Export experiment results to Confluence

This guide exports Kameleoon experiment results into a Confluence page from a single Claude conversation. It requires no script, no stored API credentials to manage, and no code. Claude connects to Kameleoon’s and Atlassian’s own remote MCP (Model Context Protocol) servers, and performs the export when you ask it to in plain language. Both MCP servers expose more than read-only tools. The Kameleoon MCP server can also start, pause, stop, or delete experiments and feature flags. The Atlassian MCP server reaches beyond Confluence: it also covers Jira and Bitbucket, and includes tools to create and update Confluence pages and comments. Whichever Jira and Bitbucket access you grant it during authorization, that access travels with the connection, not just the Confluence page this export touches. Review any action either connector proposes before approving it.

Prerequisites

A Kameleoon account with access to the experiment being exported
A Confluence space you can edit, on a site connected to an Atlassian account
A Claude account with Connectors available

Step 1: Add the connectors

Two paths get you to the same result. Use whichever matches how you already work with Claude.

Connect via a coding tool

Connect the Kameleoon MCP server to your assistant. See Kameleoon MCP server for the registration and authentication steps for Claude Code, Cursor, Codex, and Antigravity.
Connect the Atlassian MCP server. In Claude Code, run:
claude mcp add --transport http atlassian https://mcp.atlassian.com/v1/mcp/authv2
then run /mcp inside Claude Code and authorize through the browser. See Set up clients for Cursor, VS Code, and other client-specific setup steps.
Start a new conversation so the assistant loads the tools from both servers.

Connect via the Claude app

Open your account’s Connectors settings in the Claude app (Customize > Connectors).
Add the Kameleoon connector and authorize it through OAuth. If it isn’t listed in the connector directory, add it as a custom connector using the URL https://mcp.kameleoon.com/mcp, then authorize it.
Add the Atlassian connector. It’s available directly in the connector directory, so search for Atlassian and add it from there, then authorize it and choose which Jira, Confluence, and Bitbucket access to grant. If your organization’s directory doesn’t list it, add it as a custom connector using the URL https://mcp.atlassian.com/v1/mcp/authv2 instead.
Start a new conversation, then use the plus button in the chat to enable both connectors for it. Connectors enabled in one conversation don’t carry over to a new one.

On a Team or Enterprise plan, adding a connector yourself may not be available. A Request button instead of Add means your organization restricts who can add connectors, and an administrator needs to approve the request first.

Step 2: Prepare the Confluence page

Confluence has no database schema to build in advance the way Notion and Airtable do. You only need a space you can write to and its numeric space ID, which isn’t visible in the space’s URL the way a Notion database ID or an Airtable base ID is. Find it under Space settings in Confluence, or ask your assistant to look it up once both connectors are active. Give the assistant a page title (or let it default to Experiments) and the space ID in the prompt below. On the first run, the assistant creates the page with this table. On every later run, it finds the existing page and updates it.

Confluence column	Source	Transformation
Experiment Name	`experiment.name`	Direct. Used as the upsert key.
Status	`experiment.status`	Mapped, case-insensitively: `active → Running`, `draft`/`planned → Implementing`, `stopped`/`diverted → Completed`, `paused → Defunct`.
Start date	`experiment.dateStarted`	Date-time truncated to an ISO date (`YYYY-MM-DD`).
End date	`experiment.dateEnded`	Date-time truncated to an ISO date.
Notes	`experiment.description`	Direct.
Actual	best variation `improvementRate`	Direct (measured uplift, %).
Probability	best variation Bayesian success probability	Bucketed: `≥95 → 80% - High`, `≥80 → 50% - Medium`, else `20% - Low`.
Result	Bayesian success probability + `improvementRate`	probability `≥ 95` and uplift > 0 → `Success`; `≥ 95` and uplift < 0 → `Failure`; otherwise `Inconclusive`.

