Connect Mixpanel MCP to your AI Agents

Conversational analytics on your Mixpanel events, funnels, and retention.

Example prompts

Everything you can ask your agent about your Mixpanel data.

Ask in plain English — Sequel translates it into a Mixpanel query, runs it, and your agent hands back the answer. No SQL, no dashboards, no CSV exports.

Connect to your agent
What you can ask your agent
  • How many users completed 'Subscription Started' last week, by plan?
  • Show me the signup → activation funnel conversion for the past 30 days.
  • Compare DAU for iOS vs Android over the last 60 days.
  • What's the average value of 'Order Placed' broken down by country?
  • Which step in funnel 4321 has the biggest drop-off?

Sequel CLI

Install Mixpanel MCP to your agent

The Sequel CLI installs Mixpanel into your coding agent in one step: it signs you in, provisions a scoped API key, and writes the right config to the right place. No keys to copy by hand.

Prefer npm? npm install -g @sequelsh/cli@latest works too.

  1. 1

    Install the Sequel CLI

    One line installs the latest CLI with whatever package manager you have.

    curl -fsSL https://sequel.sh/install | sh
  2. 2

    Sign in

    Authenticate in your browser and pick an organization.

    sequel login
  3. 3

    Install into your agent

    Run install and pick an agent from the list, or target one directly by its slug.

    sequel install
    • Claude Code
      sequel install claude-code
    • Claude
      sequel install claude
    • Cursor
      sequel install cursor
    • VS Code
      sequel install vscode
    • Windsurf
      sequel install windsurf
    • Zed
      sequel install zed
    • Codex
      sequel install codex
    • OpenClaw
      sequel install openclaw
    • Hermes
      sequel install hermes

Skills

Playbooks that run on Mixpanel

Step-by-step procedures your agent follows on your Mixpanel data, instead of improvising a query. Browse all skills

Can we trust our product metrics

Two tools show different numbers for the same metric and nobody knows which to trust. Reconciles the disputed figure against a system of record such as Stripe or the application database, then finds the instrumentation defect behind it: duplicate and near-duplicate event names, events that silently stopped firing, missing properties and system-fired events counted as user actions.

Feature launch impact readout

Measure whether a feature we just shipped actually landed: adoption against an eligible denominator of users who could even reach it, breadth, depth and sustained use after the release, a pre and post impact estimate with or without a control group, and a keep, iterate or sunset call.

Find our activation moment

Discover the aha moment: which early in-product action, taken in the first session or first week, most separates users who stay from users who vanish, how to score candidate first actions by retention lift, and what share of new signups ever reach it. Covers activation metric definition, activation rate and time to value.

Is our retention curve flattening

Do users keep coming back, or do they drift away after the first week or month? Builds the cohort retention curve for your key action at its natural frequency and judges whether it flattens to a plateau or decays to zero, covering N-day versus unbounded retention, repeat usage, leaky bucket diagnosis and how the plateau compares to published benchmarks.

Product qualified lead handoff list

Build a ranked list of self-serve accounts sales should contact this week: join product usage to CRM and billing, exclude accounts already paying or already in an open deal, and attach the named trigger event and the evidence for each row.

Read out an A/B test honestly

Is this A/B test result real, or statistically significant only by accident? Runs the validity gates before reporting anything (sample ratio mismatch, peeking, power and minimum detectable effect, novelty), then gives absolute and relative lift with a confidence interval, a ship or kill call, and the winning variant extended through to revenue instead of stopping at the proxy metric.

Frequently asked questions

How do I connect Mixpanel to Sequel?

Create a Service Account in Mixpanel under Organization Settings → Service Accounts with at least project-level read access. Copy the username (looks like `my-sa.abc123.mp-service-account`) and the secret shown once at creation, then paste both into Sequel along with your numeric project ID and region.

Why a Service Account instead of an OAuth login?

Service Accounts are the long-lived, scope-limited credential Mixpanel recommends for programmatic access. They can be revoked or rotated independently of any user's login, and they don't break when a teammate leaves. Sequel uses HTTP Basic auth with the service account username and secret.

Which Mixpanel region should I pick?

Use the region your project was created in. US projects use mixpanel.com / data.mixpanel.com. EU projects use eu.mixpanel.com / data-eu.mixpanel.com. The region toggle simply chooses which API host Sequel calls — pick the wrong one and you'll see 401s or empty results.

Can Sequel modify data in Mixpanel?

No. The connector only ships read-only tools: segmentation, funnel report, list funnels, list events, list event properties, and raw event export. The agent has no write surface, so it cannot create reports, funnels, or cohorts.

Does Sequel support saved funnels?

Yes. List saved funnels by project, then run any of them with a custom date range or breakdown — no need to rebuild the steps. You can also describe a new step sequence in English and Sequel will compute conversion on the fly.

Will raw event exports hit Mixpanel's rate limits?

Raw exports are slower and more expensive than aggregated queries. Sequel keeps them bounded by date and event name, and queues large fan-outs to stay within Mixpanel's per-project concurrency limits. For very wide exports, expect noticeable waits.

How fresh is the data?

Mixpanel ingests events with low latency, and Sequel hits the JQL/Query API live — so data freshness matches what Mixpanel shows in the UI. Raw exports typically lag a few minutes behind ingest.

Can I combine Mixpanel with my warehouse or revenue data?

Yes. A common setup is Mixpanel + Stripe + Postgres — ask one question that joins product behavior, revenue, and account metadata, and Sequel reasons across all three sources at once.

Start analyzing your Mixpanel data today

Connect in minutes — then ask across all your data, not just one source.