Skip to main content
Glama

gate402-mcp

x402 trust

MCP server for Gate402 — pay-per-call agent APIs over HTTP 402 (x402 / USDC on Base). Gives any MCP client a full toolkit with no signup and a free tier on first runs — GPU/CPU compute, open-model LLM inference, a GPU marketplace, a complete Base trading loop (discover → vet → read → execute), and web tools:

Tool

What it does

Price

gate402_infer

Open-model LLM inference (Llama 3.1 8B/70B, Qwen 2.5, Mistral); billed on actual tokens, signed usage receipt

per-token (from $0.05/1M in)

gate402_compute

Metered GPU/CPU for a container job, with a signed execution receipt. Not available yet: answers 503 until a compute provider is connected

per-second

gate402_scrape

Fetch any public URL, render JS, strip nav/ads → clean LLM-ready Markdown

$0.002

gate402_scrape_stealth

Cloudflare/anti-bot fetch via residential proxies + CAPTCHA solving → clean Markdown

$0.05

gate402_minify

Compress text to cut downstream LLM token spend (~40%)

$0.005 / 10k tok

gate402_dedup

Semantic vector-cache lookup (exact + cosine)

$0.001 hit / $0.003 miss

gate402_onchain

On-chain wallet/token intel on Base (balances, EOA/contract, tx count, token metadata)

$0.01

gate402_dex

Live DEX price / liquidity / 24h volume for a Base token

$0.01

gate402_news

Recent news headlines + bull/bear sentiment for a ticker/topic

$0.02

gate402_edgar

Latest SEC EDGAR filings (10-K/10-Q/8-K) for a US ticker

$0.02

gate402_token_risk

Rug/tradeability verdict for a Base token (score + SAFE/CAUTION/AVOID; liquidity, honeypot sim, holder concentration)

$0.03

gate402_momentum

Momentum + order-flow signal for a Base token (RISING/FALLING + ACCUMULATION/DISTRIBUTION, honeypot-gated)

$0.02

gate402_best_swap

Best-execution intel: which DEX pool + estimated price impact for a trade size

$0.02

gate402_launches

Radar of the freshest Base token launches, lightly pre-screened

$0.02

…plus three free tools that run locally in this process (pure compute — no payment, no key, no network):

Tool

What it does

gate402_token_count

Estimate the token count of a string (budget your context window)

gate402_html_to_md

Convert an HTML string you already have into clean Markdown

gate402_json_repair

Coerce malformed / LLM-mangled JSON into valid JSON

GPU marketplace

Gate402 also routes to third-party GPU/compute providers with escrow-on-success — the buyer is charged only if the provider delivers. Providers keep 85% of each call, paid out on-chain in USDC on Base.

Tool

What it does

gate402_market_infer

LLM inference via the marketplace, routed to the cheapest healthy provider. x402-native (settles on-chain per call — needs an x402 client, not the free-tier key)

gate402_providers

Free. Browse active providers — model, per-call price, reliability

gate402_become_provider

Free. List your own GPU and earn 85%/call — returns the exact sign-message + registration request

How billing works

On first use the server self-claims a free-credit API key from Gate402 and caches it at ~/.gate402-mcp/key.json. Calls draw down that credit. When it runs out, tools return a top-up link instead of failing. To skip the free tier, set GATE402_API_KEY to a funded account (top up).

The payment is the auth — there are no accounts to create.

Related MCP server: AfaAgent x402 API Suite

Install

npm install -g gate402-mcp

Or run without installing via npx gate402-mcp.

Configure your MCP client

Claude Desktop / Claude Code

Add to your MCP config (claude_desktop_config.json, or claude mcp add):

{
  "mcpServers": {
    "gate402": {
      "command": "npx",
      "args": ["-y", "gate402-mcp"]
    }
  }
}

Cursor / Cline / Windsurf

Same shape — point the MCP server command at npx -y gate402-mcp.

