# futureagievals-mcp

Hosted MCP server connecting your agent to the AI SDK evaluation dashboard. It checks eval gates, summarizes failed traces, and records run receipts.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: clauxel
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/clauxel/futureagi-evalops-mcp-mcp
- Updated: 2026-05-25
- HTML: https://affpapa.org/en/nevibecode/mcp/futureagievals-mcp

## What it does

The server acts as a remote HTTP gateway to an evaluation platform via `https://futureagievals.clauxel.com/mcp`. It equips coding assistants with four tools: `run_eval_gate`, `summarize_trace_failures`, `compare_eval_versions`, and `issue_eval_receipt`. Agents use them to evaluate prompts and SDK logic before production releases, track behavioral changes, and document run evidence.

Production requests fail without a valid bearer token from the provider site. Because this is a hosted remote service rather than a local binary, your MCP environment must support streamable HTTP transports.

## Requirements
- Remote server: `https://futureagievals.clauxel.com/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http futureagievals-mcp https://futureagievals.clauxel.com/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add futureagievals-mcp --url https://futureagievals.clauxel.com/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "futureagievals-mcp": {
            "url": "https://futureagievals.clauxel.com/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "futureagievals-mcp": {
            "type": "http",
            "url": "https://futureagievals.clauxel.com/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http futureagievals-mcp https://futureagievals.clauxel.com/mcp
```

## When to use it
- Execute automated evaluation gates before releases
- Summarize failure root causes from execution traces
- Compare performance metrics across evaluation versions
- Generate verifiable run receipts for workflow audits

## Example prompts
- Execute `run_eval_gate` against the staging test suite.
- Analyze recent run failures using `summarize_trace_failures`.
- Run `compare_eval_versions` between current and previous runs, then issue a receipt.

## Similar
- [evalscopebench-mcp](https://affpapa.org/en/nevibecode/mcp/evalscopebench-mcp.md) — MCP server
- [eval-harness](https://affpapa.org/en/nevibecode/skills/eval-harness.md) — skill · 4700 installs
- [write-judge-prompt](https://affpapa.org/en/nevibecode/skills/write-judge-prompt.md) — skill · 1100 installs
- [ai-system-testing](https://affpapa.org/en/nevibecode/skills/ai-system-testing.md) — skill · 1100 installs
- [generate-synthetic-data](https://affpapa.org/en/nevibecode/skills/generate-synthetic-data.md) — skill · 1000 installs
- [inngest-agent-evals](https://affpapa.org/en/nevibecode/skills/inngest-agent-evals.md) — skill · 556 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md