Gram

v1.0.0Monitoring & Observabilitystable

Securely scale AI usage across your organization. Control plane for building, securing and monitoring your agents, mcp and skills.

agentsaisdkclisgolangmcp
Share:
241
Stars
0
Downloads
0
Weekly
0/5

What is Gram?

Gram is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to securely scale ai usage across your organization. control plane for building, securing and monitoring your agents, mcp and skills.

Securely scale AI usage across your organization. Control plane for building, securing and monitoring your agents, mcp and skills.

This server falls under the Monitoring & Observability category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Securely scale AI usage across your organization. Control pl

Use Cases

Build and secure AI agents, MCP, and skills across organizations.
Monitor agent performance and resource usage.
Implement governance and audit trails for AI systems.
speakeasy-api

Maintainer

LicenseAGPL-3.0
Languagetypescript
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx gram

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Gram

Gram is a control plane for building, securing, and monitoring AI agents and MCP tools at organizational scale, created by Speakeasy API. It lets teams define custom MCP tools from TypeScript functions or OpenAPI documents, enforce access controls, chain tools into higher-order compositions, and observe agent usage across the organization. Platform and infrastructure teams use Gram to govern AI tool access without requiring each developer to manage their own MCP server configurations.

Prerequisites

  • A Gram account — sign up at getgram.ai
  • Node.js 18+ for creating Gram Functions
  • curl available for installing the Gram CLI
  • An MCP-compatible client such as Claude Desktop or Claude Code
1

Install the Gram CLI

Download and install the Gram CLI using the official install script. This adds the 'gram' binary to your PATH.

curl -fsSL https://go.getgram.ai/cli.sh | bash
2

Authenticate with Gram

Log in to your Gram account via the CLI. This stores credentials locally so subsequent commands are authorized.

gram auth
3

Create a new Gram Function project

Scaffold a new TypeScript-based tool project using the Gram Functions framework. Follow the interactive prompts to name and configure your project.

npm create @gram-ai/function@latest
4

Define and deploy a tool

Edit the generated project to define your tool using the Gram SDK, then deploy it to the Gram platform. Deployed tools are immediately available to authorized MCP clients.

5

Connect your MCP client to Gram

Configure your MCP client to connect to your Gram-managed endpoint. The CLI can output the correct config snippet for your client type.

gram mcp config

Gram Examples

Client configuration

Example Claude Desktop config pointing to a Gram-managed MCP endpoint. Replace the URL with the one provided by 'gram mcp config'.

{
  "mcpServers": {
    "gram": {
      "command": "gram",
      "args": ["mcp", "serve"]
    }
  }
}

Prompts to try

Prompts that exercise Gram-managed tools once your agent and tools are deployed.

- "List all available Gram tools for my organization."
- "Run the 'add' tool with inputs a=5 and b=3."
- "Show me the audit log of tool invocations from the past 24 hours."
- "Which agents are currently registered in the Gram control plane?"
- "Deploy the new version of my OpenAPI-backed tool to Gram."

Troubleshooting Gram

gram auth fails or credentials are not recognized

Ensure you have an active Gram account at getgram.ai. If the CLI was just installed, open a new terminal session so the PATH update takes effect. Run 'which gram' to confirm the binary is found.

Deployed tools do not appear in the MCP client

Run 'gram mcp config' to get the current connection details and verify they match what is in your client config. Restart your MCP client after any configuration change.

npm create @gram-ai/function fails

Ensure npm 7+ is installed (it ships with Node.js 15+). Run 'npm --version' to check. Update Node.js if necessary, then retry the scaffold command.

Frequently Asked Questions about Gram

What is Gram?

Gram is a Model Context Protocol (MCP) server that securely scale ai usage across your organization. control plane for building, securing and monitoring your agents, mcp and skills. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Gram?

Follow the installation instructions on the Gram GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Gram?

Gram works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Gram free to use?

Yes, Gram is open source and available under the AGPL-3.0 license. You can use it freely in both personal and commercial projects.

Browse More Monitoring & Observability MCP Servers

Explore all monitoring & observability servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "gram": { "command": "npx", "args": ["-y", "gram"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Gram?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides