# Speakeasy Platform MCP

Links coding agents to the Speakeasy AI Control Plane to enforce tool permissions, scrub sensitive data, and monitor token expenses across teams.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed · ★ 275
- Author: speakeasy-api
- License: AGPL-3.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/speakeasy-api/gram
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/mcp/speakeasy-platform-mcp

## What it does

The server interfaces coding assistants with the Speakeasy AI Control Plane over HTTP. Teams use it to govern how assistants discover and call tools, backing access rules with OAuth 2.1 and identity providers such as Okta or Azure AD.

Telemetry relies on OpenTelemetry standards. The gateway screens outgoing prompts and model answers on the fly, redacting personal records and blocking prompt injection attempts. Every tool invocation gets indexed. That makes token budgets and runtime costs straightforward to review.

Connecting to the hosted endpoint requires only `https://ai.speakeasy.com/platform-mcp`. For self-hosting the open-source repository, your machine must run Docker and Mise to execute `./zero`. Without active containers, the local development backend will not boot.

## Requirements
- Remote server: `https://ai.speakeasy.com/platform-mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http platform https://ai.speakeasy.com/platform-mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add speakeasy-platform-mcp --url https://ai.speakeasy.com/platform-mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "speakeasy-platform-mcp": {
            "url": "https://ai.speakeasy.com/platform-mcp"
        }
    }
}
```

### GitHub Copilot

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

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http speakeasy-platform-mcp https://ai.speakeasy.com/platform-mcp
```

## When to use it
- restrict agent tool access with enterprise role policies
- strip API keys and personal data from model prompts
- audit token expenditures across individual coding sessions
- track tool execution history through searchable access logs

## Example prompts
- Check recent agent sessions and summarize token costs
- Inspect the audit log for blocked prompt injection attempts
- List all connected MCP tools and their permission policies

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 386989
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33918
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81115
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119160
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47392
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43872

---
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