Agent frameworks (LangChain, CrewAI, LlamaIndex)

gate402 is a standard stdio MCP server, so any framework with an MCP adapter can load all 17 tools:

LangChain / LangGraph (langchain-mcp-adapters):

from langchain_mcp_adapters.client import MultiServerMCPClient
client = MultiServerMCPClient({"gate402": {"command": "npx", "args": ["-y", "gate402-mcp"], "transport": "stdio"}})
tools = await client.get_tools()   # feed into your agent

CrewAI (MCPServerAdapter):

from crewai_tools import MCPServerAdapter
from mcp import StdioServerParameters
params = StdioServerParameters(command="npx", args=["-y", "gate402-mcp"])
with MCPServerAdapter(params) as tools:
    ...  # pass tools to your Crew

LlamaIndex (McpToolSpec):

from llama_index.tools.mcp import BasicMCPClient, McpToolSpec
tools = McpToolSpec(client=BasicMCPClient("npx", args=["-y", "gate402-mcp"])).to_tool_list()

Once wired, the agent calls the tools autonomously; the free-credit key is claimed on first use, and paid tools draw down from it.

Environment variables

Var

Default

Purpose

GATE402_API_KEY

(unset)

Use a funded account instead of the free tier.

GATE402_BASE_URL

https://gate402.app

Override the gateway (self-hosting / testing).

GATE402_CONFIG_DIR

~/.gate402-mcp

Where the cached free key is stored.

Develop

npm install
npm run build
npm start          # or: npm run dev

License

MIT

Available Tools

7 tools
gate402_dedupA

Semantic vector-cache lookup: exact-match then 0.88 cosine similarity. Returns a cache hit/miss for a query, sub-10ms. Pay-per-call ($0.001 hit / $0.003 miss).

Inputs: query (required), vector, namespace, storeOnMiss.

gate402_html_to_mdA

FREE. Convert an HTML string you already have into clean Markdown. (To FETCH a live page instead, use gate402_scrape.) No payment required.

Inputs: html (required).

gate402_json_repairA

FREE. Coerce malformed / LLM-mangled JSON (trailing commas, single quotes, unquoted keys) into valid JSON. No payment required.

Inputs: text (required).

gate402_minifyA

Compress text to cut downstream LLM token spend (~40%): strips filler, collapses JSON, densifies prose. Pay-per-call ($0.005/10k tokens).

Inputs: text (required), format, aggressive.

gate402_scrapeA

Fetch any public URL, render client-side JS, strip nav/ads, and return clean LLM-ready Markdown. Pay-per-call ($0.002) via Gate402; free tier on first runs.

Inputs: url (required).

gate402_scrape_stealthA

Hardened headless fetch for JS-heavy or lightly bot-protected pages. Use when gate402_scrape is blocked or returns little content. Pay-per-call ($0.05).

Inputs: url (required).

gate402_token_countB

FREE. Estimate the token count of a string (cl100k/o200k tokenizer). Use to budget context windows. No payment required.

Inputs: text (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 7 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: dedup for cache lookup, html_to_md for conversion, json_repair for fixing JSON, minify for compression, scrape for general scraping, scrape_stealth for hardened scraping, and token_count for token counting. The only potential overlap is between scrape and scrape_stealth, but their descriptions explicitly differentiate them by use case, so no ambiguity.

    Naming Consistency4/5

    All tools share the 'gate402_' prefix, which is consistent. However, suffixes vary: some are single verbs (dedup, minify, scrape), others are compound with underscores (html_to_md, json_repair, token_count), and one is a phrase (scrape_stealth). This mixed pattern is minor but prevents a perfect score.

    Tool Count5/5

    With 7 tools, the server covers a well-scoped set of utilities: caching, conversion, repair, minification, scraping (two variants), and token counting. Each tool addresses a distinct need without redundancy, and the count feels appropriate for a utility-focused MCP server.

    Completeness4/5

    The tool set covers core text processing and web scraping tasks, but there are minor gaps such as the lack of a general format converter (e.g., CSV to JSON) or PDF extraction. However, the existing tools form a coherent set for the advertised functionalities, and no critical operations are missing for the apparent scope.

    Maintenance

    ActivityStale
    ResponsivenessUsually handled

    Related MCP Connectors

    Related MCP Servers