Is the checkout experiment winning yet, and which flags are on in production?
Optimizely runs your experiments and feature flags, and Memtro reads them so anyone can ask how a test is going. It reads projects, experiments with status, traffic, variations and metrics, and results with lift and statistical significance. It reads personalization campaigns, audiences, a project's change history, and flags with their state per environment. When asked and granted, it can create, edit, start, pause or archive experiments, manage audiences and campaigns, and turn flags on or off. Optimizely can tell Memtro when a flag changes, and a job can start on it. Who made each change is not returned.
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right Optimizely tool, reads what it needs, and answers with the source cited.
These tools change something in Optimizely. The model is told to use them only when you ask for exactly that, and an administrator grants each one by name to the teams that may use it; a personal connection acts only after its owner allows it.
Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in Optimizely, choosing the same kinds; the tab shows Optimizely's own steps. Each event can start a job, with the event as its input. Optimizely signs every delivery, and one whose signature does not check out is refused before it is read.
Events and webhooksOnce Optimizely is connected, these are the kinds of things people ask. Plain questions, no syntax. If the answer lives partly in another system you have connected, Memtro looks there too.
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.
Sarah Chen, growth lead at Acme Ltd, asked whether the new pricing page was winning. Memtro read the experiment's results and gave each variation's conversion rate, lift and significance, and how many visitors were still needed.
Dan Fletcher, engineering manager at Northwind, asked Memtro to turn off the new search flag in production after errors rose. Memtro turned its rules off there, and the SDKs fell back to the default straight away.
The teams that lean on Optimizely most, a question each of them asks, and what they connect beside it.
A record in Optimizely is rarely the whole story. Connect the systems that hold the rest and Memtro checks them together, then remembers what it learned so the next question starts further along.
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
CallRail
Fullstory
WhatConverts
Lightdash
Microsoft Clarity
Eppo
Holistics
MouseflowIn Optimizely, open Profile, then API Access, and generate a new token. Copy it straight away, as it is shown once.
In Memtro, open Connections → Optimizely and paste the Personal access token.
The token has your access in every project. An administrator grants the experiment, audience and flag actions by name. Events are turned on from the connection's Events tab, then added in each project's Webhooks settings.
Ask how the current checkout experiment is doing.
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The Optimizely tools are the same everywhere.
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. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.
A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.
Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.
Personal access token. In Optimizely, open Profile, then API Access, and generate a new token. Copy it straight away, as it is shown once.
Yes. Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in Optimizely, choosing the same kinds; the tab shows Optimizely's own steps. A job set to start "When an event arrives" then runs once for each matching event, with the event as its input. The run treats the event as outside data, never as instructions, and anything the run remembers is kept as unverified.
Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.
Sign in, add a provider key, create an API key and point your tools at Memtro. It takes about five minutes.
7-day free trial · card on file, nothing charged until it ends · cancel any time
