Agent Native

install

This skill sets up Mixpanel integration for a project by configuring one of three engines: Mixpanel MCP server, the headless Python SDK, or the user's own…

by MixpanelOfficial publisherNot reviewed

Identity proven by GitHub's verified organisation badge (mixpanel)

Agents
Claude Code
License
Apache-2.0 (permissive)
Pin
dfb891bpinned Oct 9, 2026
Context cost
about 1,200 tokenswhen the skill loads
GitHub stars
15on mixpanel/ai-plugins
Scripts
None found

What it does

“Set up Mixpanel for this project — set up an engine for the other Mixpanel skills to use (Mixpanel MCP server, mixpanel-headless Python SDK, or the user's own integration). Use whenever the user asks to set up, install, configure, or connect Mixpanel, switch Mixpanel engine or region, or when any Mixpanel skill finds no engine set up. Trigger phrases: "set up mixpanel", "configure mixpanel", "connect mixpanel", "install mixpanel", "switch mixpanel region", "use mixpanel headless", "change mixpanel engine", "add another engine". Do NOT use for adding Mixpanel tracking code to an application (use tracking-implementation) or for running analytics queries. Interactive — asks the user to choose; do not run in non-interactive sessions.”
The author's description, from SKILL.md

Generated summary from the skill's own text

This skill sets up Mixpanel integration for a project by configuring one of three engines: Mixpanel MCP server, the headless Python SDK, or the user's own integration. It guides the user through engine selection, authentication, and verification, then records preferences for future sessions.

When to use it

  • User asks to install, configure, or connect Mixpanel to their project
  • Need to switch Mixpanel region or change between multiple engines
  • Another Mixpanel skill detects no engine is configured
  • User wants to set up headless SDK or MCP server for analytics

When not to

  • Adding Mixpanel tracking code to an application source code
  • Running analytics queries or reports
  • Non-interactive sessions like CI pipelines without user input

Install

These commands install from the publisher's repository and need no account.

skills CLI

npx skills add mixpanel/ai-plugins --skill install

Installs from the repository's default branch, which may have changed since the commit we checked.

GitHub CLI 2.90 or later

gh skill install mixpanel/ai-plugins install@dfb891bb6f754b5dcab2131b7311e7ccb9bf8c3e

Installs the commit we checked (dfb891b).

Copy by hand

Download the folder at dfb891b and put it in .claude/skills/install for Claude Code and in .agents/skills/install for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.

Skill folder on GitHub at dfb891b

Copy for your agent

A prompt for any coding agent. It installs this skill from GitHub at the pinned commit, copies the files as they are, and runs nothing unless you ask.

Prompt

Install the agent skill "install" by Mixpanel.

Source, pinned to commit dfb891bb6f754b5dcab2131b7311e7ccb9bf8c3e:
https://github.com/mixpanel/ai-plugins/tree/dfb891bb6f754b5dcab2131b7311e7ccb9bf8c3e/plugins/mixpanel/skills/install

1. Download that folder at exactly this commit. Do not use the default branch.
2. Copy every file unchanged, keeping the folder structure, into .claude/skills/install/ for Claude Code and .agents/skills/install/ for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. Create the folders if they do not exist.
3. Do not edit the files, and do not run any script in the folder unless I ask you to.
4. When you are done, list the files you copied and where you put them.

Listing and checks: https://www.agentnative.dev/skills/mixpanel/install

SKILL.md at the pin

---
name: install
description: >
  Set up Mixpanel for this project — set up an engine for the
  other Mixpanel skills to use (Mixpanel MCP server, mixpanel-headless
  Python SDK, or the user's own integration). Use whenever the user asks to set up, install,
  configure, or connect Mixpanel, switch Mixpanel engine or region, or
  when any Mixpanel skill finds no engine set up. Trigger phrases:
  "set up mixpanel", "configure mixpanel", "connect mixpanel", "install
  mixpanel", "switch mixpanel region", "use mixpanel headless", "change
  mixpanel engine", "add another engine". Do NOT use for adding Mixpanel tracking code to an
  application (use tracking-implementation) or for running analytics
  queries. Interactive — asks the user to choose; do not run in
  non-interactive sessions.
compatibility: "Works in any project. Sets up an engine for the other Mixpanel skills to use."
metadata:
  engine: none
---

# Mixpanel Install

> **No engine required** — this skill is what _sets engines up_.

Set up how this project talks to Mixpanel. Engines can coexist — a project may have more than one; this skill adds one and can record which to prefer. How the other skills resolve an engine (the precedence ladder, preference locations) lives in [`../../ENGINE.md`](../../ENGINE.md) — read it before starting.

