Bring Ab Testing
to Pydantic AI
Learn how to connect VWO to Pydantic AI and start using 10 AI agent tools in minutes. Fully managed, enterprise secure, and ready to use without writing a single line of code.
GDPR Free for Subscribers
What is the VWO MCP Server?
Connect your VWO (Visual Website Optimizer) account to any AI agent and take control of your experimentation and feature rollout workflows through natural conversation.
What you can do
- Optimization Campaigns — List and monitor active A/B tests and personalization campaigns directly from your agent
- Feature Management — List all feature flags and toggle their status across different environments for instant rollouts
- Experiment Results — Retrieve statistical results for any campaign, including performance metrics and significance levels
- Conversion Tracking — Browse conversion goals and KPIs being tracked to understand your optimization impact
- Audience Segmentation — List pre-defined segments to see how you are targeting different user groups
- Environment Control — Manage features across staging, production, and other configured VWO environments
How it works
1. Subscribe to this server
2. Enter your VWO API Token and Account ID
3. Start managing your experiments and feature flags through Claude, Cursor, or any MCP-compatible client
No more jumping between experiment dashboards to check test results. Your AI agent becomes your optimization engine.
Who is this for?
- Growth Managers — monitor experiment results and statistical significance without manual data exports
- Product Managers — manage feature rollouts and toggle flags across environments during launches
- Data Analysts — quickly surface conversion goal metrics and audience segment performance
- CRO Specialists — audit active campaigns and identify optimization opportunities through simple chat commands
Built-in capabilities (10)
Retrieves details for a specific VWO campaign
Retrieves statistical results for a VWO campaign
Retrieves configuration for a specific VWO feature flag
Retrieves details about the authenticated VWO account
Lists all audience segments
g. clicks, revenue) are being tracked. Lists all conversion goals and metrics
Lists all VWO feature flags
Lists all VWO A/B test and personalization campaigns
g. Production, Staging) for feature flag management. Lists all configured VWO environments
Provide the feature ID and the desired enabled status (true/false). Enables or disables a VWO feature flag
Why Pydantic AI?
Pydantic AI validates every VWO tool response against typed schemas, catching data inconsistencies at build time. Connect 10 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 VWO integration code
- —
Structured output guarantee: Pydantic AI ensures tool results conform to defined schemas, eliminating runtime type errors
- —
Dependency injection system cleanly separates your VWO connection logic from agent behavior for testable, maintainable code
VWO in Pydantic AI
VWO and 3,400+ other MCP servers. One platform. One governance layer.
Teams that connect VWO 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 | 3,400+ 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 VWO in Pydantic AI
The VWO 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 10 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
VWO for Pydantic AI
Every tool call from Pydantic AI to the VWO MCP Server is protected by DLP redaction, cryptographic audit chains, V8 sandbox isolation, kill switch, and financial circuit breakers.
Frequently asked questions
Can I check the winning variation of an A/B test through the agent?
Yes. The get_campaign_results tool allows your AI agent to pull detailed performance metrics for any experiment, showing improvement percentages and statistical significance for each variation compared to the control.
Is it possible to enable or disable feature flags via chat?
Absolutely. Using the toggle_feature_flag tool, you can instruct your agent to enable or disable any feature flag for a specific rollout or emergency kill-switch scenario, ensuring fast control without opening the VWO dashboard.
How can I see what conversion goals are being tracked for my campaigns?
You can use the list_conversion_goals tool to retrieve all metrics and KPIs (like clicks, form submissions, or revenue) currently being monitored in your VWO account.
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 VWO MCP integration works identically with OpenAI, Anthropic, Google, or any supported provider.
MCPServerHTTP not found
Update: pip install --upgrade pydantic-ai