Optimizely MCP server

What the model can do

Optimizely Web and Feature Experimentation: projects, experiments with status, traffic, variations and metrics, experiment results (rate, lift and statistical significance per variation), personalization campaigns, audiences, the change history of a project, and feature flags with their on or off state per environment; when you ask, can create and edit experiments, start, pause, resume or archive them, create and edit audiences, start, pause or resume campaigns, create flags, turn a flag on or off in an environment, and archive or restore flags.

Reads

• optimizely_projectsOptimizely projects: id, name, platform (web, custom, ios, android), status, and whether it uses feature flags.
• optimizely_experimentsExperiments in an Optimizely project or campaign: name, key, type, status (running, paused, not_started, concluded, archived), traffic, audiences…
• optimizely_experimentOne Optimizely experiment in full: status, traffic and holdback, audiences, pages, variations with weights…
• optimizely_experiment_resultsResults of an Optimizely experiment: visitors, and per metric and variation the conversion rate, lift over the baseline, statistical significance…
• optimizely_campaignsPersonalization campaigns in an Optimizely project: name, status, holdback, whether concluded and the outcome recorded, and dates.
• optimizely_audiencesAudiences in an Optimizely project: id, name, description, conditions (as Optimizely stores them), how many experiments use each…
• optimizely_changesThe change history of an Optimizely project in a time window (default the last 7 days): what changed (experiment, flag, ruleset, audience…)…
• optimizely_flagsFeature flags in an Optimizely Feature Experimentation project: key, name, description…

Acts, when you ask

These change something in Optimizely. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• optimizely_create_experimentCreates an experiment in an Optimizely project, not started: nobody sees it until it is started.
• optimizely_update_experimentEdits an Optimizely experiment's name, description, traffic, holdback, audiences or schedule.
• optimizely_start_experimentStarts an Optimizely experiment: it is published and shown to live visitors (or users) at its traffic allocation straight away.
• optimizely_pause_experimentPauses a running Optimizely experiment: visitors stop seeing it and stop counting toward its results until it is resumed.
• optimizely_resume_experimentResumes a paused Optimizely experiment, showing it to live visitors again.
• optimizely_archive_experimentArchives an Optimizely experiment: it stops running for everyone and leaves the active list (it can be unarchived in Optimizely).
• optimizely_create_audienceCreates an audience in an Optimizely project from targeting conditions; it targets nobody until an experiment or campaign uses it.
• optimizely_update_audienceEdits an Optimizely audience's name, description or conditions, or archives it (archived true) or restores it.
• optimizely_set_campaign_stateStarts, pauses or resumes an Optimizely personalization campaign; start and resume show its experiences to live visitors, pause stops them.
• optimizely_create_flagCreates a feature flag in an Optimizely Feature Experimentation project.
• optimizely_enable_flagTurns on a feature flag's rules in one environment of an Optimizely project, so the SDKs in that environment start serving its rules to real users…
• optimizely_disable_flagTurns off a feature flag's rules in one environment of an Optimizely project…
• optimizely_archive_flagsArchives the named feature flags in an Optimizely project; Optimizely refuses flags that code still references unless override_code_references is set.
• optimizely_unarchive_flagsRestores the named archived feature flags in an Optimizely project.

Connect it

First connect Optimizely on Dashboard → Connections, then add Memtro to your client once. The Optimizely tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared Optimizely connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Is the checkout button experiment significant yet?
• › Which experiments are running in the web project?
• › Which flags are on in production?
• › What changed in the mobile project this week?
• › Pause the homepage hero experiment.

Questions

Is this an official Optimizely MCP server?

No. It is Memtro's MCP server with Memtro's Optimizely connector behind it: the connector reads Optimizely through its API with the credentials you connect on Dashboard → Connections, and Optimizely publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in Optimizely?

Only when you ask, and only what it has been granted: create an experiment (not started); edit an experiment's name, description, traffic, audiences or schedule, optionally publishing it; start an experiment, showing it to live visitors; pause a running experiment; resume a paused experiment; archive an experiment; create an audience; edit or archive an audience; start, pause or resume a personalization campaign; create a feature flag; turn a flag's rules on in one environment; turn a flag's rules off in one environment; archive flags by key; restore archived flags by key. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Analytics

• Google Analytics 4 MCP server
• Plausible Analytics MCP server
• Umami MCP server
• Google Search Console MCP server
• Segment MCP server
• App Store Connect MCP server