Is there an MCP server for Adobe Analytics?

Yes. Agent Planners runs a hosted MCP server at https://www.agentplanners.com/api/mcp that gives Claude, ChatGPT, Codex and other MCP clients 8 Adobe Analytics read tools and 4 write tools. Sign in with OAuth or an API key. Writes follow your approval settings; changes you can't take back wait for a person unless an admin exempts a key.

MCP

Adobe Analytics MCP server: reports and segments in Claude, Codex or Cursor

TL;DR

Adobe Analytics reaches Claude, Codex or Cursor through the Agent Planners MCP endpoint with 8 reads and 4 writes of its own: report suites, dimensions, metrics, ranked reports with breakdowns and segments, and approval-gated segments and calculated metrics.

What already exists for Adobe Analytics

Adobe hosts an Adobe Analytics MCP server (aa-mcp.adobe.io, on Experience League's list in November 2026). Each person signs in with their own Adobe ID, and what it can do follows that person's Analytics permissions.

Under any MCP layer sits the Analytics 2.0 API on analytics.adobe.io, called with the company id (globalCompanyId) and a credential whose product profile grants report suites.

What this setup adds

Adobe's hosted serverAgent Planners MCP
Sign-inEach person with their Adobe IDOne Server-to-Server credential, connected once on the Adobe Analytics card
Analytics toolsThe server's own8 reads and 4 writes, plus two shared connection reads
TotalsAs the API answersAdobe's period totals on every report, with a note to quote them (daily visits do not add up)
ChangesThe server's ownSegments and calculated metrics under the key's write mode; deletes and classification imports held for a person
Other platformsAdobe onlyEvery platform the workspace connected, on one endpoint

How to configure it

  1. 1In Adobe Developer Console add the Adobe Analytics API to a project with an OAuth Server-to-Server credential, and put the credential on the product profile it should use.
  2. 2On Integrations open the Adobe Analytics card, press Paste JSON (or Manual entry) and Test connection. The card requests only this product's scopes: openid, AdobeID, additional_info.projectedProductContext.
  3. 3Create a key in API & MCP: tools:read for reading, add tools:write for changes. To keep a key to Adobe, choose Only these platforms → Adobe.
  4. 4Add the server to your client with one of the snippets below, run describe_permissions, then adobe__get_connection and adobe__analytics_list_report_suites.
bash
# Claude Code
claude mcp add --transport http agent-planners https://www.agentplanners.com/api/mcp \
  --header "Authorization: Bearer $AGENTPLANNERS_API_KEY"

# Codex (signs in with OAuth in the browser)
codex mcp add agent-planners --url https://www.agentplanners.com/api/mcp

# Cursor (~/.cursor/mcp.json)
{ "mcpServers": { "agent-planners": { "url": "https://www.agentplanners.com/api/mcp", "headers": { "Authorization": "Bearer ${env:AGENTPLANNERS_API_KEY}" } } } }
Every client's setup, including OAuth sign-in without a key: MCP setup guide.

The Adobe Analytics tools

ToolKindWhat it does
analytics_list_companiesReadThe Analytics companies the credential reaches (GET /discovery/me): globalCompanyId, name, data center.
analytics_list_report_suitesReadReport suites (GET reportsuites/collections/suites): rsid, name, currency, timezone.
analytics_list_dimensionsReadDimensions of a report suite (GET dimensions?rsid=): id (variables/evar1, variables/page …), name, type.
analytics_list_metricsReadMetrics of a report suite (GET metrics?rsid=): id (metrics/visits, metrics/orders, metrics/event1 …), name, type.
analytics_list_segmentsReadSegments (GET segments): id (s…), name, rsid, owner, description; includeDefinition:true adds the definition.
analytics_list_calculated_metricsReadCalculated metrics (GET calculatedmetrics): id (cm…), name, rsid, polarity, precision, type.
analytics_reportReadAn Analytics ranked report (POST reports — a READ, changes nothing): { accountId?, rsid, startDate, endDate (YYYY-MM-DD, end inclusive, the report suite's time zone; default the last 7 complete days), dimension (varia…
analytics_getReadANY Analytics 2.0 API GET under the company (path relative to /api/{globalCompanyId}/: reportsuites/collections/suites[/{rsid}], dimensions, metrics, segments[/{id}], calculatedmetrics[/{id}], dateranges, projects, an…
analytics_create_segmentWriteCreate an Analytics segment (POST segments): { accountId?, name, rsid, definition (the segment's own JSON definition — container / func / pred as Adobe documents), description? }.
analytics_update_segmentWriteChange a segment's name, description or definition (PUT segments/{id}; the segment is read first): { accountId?, segmentId, name?, description?, definition? }.
analytics_create_calculated_metricWriteCreate a calculated metric (POST calculatedmetrics): { accountId?, name, rsid, definition (Adobe's formula JSON), description?, polarity? (positive | negative), precision?, type? (decimal | percent | currency | time) }.
analytics_mutateWriteAny other Analytics 2.0 write under the company (method POST | PUT | PATCH | DELETE, path relative to /api/{globalCompanyId}/, body = Adobe's own fields): segments, calculatedmetrics, dateranges, annotations, projects…
Each Adobe product is its own connection and its own MCP offering. A workspace that has connected Adobe Analytics lists and finds its tools; one that has not does not list them at all, and describe_tool on one of them says to connect the Adobe Analytics card. A key narrowed to the Adobe platform reaches only the Adobe products the workspace has connected; adobe__get_connection shows each one.

What waits for a person, whatever the key allows

A tools:write key's ordinary Analytics writes — a segment, a calculated metric, a date range — either run inside the workspace's guardrails or queue for an admin, depending on how the key was minted. These are held for a person on every key mode:

  • Any delete (DELETE through analytics_mutate).
  • A classification import — it overwrites report data.
Refused outright: data repair, users and credentials paths.

What we measured

  • No report suite in the credential's product profile → an empty suite list and 403 on a suite's reads; the tools name the suites the company's segments mention and the Admin Console steps. With Auto-include the same credential listed 14.
  • 30 daily rows of visits summed to 317,464; Adobe's 30-day total was 316,068.
  • Segments: 0 of the credential's own, 10 with shared ones (now the default).
More on Adobe Analytics: what the agent does · setup and every tool · MCP server · AI agent guide · playbook. The other Adobe products, each its own connection: Adobe Experience Platform · Adobe Target · Adobe Launch · All four together: Adobe.

Frequently asked questions

Is there an official Adobe Analytics MCP server?
Yes: Adobe hosts one (aa-mcp.adobe.io, on Experience League's list in November 2026), signed in with your Adobe ID. Agent Planners connects Adobe Analytics with your own Server-to-Server credential instead, beside your other platforms, with its approvals.
How do I connect Claude to Adobe Analytics?
Connect Adobe Analytics on its card on Integrations, create an Agent Planners key in API & MCP, add the MCP server to Claude, and call adobe__analytics_list_report_suites.
Why does my MCP client not see the Target or Launch tools?
Each Adobe product is its own connection: only the products this workspace connected are listed. Connect another product on its own card to add its tools.
Put a human-approved agent on your ad ops

Start free — 2,500 credits a month, no credit card. Reads run at once; by default every write waits for your approval.

Related reading