决策质检 M2M API · Decision QC M2M API

An AI-run "devil's advocate" QC service for other AIs. When your agent makes a decision, it immediately gives the sharpest counter-argument — pay per call, settled in USDT in real time. An AI-run "devil's advocate" QC service for other AIs. Your agent makes a decision; it returns the sharpest counter-argument. Pay-per-call, settled in USDT.

English below. Chinese first.

What is this / What it is

Before your AI agent makes a major decision, call this endpoint and you'll get a structured "counter-opinion" — the strongest rebuttal, key risks, and one question you should ask yourself. It helps you (and your agent) avoid pitfalls.

Before your AI agent commits to a big decision, call this endpoint to get a structured counter-argument: the strongest rebuttal, key risks, and one question you should ask yourself. Helps you (and your agent) avoid blind spots.

• Powered by DeepSeek, genuinely generated counter-arguments (not templates)

• 0.1 USDT(TRC20) per call, pay before service (TRC20 minimum deposit is 0.1 USDT, so a single call is a single credited transfer)

• Payments enter the seller's wallet in real time, verified on-chain, no escrow or commission

• Also an MCP server, automatically discoverable and callable by clients such as Claude/Cursor/Cline

Powered by DeepSeek (real generated counter-arguments, not templates). 0.1 USDT(TRC20) per call, pay-first. Payments land in the seller wallet in real time, verified on-chain, no escrow. Also an MCP server, discoverable by Claude/Cursor/Cline etc.

Related MCP server: multi-model-mcp

Online address / Live endpoint

https://random-monetize.onrender.com

Landing page (switchable between Chinese and English): https://random-monetize.onrender.com/

Landing page (zh/EN toggle): https://random-monetize.onrender.com/

How to call it / How to call

REST (simplest)

curl -X POST https://random-monetize.onrender.com/api/argue \
  -H "Content-Type: application/json" \
  -d '{"decision":"Quit my stable job to run a one-person company?","caller":"your-agent-id","tx_hash":"YOUR_USDT_TX_HASH"}'

Payment method / Payment: first transfer 0.1 USDT(TRC20) to the seller wallet (TRC20 minimum deposit is 0.1 USDT, so a single transfer is a single credited transaction), get the transaction hash tx_hash, then call.

Payment: first send 0.1 USDT(TRC20) to the seller wallet (TRC20 minimum deposit is 0.1 USDT per transfer, so one payment = one call), then call with the resulting tx_hash.

• Seller receiving wallet / Seller wallet (TRC20): TYxynR5V17FYB49osvU3rUkiD7vW2ov634

• Unit price / Price: 0.1 USDT per call (TRC20 minimum deposit 0.1 USDT)

• Calling directly without payment will be rejected / Without payment: rejected with payment_required

MCP (standard, automatically discovered by AI clients)

Endpoint / Endpoint: POST https://random-monetize.onrender.com/mcp (protocol 2024-11-05)

Tool / Tool: decision_qc, Parameters / params: decision / caller / tx_hash

Python embedding in your agent / Embed in your agent

See buyer_agent.py — embed the call_qc() inside it into your workflow; whenever the agent is about to make a decision, run QC first.

See buyer_agent.py — embed call_qc() into your workflow; QC before every decision.

Why it's worth connecting / Why wire it in

• Your agent gets a pair of "opposing eyes" when making decisions, and one less crash

• Extremely low cost (0.1 USDT ≈ ¥0.72), far cheaper than one wrong decision

• Machine-to-machine, zero human labor, pay and use immediately

Your agent gets a "second opposing eye" before deciding, for a tiny cost (0.1 USDT ≈ ¥0.72) — far cheaper than one bad call. Machine-to-machine, zero human in the loop.

Honest boundaries / Honest note

The service, payment collection, on-chain payment verification, and MCP discovery channel are all ready. Current real external paid calls are 0 — this is not a code issue, it's "not yet discovered and used by enough agents". If you use it and find it valuable, forwarding it to friends who also run agents is the best promotion.

The service, payout, on-chain payment verification, and MCP discovery are all live. Real external paid calls are currently 0 — not a code issue, just "not yet discovered by enough agents". If you use it and find it useful, passing it to fellow agent-runners is the best promo.

Local run / Run locally

pip install -r requirements.txt   # 零依赖，仅标准库也可跑
export DEEPSEEK_API_KEY=你的key    # 不填则返回演示反对意见
python seller.py                   # 监听 0.0.0.0:8001

Check the ledger / View ledger: open in browser http://127.0.0.1:8001/status

See ledger at http://127.0.0.1:8001/status

Deployment / Deploy

Deployed to Render (render.yaml at repo root). After code changes, you must click Manual Deploy → Deploy latest commit on the Render service page for it to take effect (git push does not auto-deploy).

Deployed on Render (render.yaml at repo root). After pushing, click Manual Deploy → Deploy latest commit on Render (git push alone does not auto-deploy).

File list / Files

• seller.py: Decision QC API + MCP server (seller AI)

• buyer_agent.py: real external caller template, embeddable into your agent

• requirements.txt: zero dependencies (standard library only); on-chain auto-transfer dependencies see requirements-payout.txt

• smithery.yaml: MCP marketplace release configuration

• render.yaml: Render deployment configuration