VERIFIED

Growthbook

Safety: 55/100

0.0(0)

The GrowthBook MCP server allows LLM clients to interact with GrowthBook for feature flag management and experiment configuration, secured by an API key.

MCP ServersFreemium

Boost this tool

Subscribe to listing upgrades or segmented pushes.

Overview

The GrowthBook MCP server allows LLM clients to interact with GrowthBook for feature flag management and experiment configuration, secured by an API key.

The GrowthBook MCP server is relatively safe for read-only operations and controlled write operations. However, the risk of unauthorized modifications exists if the API key is compromised. Exercise caution when granting write access and monitor changes to feature flags.

Performance is primarily limited by the GrowthBook API's response times. Consider caching frequently accessed data to improve performance.

Cost is primarily determined by GrowthBook's pricing model, which may be based on usage or feature tiers. Monitor API usage to avoid unexpected costs.

TypeScript19 starsMITCI/CD

Connections & Capabilities

Connects To

GitHubTwitter

Capabilities

ReadWrite

Client Compatibility

Claude Desktop

full

Cursor

untested

Windsurf

untested

VS Code (Copilot)

untested

Quickstart

Claude Desktop Config

{
  "mcpServers": {
    "growthbook": {
      "command": "npx",
      "args": ["-y", "@growthbook/mcp"],
      "env": {
        "GB_API_KEY": "YOUR_API_KEY",
        "GB_API_URL": "YOUR_API_URL",
        "GB_APP_ORIGIN": "YOUR_APP_ORIGIN",
        "GB_USER": "YOUR_NAME"
      }
    }
  }
}

Tools (4)

safe

get_experiment

Retrieves details about a specific GrowthBook experiment.

Read-only operation, no side effects.

moderate

create_feature_flag

Creates a new feature flag in GrowthBook.

Non-destructive write operation, but can impact application behavior.

high

update_feature_flag

Modifies an existing feature flag in GrowthBook.

Structural modification that can significantly alter application behavior.

safe

list_experiments

Lists all experiments in GrowthBook.

Read-only operation, no side effects.

Security

Auth Method

API Key

Known Risks

!Compromised API key could lead to unauthorized modification of feature flags and experiments.
!Misconfigured feature flags could negatively impact application behavior.
!Lack of input validation could lead to unexpected errors or security vulnerabilities.
!Reliance on environment variables for configuration can be problematic if not managed securely.

Safety Score

moderate risk

Positive

+Requires API key for authentication.
+Capabilities are limited by the API key's permissions.
+No direct code execution or shell access.
+Primarily interacts with GrowthBook's API, not the underlying system.

Risks

-Write operations are possible, potentially impacting feature flag configurations.
-Exposure of API key could lead to unauthorized modifications.
-Lack of sandboxing could lead to unintended consequences from misconfigured flags.
-Email address is required, raising PII concerns.

Use Cases

Feature Flag Management

-Creating new feature flags for A/B testing.
-Modifying existing feature flags to control feature rollout.
-Retrieving feature flag configurations for decision-making.

Experiment Configuration

-Creating new experiments to test different variations of a feature.
-Retrieving experiment details to analyze results.
-Managing experiment settings, such as traffic allocation.

Integration with LLM Agents

-Automating feature flag creation based on user stories.
-Dynamically adjusting feature flags based on real-time data.
-Generating experiment configurations based on predefined templates.

Who Is This For?

Best For

Teams using GrowthBook for feature flag management.
Automating feature flag creation and modification.
Integrating feature flag management into LLM-powered workflows.
Experimenting with different feature variations.

Not Ideal For

Applications that require real-time feature flag updates.
Environments with strict security requirements that prohibit the use of API keys.
Teams that do not use GrowthBook for feature flag management.

Autonomy & Execution

Default ModeRead and Write enabled

Autonomy is limited by the permissions associated with the API key. Ensure the API key has appropriate restrictions to prevent unintended consequences.

StatelessYes

Retry LogicNot documented

ConcurrencySingle-threaded

Production Tip

Implement thorough testing and monitoring of feature flags and experiments to minimize the risk of negative impacts.

FAQ

What is the GrowthBook MCP server?

It's a server that allows LLM clients to interact with GrowthBook for feature flag and experiment management.

What environment variables are required?

GB_API_KEY (GrowthBook API key) and GB_EMAIL (your email address used with GrowthBook).

What happens if my API key is compromised?

Unauthorized users could modify feature flags and experiments, potentially impacting application behavior.

Can I use a PAT (Personal Access Token) instead of an API key?

Yes, but MCP server capabilities are limited by the PAT's permissions.

What are the default values for GB_API_URL and GB_APP_ORIGIN?

GB_API_URL defaults to `https://api.growthbook.io` and GB_APP_ORIGIN defaults to `https://app.growthbook.io`.

How do I add the MCP server to my AI tool?

Refer to the official GrowthBook documentation for a complete guide.

Is it possible to rollback changes made via the MCP server?

Rollback support depends on GrowthBook's capabilities and is not directly managed by the MCP server.

Metrics

Discovered2/14/2026

Reviews0

Saved By0 users

Related Tools

Growthbook - AI Tool Review & Alternatives | Flaex AI