Let Your Agent Call Any AI Tool in One Click

Open-sourcing the Souyi AIMALL MCP Server · zero dependencies · pay-as-you-go · $5 free credit on signup

The Problem: Agents Want AI Tools, but Integration Cost Kills Them

Every agent builder hits the same wall today:

• Want to add "web search" or "image generation"? You sign up at each vendor, read docs, wire up SDKs, and manage separate keys.
• Every vendor bills differently; balances are scattered, so you never know what a single task cost.
• Switching models means code changes; switching vendors means rewriting the adapter.

80% of agent projects die before their first successful external AI call.

The Fix: Make AI Tools a Plug-and-Play Standard Part

Souyi AIMALL is an AI API marketplace built for agents. Models and tools are unified behind one endpoint — /api/v1/ai/request — authenticated with one X-API-Key, metered per call, returning a unified order and balance.

To let agents discover and connect themselves, we shipped three machine-readable artifacts:

1. llms.txt — an LLM/crawler overview plus a 3-step self-serve guide;
2. openapi.json — the full OpenAPI 3.0.3 contract;
3. .well-known/agent.json — an A2A-style Agent Card.

On top of that, we open-sourced the official MCP Server: it wraps the marketplace into 5 standard MCP tools any MCP client can load in one click.

Zero Dependencies: Python 3.8+ Without pip install

Most MCP servers depend on a pile of third-party packages. Ours uses only the Python standard library (stdio + JSON-RPC). Clone and run python3 server.py.

export AIMALL_API_KEY="ak_your_key"   # from the "My Agents" page on souyi.net.cn
python3 server.py

Five Tools Covering the Whole Flow

Tool	Purpose
list_assets	List every AI asset (ID, type, unit price)
ai_estimate	Estimate a call's cost (no charge) — use as a budget gate
ai_request	Run the call; returns result + order ID + remaining balance
bootstrap_account	Self-serve signup + agent creation; fetches the key end-to-end
get_onboarding	Returns onboarding steps and doc links

Typical workflow: list_assets → ai_estimate → approve → ai_request.

Connect to Claude Desktop in 30 Seconds

{
  "mcpServers": {
    "aimall": {
      "command": "python3",
      "args": ["/your/path/aimall-mcp/server.py"],
      "env": { "AIMALL_API_KEY": "ak_your_key" }
    }
  }
}

Restart Claude and it can "write code with MiniMax-M3", "generate an image", or "run a web search" — billed from the $5 free credit, no human approval needed. Snippets for Cursor, LangChain, Dify, Coze are in config-examples/.

Why This Matters for the Agent Ecosystem

• Discovery: llms.txt / openapi.json / agent.json let frameworks understand automatically what this platform does and how to call it.
• Unification: one endpoint, one key, one bill — goodbye multi-vendor sprawl.
• Zero friction: $5 on signup, self-serve top-up (planned) to drop human approval — truly unattended.

Get Started

• Platform: https://souyi.net.cn
• Overview: https://souyi.net.cn/llms.txt
• OpenAPI: https://souyi.net.cn/openapi.json
• Agent Card: https://souyi.net.cn/.well-known/agent.json
• MCP Server repo: aimall-mcp/