Step 3: Prompt the assistant

The Kameleoon MCP tool returns its own precomputed winner alongside the raw results, but it judges significance against this account’s configured reliability threshold, not this schema’s fixed thresholds. The prompt below works around that gap by asking for the raw numbers instead. Replace the experiment ID and the Confluence space ID with your own before sending it.

Using the Kameleoon MCP tools, get the results for experiment XXXXXX. In the
returned data, skip the "_reference" variation and read each remaining
variation's raw reliability and improvement rate for the experiment's
primary goal (not the tool's own winner summary, which judges significance
against this account's configured reliability threshold rather than the
fixed thresholds below). If the same reliability and improvement numbers
appear more than once under different field names, use the plain
percent-scale value tied to that goal, not a 0-1 fraction of it. Pick the
variation with the highest improvement rate as the best performer.

Then, using the Atlassian MCP tools, find a Confluence page titled
"Experiments" in space ID XXXXXXX. If it exists, read its table and keep
every row you didn't just touch exactly as it is: updating this page
replaces its whole body, so reproduce every other row unchanged and only
add or update the one row for this experiment. If the page doesn't exist,
create it with a full-width table using the header below (a full-width
layout keeps the header labels from wrapping mid-word across 8 columns).

Upsert a row matching on "Experiment Name" against the experiment's name:
update the matching row if one exists, or add a new one if not. Set the
columns as follows:
- Experiment Name: the experiment's name
- Status: map a status of active to Running, draft to Implementing, stopped
  to Completed, and paused to Defunct (case-insensitive; confirm the exact
  tokens your account returns)
- Start date / End date: the experiment's start and end dates, formatted as
  YYYY-MM-DD
- Notes: the experiment's description
- Actual: the best variation's improvement rate
- Probability: 80% - High if the reliability is 95 or above, 50% - Medium if
  it's 80 or above, otherwise 20% - Low
- Result: Success if the reliability is 95 or above with a positive
  improvement, Failure if the reliability is 95 or above with a negative
  improvement, otherwise Inconclusive

Don't add or change any columns outside this list, and don't touch any
content on the page outside the table.

A few things to know if you adjust this prompt:

Don’t ask for the tool’s own winner summary instead of the raw numbers. It judges significance against your account’s own threshold, not this schema’s fixed 95%/80% buckets.
Keep each experiment’s name unique on the page. The upsert matches on that column, and may update the wrong row, or create a duplicate, if more than one row shares that name.
The instruction to preserve every other row matters more here than in the Notion or Airtable versions of this prompt. Confluence has no per-row write, so an update that forgets an existing row deletes it from the page.
Kameleoon’s results can repeat the same reliability and improvement numbers elsewhere in the response under a different name or scale. The prompt already specifies which one to use.

Step 4: Automate it

To keep the page current without repeating the export manually, set up a recurring scheduled task with the same prompt. Each scheduled run starts a new session with no memory of prior runs, so include the experiment ID (or a request to list all running and recently completed experiments) and the space ID every time. A daily or weekly cadence matches how often most teams need this export.

Notes

An AI assistant interprets this mapping fresh on every run rather than executing a fixed function. Check the page it writes against the values in Kameleoon, particularly the first few runs.
Confluence stores a page’s content as plain HTML, not typed properties the way Notion and Airtable do. Nothing validates a value against a column’s expected type before it’s written, so a bad mapping just lands as the wrong text in a cell instead of raising an error. Spot-check the table more carefully after the first run than you would a typed Notion or Airtable field.
The Atlassian connector’s write access, combined with its reach into Jira and Bitbucket beyond Confluence, means proposed actions deserve a closer look than just the table edits. Review any Kameleoon-side action separately too, since that connector can affect running experiments and feature flags.
For an export you run on a schedule or that must reproduce identically every time, a fixed script is a more predictable alternative to an AI assistant interpreting the same prompt on each run.

​Prerequisites

​Step 1: Add the connectors

​Connect via a coding tool

​Connect via the Claude app

​Step 2: Prepare the Confluence page

​Step 3: Prompt the assistant

​Step 4: Automate it

​Notes