Agent Tools (x402 Discovery)

MCP Tools

Discover and call x402 paid services (token signals, on-chain Q&A, DeFi plans, LLM chat, etc.) via the agent-tools.cloud directory. Use when the user wants on-demand paid APIs that settle in USDC on Base — no signup, no API keys, pay per call. Tools - `search(intent)` finds matching services by natural language, `get(slug)` returns the full call template (URL, schema, price, x402 bazaar info), `list_categories()` browses, `stats()` shows directory size.

Install

openclaw skills install agent-tools-cloud

agent-tools

Discovery + payment shim for the x402 paid-API ecosystem. Lets an agent find a paid service by intent, read its full call template, and call it (pay-per-call USDC on Base).

Backed by agent-tools.cloud, an open directory of 2 000+ paid x402 endpoints aggregated from awesome-x402, x402scan, x402.org/ecosystem, and a handful of self-hosted services on the same host (LLM chat, token signal, on-chain Q&A, DeFi planner …).

When to use

Use this skill when the user asks for any of:

"find an API that does X" / "is there a paid service for Y"
"call a token signal / on-chain query / DeFi planner / chat completion"
"what's available on x402" / "what x402 services exist"
Any task that needs a paid third-party API and the user has not pinned a specific provider — search first, then decide.

Do not use this skill for:

Free, well-known APIs (CoinGecko, Defillama public). Call those directly.
The user explicitly named a non-x402 provider (OpenAI, Anthropic, etc.).

How to use

1. Find a service

The MCP server (installed via the installer specs above) exposes four tools:

Tool	Purpose
`search(intent, top_k?, max_price_usd?, category?)`	Natural-language search across the directory
`get(slug)`	Full record for one service (URL, schema, price, x402 bazaar info)
`list_categories()`	Browse available categories
`stats()`	Live directory size + health

Most workflows start with search, pick the highest-ranked result whose price_usd fits the user's budget, then get(slug) to read the full call template.

2. Read the call template

For services hosted on agent-tools.cloud the get() payload includes the full x402 v2 extensions.bazaar block — request body example, JSON Schema, output example — so the agent can construct a request without trial-and-error.

Third-party entries are passed through as scraped; some include bazaar metadata, some don't. When in doubt, call the endpoint with an empty body and read the 402 challenge for guidance.

3. Call + pay

POST to the endpoint with no payment header → receive HTTP 402 + payment-required header (base64-encoded x402 v2 challenge).
Decode, sign with a Base-mainnet USDC wallet, attach as X-Payment header.
Re-POST → receive 200 + the actual response.

Use any x402-compatible payment lib (x402-axios, x402-fetch, the Python x402.payment helpers, …) to handle steps 1–3.

Configuration

Set in ~/.openclaw/openclaw.json (optional):

{
  "skills": {
    "entries": {
      "agent-tools": {
        "enabled": true,
        "env": {
          "AGENT_TOOLS_API_BASE": "https://agent-tools.cloud"
        }
      }
    }
  }
}

AGENT_TOOLS_API_BASE lets you point at a self-hosted directory if you ever deploy your own.

Cost

The directory itself is free (search, get, list_categories, stats all unauthenticated, no rate limit announced). Only the underlying paid services charge — typically $0.001–$0.50 per call.

Safety

The skill never auto-pays. It only discovers and returns call templates. The agent / user is in full control of which 402 challenge to actually settle.
Always show price_usd to the user before paying.
Cap max_price_usd in search() if the user mentioned a budget.

Project home: https://agent-tools.cloud
Source: https://github.com/JoursBleu/agent-tools-mcp
PyPI: https://pypi.org/project/agent-tools-mcp/
x402 spec: https://x402.org