Integration · Research and analyze

Fullstory and Meteor AI agents

Connect Fullstory to a Met: find where users get frustrated, build funnels and metrics, and read real sessions with the official MCP.

Book a demo See every integration

How does Fullstory connect to Meteor?

With the Fullstory ready-made connector, your Met queries how users behave on your site or app with the account the admin authorizes. It finds the strongest frustration signals on a page or a funnel, measures how many users each one affects, builds segments, metrics, funnels and journeys from a plain language description, and goes down to real sessions to see what happened: the events in order, a screenshot of the moment or the accessibility tree. It is the official Fullstory server, in beta: it needs StoryAI enabled in your org, and usage runs on your Fullstory plan.

What it automates

What a Meteor Met does with Fullstory

Learn where users get stuck

You ask the Met what is frustrating users at checkout. The Met finds the page, looks for the strongest frustration signals there and tells you how many users each one affects and whether it is rising against the previous period.

See what the user saw

With a session link, the Met opens it, reads its events in order and takes the screenshot of the moment of the error, so it can tell you which button was tapped and which message appeared. When it is done, it closes the session.

Measure a funnel without building the report

You describe the steps, for example view product, add to cart and pay, and the Met builds the funnel, runs it and gives you each step's conversion. If you want the reason, it brings sessions from people who dropped off at a step.

Answer a question with a metric

"How many iPhone users used search this week?" The Met turns the question into a metric, runs it and leaves you the link to review it in Fullstory.

33 tools

Everything your Met can ask Fullstory for

Each one is a real action in the connector. You can switch off the ones you would rather it never used.

Tool	What it does
`discover_org_context`	Searches your org's pages, named elements, events and variables to get their IDs.
`get_pages`	Lists page definitions, both the ones your team created and detected ones, with an optional filter.
`build_segment`	Turns a plain language description into a segment, saves it and returns its link.
`update_segment`	Refines an existing segment with an instruction and saves the new version under another ID.
`get_segment`	Looks up saved segments by ID or name and describes their definition.
`build_metric`	Turns a question into a single number, a ranking or a trend metric.
`update_metric`	Refines an existing metric, such as its aggregation or a filter, and returns a new ID.
`compute_metric`	Runs a metric on your data, with an optional segment and period, and returns the result.
`get_metric`	Looks up saved metrics by ID or name and describes their definition.
`get_view_counts`	Says how many times up to 10 saved segments or metrics were viewed, up to 90 days back.
`build_funnel`	Turns a sequence of actions described in words into an ordered funnel and saves it.
`update_funnel`	Refines a funnel's steps, conversion window, period or aggregation and returns a new one.
`get_funnel`	Looks up saved funnels by ID or name, with their step count and description.
`compute_funnel`	Runs a funnel and returns the count at each step and the conversion between steps.
`get_funnel_sessions`	Brings the sessions that completed a funnel step or the ones that dropped off there.
`build_journey`	Builds a journey around an anchor event, looking before or after it, and saves it.
`get_journey`	Looks up saved journeys by ID or name and describes them in plain language.
`compute_journey`	Runs a journey and returns the most common paths and the share at each step.
`get_journey_sessions`	Brings the sessions that went through a journey's anchor event or a specific path.
`update_journey`	Refines a saved journey's period, direction or segment and returns a new one.
`get_sessions`	Brings the sessions that match a metric or a segment, with their replay link.
`get_session_events`	Returns every event of a session in order, such as clicks, navigation, errors and network.
`session_open`	Opens a session for visual inspection and returns a summary of its events.
`session_screenshot`	Renders a screenshot of what the user saw on a page at a moment of the open session.
`session_get_a11y_tree`	Returns the page's accessibility tree at a moment of the open session.
`session_diff`	Compares the accessibility tree between two moments to see what changed on the page.
`session_close`	Closes the open session and frees its resources; it must be called when done.
`discover_groups`	Finds the strongest frustration signals in your product, a page, a segment or a funnel.
`get_opportunity_stats`	Computes the affected users, breakdowns and rate of change for a frustration signal.
`get_sessions_for_opportunity`	Brings real sessions where a specific frustration signal was observed.
`get_opportunities`	Brings the opportunities StoryAI already ranked, with a window from 24 hours to 30 days.
`get_opportunity`	Returns the full record of a StoryAI opportunity by its identifier.
`get_managed_funnels`	Lists the funnels Fullstory manages for your org, with their ID and name.

The limits

What the Fullstory integration does NOT do

We would rather say it here than have you find out during rollout.

It does not ask for confirmation in the chat before saving a new segment, metric, funnel or journey. The admin decides in Connections which actions stay on.
It does not create dashboards, change privacy rules or manage users, billing or settings: Fullstory documents that its MCP does not do operations of that kind.
It is not for bulk exports: groupings are capped at 50 rows, and session tools bring 10 by default and 50 at most.
It does not work across orgs. Each connection sees a single Fullstory org; if you manage several, you need one connection for each.
It does not work without StoryAI. If your org has it turned off, the connection completes but no tools appear.
It does not bring heatmaps or dashboards: Fullstory says the MCP focuses on metrics, segments and session replay.

Before you start

What you need to connect Fullstory

1

Fullstory with StoryAI enabled

An admin of your Fullstory account checks or enables StoryAI in Settings, StoryAI Features. The opportunities StoryAI already ranked also need the StoryAI Premium plan.
2

Connect it in Met

In Met, open Connections → FullStory → Connect. A Fullstory window opens so you can sign in and authorize Met. Then you choose which actions stay on.

Frequently asked questions

Questions about Fullstory and Meteor

It can open and read it: it brings the events in order, takes screenshots of a moment and compares the page between two moments. It does not play the full video the way you see it in Fullstory.

It only saves definitions it builds itself, such as segments, metrics, funnels and journeys; when it refines one, Fullstory saves a new version and leaves the original as it was. It does not touch privacy, users or billing.

Almost always because StoryAI is turned off in your org. A Fullstory admin enables it in Settings, StoryAI Features, and then you connect again.

You do, with your own Fullstory plan. Meteor does not resell Fullstory: the connector uses your company's account and follows the limits Fullstory sets for that account.

Step by step