Bring Feature Flags
to Pydantic AI
Learn how to connect GrowthBook to Pydantic AI and start using 15 AI agent tools in minutes. Fully managed, enterprise secure, and ready to use without writing a single line of code.
GDPR Free for SubscribersCompatible with every major AI agent and IDE
Gemini
What is the GrowthBook MCP Server?
Connect your GrowthBook account to any AI agent to streamline your experimentation and feature management workflows through natural language.
What you can do
- Feature Management — List, create, and toggle feature flags across production and staging environments to control rollouts.
- Project Control — Organize your experimentation roadmap by managing projects, their descriptions, and specific settings.
- Environment Visibility — Audit and list all configured environments to ensure flags are deployed correctly across your stack.
- Full Lifecycle — Create, update, or delete projects and environments as your infrastructure and team needs evolve.
- Deep Inspection — Retrieve detailed metadata for specific features and projects to understand their current configuration and state.
How it works
- Subscribe to this server
- Enter your GrowthBook Secret Key
- Start managing flags and experiments from Claude, Cursor, or any MCP-compatible client
Who is this for?
- Product Managers — quickly toggle features for beta testers or check the status of an experiment without opening the dashboard
- Engineering Teams — manage flags and environments directly from the code editor to maintain development flow
- DevOps Engineers — audit environment configurations and project structures via simple natural language queries
Built-in capabilities (15)
Create a new GrowthBook environment
Create a new GrowthBook feature flag (v2)
Create a new GrowthBook project
Delete a GrowthBook environment
Delete a GrowthBook feature flag (v2)
Delete a GrowthBook project
Get a single GrowthBook feature flag (v2)
Get a single GrowthBook project by ID
g., production, staging) used for per-environment feature flag control. List all GrowthBook environments
List all GrowthBook feature flags (v2)
List all GrowthBook projects
Toggle a GrowthBook feature flag on or off
Update an existing GrowthBook environment
Partially update a GrowthBook feature flag (v2)
Edit an existing GrowthBook project
Why Pydantic AI?
Pydantic AI validates every GrowthBook tool response against typed schemas, catching data inconsistencies at build time. Connect 15 tools through Vinkius and switch between OpenAI, Anthropic, or Gemini without changing your integration code. full type safety, structured output guarantees, and dependency injection for testable agents.
- —
Full type safety: every MCP tool response is validated against Pydantic models, catching data inconsistencies before they reach your application
- —
Model-agnostic architecture. switch between OpenAI, Anthropic, or Gemini without changing your GrowthBook integration code
- —
Structured output guarantee: Pydantic AI ensures tool results conform to defined schemas, eliminating runtime type errors
- —
Dependency injection system cleanly separates your GrowthBook connection logic from agent behavior for testable, maintainable code
GrowthBook in Pydantic AI
GrowthBook and 4,000+ other MCP servers. One platform. One governance layer.
Teams that connect GrowthBook to Pydantic AI through Vinkius don't need to source, host, or maintain individual MCP servers. Every tool call runs inside a hardened runtime with credential isolation, DLP, and a signed audit chain.
Raw MCP | Vinkius | |
|---|---|---|
| Server catalog | Find and host yourself | 4,000+ managed |
| Infrastructure | Self-hosted | Sandboxed V8 isolates |
| Credential handling | Plaintext in config | Vault + runtime injection |
| Data loss prevention | None | Configurable DLP policies |
| Kill switch | None | Global instant shutdown |
| Financial circuit breakers | None | Per-server limits + alerts |
| Audit trail | None | Ed25519 signed logs |
| SIEM log streaming | None | Splunk, Datadog, Webhook |
| Honeytokens | None | Canary alerts on leak |
| Custom domains | Not applicable | DNS challenge verified |
| GDPR compliance | Manual effort | Automated purge + export |
Why teams choose Vinkius for GrowthBook in Pydantic AI
The GrowthBook MCP Server runs on Vinkius-managed infrastructure inside AWS — a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts. All 15 tools execute in hardened sandboxes optimized for native MCP execution.
Your AI agents in Pydantic AI only access the data you authorize, with DLP that blocks sensitive information from ever reaching the model, kill switch for instant shutdown, and up to 60% token savings. Enterprise-grade infrastructure, zero maintenance.
* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure
How Vinkius secures
GrowthBook for Pydantic AI
Every tool call from Pydantic AI to the GrowthBook MCP Server is protected by DLP redaction, cryptographic audit chains, V8 sandbox isolation, kill switch, and financial circuit breakers.
Frequently asked questions
Can I enable or disable a feature flag using the AI?
Yes! You can use the toggle_feature tool to change the state of any flag. Simply specify the feature ID and the target environment.
How do I see which environments are available for my flags?
You can use the list_environments tool. It will retrieve all configured environments like production, staging, or development used in your GrowthBook account.
Is it possible to create a new project to group my experiments?
Absolutely. Use the create_project tool by providing a name and optional description. This helps keep your feature flags and experiments organized by team or application.
How does Pydantic AI discover MCP tools?
Create an MCPServerHTTP instance with the server URL. Pydantic AI connects, discovers all tools, and generates typed Python interfaces automatically.
Does Pydantic AI validate MCP tool responses?
Yes. When you define result types as Pydantic models, every tool response is validated against the schema. Invalid data raises a clear error instead of silently corrupting your pipeline.
Can I switch LLM providers without changing MCP code?
Absolutely. Pydantic AI abstracts the model layer. your GrowthBook MCP integration works identically with OpenAI, Anthropic, Google, or any supported provider.
MCPServerHTTP not found
Update: pip install --upgrade pydantic-ai
Explore More MCP Servers
View all →
WakaTime (Coding Stats)
14 toolsTrack your coding activity and productivity metrics directly through WakaTime — monitor projects, goals, and time spent in your IDE.
SigmaMind AI
10 toolsTrain custom computer vision models with your own images and deploy object detection and classification without ML expertise.
Amazon Bedrock KB
6 toolsConnect your AI agent to AWS Bedrock Knowledge Bases — execute semantic searches, managed RAG, and sync vector datasources natively.
arXiv Alternative
4 toolsAccess millions of scientific papers from arXiv — search by author, category, or keyword and fetch metadata directly from the open-access archive.