Skip to content
MCP server

Rein Agent Risk Scale

Self-check MCP server giving AI agents that spend money an indicative A-D risk grade

Install
Pricing
Free
Platforms
API
Runs
Remotely
Version
0.2.0
Listing updated
1 day ago

About Rein Agent Risk Scale

Rein Agent Risk Scale is a free MCP server for finance and payments. It runs as a hosted remote server, so there is nothing to install locally.

Rein grades the controls around an agent that can spend money, not the model behind it. An agent or its operator answers structured questions about who is accountable, where spend caps are enforced, payee limits, a stop switch and who holds it, human escalation, monitoring, history and incidents. Fixed published rules, with no language model, turn the answers into a grade from A to D, or NR for not rated.

The grade is explicitly self-declared and unverified, and the site says it is not a credit rating. Rated grades stay NR until the accountable principal is verified through KYB, and the site says no deployment has been rated yet. A sandbox eval with fake money can also shift the indicative grade, and a separate free three-day Harness Report tests agent behaviour.

The endpoint is remote streamable HTTP with no sign-in and three tools. The examples repository (MIT) has short samples for the plain MCP Python SDK, the OpenAI Agents SDK, LangGraph and CrewAI. Answers are stored under a pseudonymised record, and a handle is only published if you opt in.

Install

Run in a terminal

claude mcp add --transport http rein-agent-risk-scale https://rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io/mcp

Add to Settings → Connectors → Add custom connector

https://rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "rein-agent-risk-scale": {
      "url": "https://rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "rein-agent-risk-scale": {
      "type": "http",
      "url": "https://rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "rein-agent-risk-scale": {
      "serverUrl": "https://rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "rein-agent-risk-scale": {
      "type": "http",
      "url": "https://rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io/mcp"
    }
  }
}

Tools this server provides (3)

Taken from the project documentation; the running server may differ.

self_check
Scores a deployment from structured answers
get_scale
Returns the A-D scale and what each grade means
explain_grade
Says what a grade requires

What it does

Rule-based scoring
The same answers always produce the same grade; no language model is involved.
Control-focused grading
Looks at spend caps, payee limits, stop switches, monitoring and accountability, not model quality.
Explained results
Responses list the findings behind the grade and steps that would raise it.
Sandbox eval mode
Agents can run a short fake-money eval that influences the indicative grade.
Framework examples
Short samples for the MCP SDK, OpenAI Agents SDK, LangGraph and CrewAI.
Private by default
Answers are stored under a pseudonymised record unless you opt in to publishing.

Good for

  • Check the spend controls on a payments-capable agent
  • Learn what an A-grade deployment requires
  • Add a risk self-assessment step to an agent pipeline
  • Compare how controls change the grade

Pricing

Free

Free self-check; no sign-in required

Prices change. Check rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io before you buy.

Rein Agent Risk Scale FAQ

Is the grade a credit rating?

No. The site says it is indicative, self-declared, not a credit rating and not an offer of credit.

Does it need an account?

No. The MCP endpoint takes no sign-in.

Has anything been officially rated?

The site says no deployment has been rated by Rein yet, and rated grades stay NR until KYB verification.

Which examples are provided?

Python files for the plain MCP SDK, OpenAI Agents SDK, LangGraph and CrewAI.

Is Rein Agent Risk Scale free?

Yes. Rein Agent Risk Scale is free to use.

How do I add Rein Agent Risk Scale to Claude Code?

Run this command in your terminal: claude mcp add --transport http rein-agent-risk-scale https://rein-scale.mangowave-ad191d9c.eastus.azurecontainerapps.io/mcp

What are the best alternatives to Rein Agent Risk Scale?

The closest alternatives to Rein Agent Risk Scale are REA, Solana MCP by Vybe, Tapetide - Indian Stock Market MCP, Agent Roast and Midplane.

Alternatives to Rein Agent Risk Scale

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Remote MCP server exposing the Vybe Solana API to AI clients and code editors

    • 44 tools
    • Remote
    • 2.8K GitHub stars
    • Finance and payments
  3. Indian NSE and BSE stock market data for AI assistants through 48 MCP tools

    • 17 tools
    • Remote
    • Freemium
    • 150 GitHub stars
    • Finance and payments
  4. Agent Roast

    MCP server

    Offline review that flags agent changes which weaken tests, assertions or CI

    Runs locally
    • Open source
    • 4 Hacker News points
    • Developer tools
  5. Midplane

    MCP server

    Postgres gateway for AI agents with table grants, column masking and approval for writes

    Runs locally
    • Open source
    • 5 GitHub stars
    • Databases
  6. Budget Tracker

    MCP server

    Local MCP expense ledger that logs receipts and mileage, splits VAT and prepares rebilling

    • 14 tools
    • Remote
    • Freemium
    • 2 GitHub stars
    • Finance and payments