ABMeter
ABMeter is a MCP server that lets AI agents manage feature flags and A/B experiments through ABMeter’s hosted streamable HTTP endpoint.
About
ABMeter MCP (ai.abmeter/abmeter) connects Model Context Protocol clients to ABMeter’s feature-flag and A/B testing platform with workflows designed for AI agents. developers shipping SaaS or web products use it in the Grow phase when they need to toggle features, define experiments, and inspect results while coding in Claude Code or Cursor—instead of context-switching to a separate UI for every change. The registry lists a streamable-http remote at mcp.abmeter.ai, with website and repository pointers for setup. It fits teams that treat experimentation as infrastructure: ship a variant, measure impact, roll back safely. Intermediate complexity reflects that you should understand experiment design and statistical basics before letting agents mutate production flags. Pair it with your analytics stack; it is an experimentation MCP, not a full product analytics warehouse.
- Hosted MCP at https://mcp.abmeter.ai (streamable HTTP)
- Feature flagging plus A/B testing oriented to AI agent workflows
- Publisher docs at abmeter.ai/docs and connect flow at abmeter.ai/lab/connect-mcp
- Open-source repo github.com/abmeter/abmeter (registry metadata)
- Version 0.1.0—experimentation platform, not a static analytics skill
ABMeter by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add --transport http abmeter https://mcp.abmeter.ai/Add your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| Transport | HTTP |
|---|---|
| Auth | None |
What it does
Let your coding agent create flags, read experiment results, and run AI-first A/B workflows from ABMeter without clicking the dashboard for every change.
Who is it for?
Best when you already use or plan to use ABMeter and want agent-driven flag and experiment operations.
Skip if: Pre-launch products with no traffic, or teams that only need read-only Google Analytics without experimentation APIs.
What you get
After MCP connect, your agent can drive ABMeter experimentation workflows from the IDE so shipping and measuring variants stays in one loop.
- Connected ABMeter remote MCP in agent config
- Agent-operable feature flag and experiment tooling
- Tighter loop between code changes and measured variant outcomes
By the numbers
- Registry version 0.1.0
- Remote URL https://mcp.abmeter.ai (streamable-http)
- GitHub source: github.com/abmeter/abmeter
Recommended MCP Servers
1ClickReport40 tools for marketing analytics, campaign management, and AI monitoring across 6 platforms.
222wcnm BilistalkermcpTrack Bilibili creators and get the latest updates on videos, dynamics, and articles.8
Abs McpQuery the ABS in plain English — CPI, labour force, building approvals, migration, GDP.
Ad Creative IntelligenceCreative fatigue, winning-angle, competitor ad, and hook benchmark intelligence.
AdAdvisor MCP ServerQuery Meta Ads performance data — accounts, campaigns, ad sets, ads, metrics & settings.
ADCIRC Model Setup MCP ServerADCIRC model configuration debugging, parameter lookup, and validation5How it compares
Experimentation and feature-flag MCP, not a general SEO or ads reporting connector.
FAQ
Who is ai.abmeter/abmeter for?
It is for developers and small teams running ABMeter who want Claude Code, Cursor, or similar agents to work with flags and A/B tests via MCP.
When should I use ai.abmeter/abmeter?
Use it in Grow when you are iterating on funnels, pricing tests, or feature rollouts and need analytics-phase experimentation from the agent.
How do I add ai.abmeter/abmeter to my agent?
Follow ABMeter’s connect guide at https://abmeter.ai/lab/connect-mcp and add the https://mcp.abmeter.ai remote in your agent’s MCP configuration.