# Base-rate card — Atom

> Base-rate card — Atom is an MCP server for agent tooling. Pay-per-call API and MCP check that audits a base-rate card for internal consistency. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Platforms:** API
- **Website:** https://atoms.pipavia.com
- **Repository:** https://github.com/nrupala/atom-services/tree/HEAD/services/base-rate-card
- **MCP Registry name:** org.aimlds/atom-base-rate-card
- **Categories:** agent tooling
- **Listing updated:** 2026-11-18

## About

The README describes this as one atom in a collective repo of small, bounded judgment services, sold per unit delivered on the Conduit Atoms catalogue. This atom takes a base-rate card manifest covering the claim, the reference class with its counts and stated rate, a steelman, a fallacy spot-check, calibration and second-order judgment, and returns a deterministic verdict on whether the card is complete and internally consistent.

The checks are computed rather than judged by a model. The most concrete one re-derives the stated rate from the reported successes divided by class size, and another confirms the card names the fallacies it screens for. Per-check evidence comes back with the verdict so a reviewer can see which section failed.

The service does not source reference-class data or verify that the underlying counts are real. It is aimed at analysts and teams who want a repeatable audit of how a base-rate argument is documented before it is relied on.

## Features

- **Rate re-derivation check:** Recomputes the stated base rate from declared successes and class size to flag arithmetic mismatches.
- **Six-section card review:** Scores claim and evidence, reference class, steelman, fallacy spot-check, calibration and judgment separately.
- **Fallacy coverage check:** Tests whether the card addresses the named fallacies the skill screens for.
- **Per-check evidence output:** Returns each check's evidence alongside a CARD_COMPLETE, CARD_NOT_COMPLETE or INSUFFICIENT_EVIDENCE verdict.
- **Internal consistency only:** Audits the card's own record and does not source or validate external reference-class data.

## Good for

- Auditing a base-rate argument before it is used in a forecast or decision memo
- Checking that a reference class declaration includes counts and a stated rate
- Standardising review of probabilistic claims across an analyst team
- Adding a deterministic completeness gate to an agent workflow that produces base-rate cards

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http atom-base-rate-card https://atoms.pipavia.com/mcp/base-rate-card
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://atoms.pipavia.com/mcp/base-rate-card
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "atom-base-rate-card": {
      "url": "https://atoms.pipavia.com/mcp/base-rate-card"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "atom-base-rate-card": {
      "type": "http",
      "url": "https://atoms.pipavia.com/mcp/base-rate-card"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "atom-base-rate-card": {
      "serverUrl": "https://atoms.pipavia.com/mcp/base-rate-card"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "atom-base-rate-card": {
      "type": "http",
      "url": "https://atoms.pipavia.com/mcp/base-rate-card"
    }
  }
}
```

## FAQ

### Does it check whether the reference-class data is accurate?

No. The listing says it checks the card's internal consistency and does not source reference-class data.

### How is it accessed?

Over the Conduit Atoms API at /v1/base-rate-card or as an MCP server at /mcp/base-rate-card, using a gateway key.

### What does it return?

A verdict of CARD_COMPLETE, CARD_NOT_COMPLETE or INSUFFICIENT_EVIDENCE across six sections, with per-check evidence.

### How do I add Base-rate card — Atom to Claude Code?

Run this command in your terminal: claude mcp add --transport http atom-base-rate-card https://atoms.pipavia.com/mcp/base-rate-card

### What are the best alternatives to Base-rate card — Atom?

The closest alternatives to Base-rate card — Atom are BotShelf Vampire, Agent interoperability design — Atom, Agent context packaging — Atom, Zero trust evaluation verdict — Atom and Zero knowledge claim audit — Atom.

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [BotShelf Vampire](https://ai-powered.com/mcp/botshelf-vampire): BotShelf Vampire is an open-source MCP server for agent tooling.
- [Agent interoperability design — Atom](https://ai-powered.com/mcp/agent-interoperability-design-atom): Agent interoperability design — Atom is an MCP server for design and media.
- [Agent context packaging — Atom](https://ai-powered.com/mcp/agent-context-packaging-atom): Agent context packaging — Atom is an MCP server for agent tooling.
- [Zero trust evaluation verdict — Atom](https://ai-powered.com/mcp/zero-trust-evaluation-verdict-atom): Zero trust evaluation verdict — Atom is an MCP server for ai models and llm tooling.
- [Zero knowledge claim audit — Atom](https://ai-powered.com/mcp/zero-knowledge-claim-audit-atom): Zero knowledge claim audit — Atom is an MCP server for security.
- [Network exposure audit — Atom](https://ai-powered.com/mcp/network-exposure-audit-atom): Network exposure audit — Atom is a paid MCP server for security.
- [BotMarket](https://ai-powered.com/mcp/botmarket): BotMarket is a free MCP server for agent tooling.
- [WritersLogic agent-infra](https://ai-powered.com/mcp/writerslogic-agent-infra): WritersLogic agent-infra is a paid MCP server for agent tooling.

---
Source: https://ai-powered.com/mcp/base-rate-card-atom
