

API Endpoint Approx. Price
Gemini Chat /api/gemini/chat/completions ~$0.001/req
Gemini Embeddings /api/gemini/embeddings ~$0.0005/req

How x402 Works

  1. Call the API endpoint
  2. Receive 402 Payment Required with price details
  3. Send USDC payment on Base chain
  4. Retry request with payment proof
  5. Receive API response

⚠️ Important Safety Notice

x402 payments require explicit setup and should not be automated without safeguards:

  • Wallet required: You must configure a wallet with USDC on Base
  • User approval recommended: Implement confirmation flows before any payment
  • Verify recipient: The payment recipient is 0xcCb92A101347406ed140b18C4Ed27276844CD9D7 (gizmolab.eth)
  • Set spending limits: Configure maximum per-request and daily limits
  • This skill does not auto-execute payments — it provides documentation only

For implementation details, see: https://api.agentapihub.com/api/docs

Directory Categories

Category APIs Example
Search Brave, Serper, Exa, Tavily, Perplexity Web search with AI summaries
AI & ML OpenAI, Claude, Gemini, Groq, Replicate LLM inference, image gen
Communication Resend, Twilio, Slack, Discord, Telegram Email, SMS, messaging
Database Supabase, Pinecone, Qdrant, Neon, Upstash SQL, vectors, KV store
Payments Stripe, Lemon Squeezy, PayPal Payment processing
Scraping Firecrawl, Browserbase, Apify Web scraping, automation
Developer GitHub, Vercel, Linear, Sentry Code, deploy, issues
Productivity Notion, Google Calendar, Todoist Tasks, scheduling

MCP-Compatible APIs

All 50+ APIs in the directory are MCP-compatible. Filter with ?mcp=true.

API Response Format

{
  "id": "resend",
  "name": "Resend",
  "description": "Modern email API for developers",
  "category": "communication",
  "provider": "Resend",
  "docs": "https://resend.com/docs",
  "auth": "api_key",
  "pricing": "freemium",
  "pricingDetails": "3,000 free/mo, then $20/mo",
  "rateLimit": "10 req/sec",
  "mcpCompatible": true,
  "x402Enabled": false,
  "tags": ["email", "transactional", "notifications"]
}

Top APIs by Category

Search

  • Brave Search - Privacy-focused, 2k free/mo
  • Exa - Neural/semantic search for AI
  • Tavily - Built specifically for AI agents

AI & ML

  • OpenAI - GPT-4, DALL-E, Whisper
  • Anthropic Claude - Best for reasoning/coding
  • Groq - Fastest inference (500+ tok/sec)

Communication

  • Resend - Simple email API, 3k free/mo
  • Twilio - SMS/voice, industry standard
  • Slack/Discord/Telegram - Team messaging

Database

  • Supabase - Postgres + auth + storage
  • Pinecone/Qdrant - Vector DBs for RAG
  • Upstash - Serverless Redis

Usage Examples

# Find an API for sending emails
Search AgentAPI for "email" → Returns Resend, SendGrid, Twilio

# Find vector databases for RAG
Search AgentAPI for "vector embeddings" → Returns Pinecone, Qdrant, Weaviate

# Find fast LLM inference
Search AgentAPI for category "ai" + filter by latency → Returns Groq, Gemini Flash

Contributing

Submit new APIs at https://agentapihub.com (Submit API link in footer).

Built By

GizmoLab (@gizmolab_) — gizmolab.io

安全使用建议
This appears to be a benign, read-only directory for discovering APIs. Before installing: 1) Note the documented x402 pay-per-use flow — this skill only documents payments and does not perform them, but do NOT give any agent or skill access to your wallet/private keys unless you trust it and have explicit confirmation flows and strict spending limits. 2) Verify any recipient addresses yourself (the README lists 0xcCb92A... as gizmolab.eth). 3) If you intend to allow automated payments in other tooling, implement user confirmations and per-request/daily caps. 4) If you need offline or private API discovery, review the public endpoints the skill points to and confirm they meet your privacy/compliance needs.
功能分析
Type: OpenClaw Skill Name: agentapi Version: 1.0.8 The skill bundle primarily consists of documentation and examples for interacting with the AgentAPI directory. The `SKILL.md` file provides `curl` commands for browsing and searching the API, which are read-only GET requests and do not pose an immediate risk. Crucially, the skill explicitly states, 'This skill does not auto-execute payments — it provides documentation only,' mitigating any potential prompt injection risk related to the described x402 payment protocol. There is no evidence of malicious intent, data exfiltration, unauthorized command execution, or manipulative prompt injection attempts.
能力评估
Purpose & Capability
Name/description match the SKILL.md. The content is a read-only, machine-readable directory and search reference for APIs, with example curl calls to public endpoints on agentapihub.com. There are no unexpected dependencies, credentials, or binaries requested.
Instruction Scope
Instructions limit themselves to browsing/searching the public directory and showing the x402 payment flow. The SKILL.md includes a payment recipient address and warns not to automate payments; it does not instruct the agent to read local files, environment variables, or to transmit unrelated data. Caution: if you later grant an agent wallet access (outside this skill), the documented payment flow could be acted upon by other skills or tooling.
Install Mechanism
No install spec and no code files — this is instruction-only, so nothing is written to disk and no third-party packages are installed.
Credentials
The skill declares no required environment variables or credentials. The only resources referenced are public API URLs and a payment address; the SKILL.md's mention of a wallet is advisory and not a required credential of the skill itself.
Persistence & Privilege
always is false and the skill does not request persistent or elevated privileges. It does not attempt to modify other skills or system configuration.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install agentapi
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /agentapi 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.8
Added x402 badges and filter to website
v1.0.7
Free directory + x402 APIs with proper safety documentation
v1.0.6
Safety update: removed payment instructions, now documentation-only
v1.0.5
x402 billing + GizmoLab branding
v1.0.4
x402 billing + GizmoLab branding
v1.0.3
Added x402 pay-per-use billing documentation
v1.0.2
Updated branding to GizmoLab
v1.0.1
Updated domain to agentapihub.com
v1.0.0
Initial release - API directory for AI agents
元数据
Slug agentapi
版本 1.0.8
许可证
累计安装 5
当前安装数 5
历史版本数 9
常见问题

AgentAPI 是什么?

Browse and search the AgentAPI directory - a curated database of APIs designed for AI agents. Find MCP-compatible APIs for search, AI, communication, databas... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1131 次。

如何安装 AgentAPI?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install agentapi」即可一键安装,无需额外配置。

AgentAPI 是免费的吗?

是的,AgentAPI 完全免费(开源免费),可自由下载、安装和使用。

AgentAPI 支持哪些平台?

AgentAPI 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 AgentAPI?

由 Gizmolab(@gizmo-dev)开发并维护,当前版本 v1.0.8。

💬 留言讨论