This skill is **interactive**: it asks the user to choose. If the session can't ask questions (non-interactive/CI run), stop and tell the user to run `/mixpanel:install` in an interactive session instead.

The bundled references defer to the live Mixpanel docs they cite — on any conflict, trust the live page.

---

## Step 0 — Detect existing state

1. Mixpanel MCP tools listed in this session → tell the user the MCP engine is already set up (name the region if the server URL is visible) and ask: keep it, switch, or add another engine alongside it. Keep → stop here. Add another → Step 1, skipping the MCP option.
2. Otherwise run `mp --version` (one command). Success → the headless SDK is installed: ask **"MCP is not available — how should we proceed?"** with two options: (a) install MCP, (b) use headless. Installing MCP → Step 2a. Using headless → verify auth (Step 2b's verify) and jump to Step 3, including the preference note.
3. Neither → continue to Step 1.

## Step 1 — Pick an engine to set up

Ask with the client's multiple-choice question UI (one question, three options):

1. **Mixpanel MCP server** _(recommended for interactive analytics)_ — a remote server exposing Mixpanel tools; OAuth login; best when a person is in the loop.
2. **Headless SDK** — the [`mixpanel-headless`](https://docs.mixpanel.com/docs/mixpanel-headless) Python package; the full Mixpanel platform as a Python object; best for scripted, CI, or coding-agent workflows.
3. **I have my own integration** — the user already has their own way to reach Mixpanel.

## Step 2a — MCP path

1. Ask the **region**: US / EU / India (URL map in [`references/mcp-setup.md`](references/mcp-setup.md)). If the user is unsure, their Mixpanel URL tells them (`eu.mixpanel.com` → EU, `in.mixpanel.com` → India; otherwise US).
2. Register the server with the current client following [`references/mcp-setup.md`](references/mcp-setup.md) — the standard recipe per client, plus the fallback transport. No further questions; just run the standard installation.
3. **Verify**: list the server's tools. If authentication is pending, have the user complete the client's OAuth flow (per the reference), then re-check.

## Step 2b — Headless path

1. Run the one-line availability check from [`references/headless-setup.md`](references/headless-setup.md); install the SDK if it's missing.
2. Authenticate per the reference: run `mp login` yourself — it picks the right flow from the environment (opens the user's browser for OAuth, or uses service-account / bearer-token env vars when set). Don't ask the user to run it; just tell them to complete the login in the browser. Credentials never go in tracked files.
3. **Verify** with `mp account test`. On failure, surface the exact error and fix auth before wrapping up.

## Step 2c — Own-integration path

Nothing to install and nothing to interrogate — assume the user provides the context needed to act (in the conversation, or in the project's agent instructions like `CLAUDE.md`). Acknowledge, skip Step 3's verification, but do offer Step 3's preference note so future sessions pick the integration up.

## Step 3 — Confirm and wrap up

Always complete all three points — even when this skill was triggered mid-task by another skill, finish them **before** resuming the original request.

1. Re-run the verification for the chosen path and summarize what is now available — the engine just set up (engine, region, where it lives — e.g. the repo's `.mcp.json` or the Python environment) plus any engine the project already had.
2. **Preference note** — if the user ended on an engine other than MCP, or more than one engine is now available, ask: "Want me to leave a note of this preference so you don't have to specify it each time?" If yes, ask where — project level or user level — and append one plain line (e.g. `For Mixpanel, use the headless SDK.`) to the matching file per [`../../ENGINE.md`](../../ENGINE.md): project → the project's `CLAUDE.md` (or `AGENTS.md` in Cursor); user → `~/.claude/CLAUDE.md` (in Cursor, point them to Settings → Rules). Never write without asking. A note is only needed to prefer something other than MCP.
3. Suggest a next step: try a skill like `analyze-report`, `deep-research`, or `tracking-implementation`. If the headless engine was installed, also mention the SDK's companion plugin (see [`references/headless-setup.md`](references/headless-setup.md)).

From mixpanel/ai-plugins at dfb891b, shown as plain text under its license: Apache-2.0 (permissive). The license text is in LICENSE.

View SKILL.md on GitHub at dfb891b

Files

3 files, 12 KB in total. Each file has a sha256 in the pin's manifest; the first 12 characters are shown.

Skill folder on GitHub at dfb891b
All 19 skills from Mixpanel

More from Mixpanel

  • auth

    Manages Mixpanel credentials, authentication, and account configuration for the mixpanel_headless library and mp CLI.

    Contains scripts

    Claude Code · about 3K tokens · pinned 6 days ago

  • dashboard-expert

    This skill analyzes, builds, modifies, and adds explainer text to Mixpanel dashboards using the mixpanel_headless Python library.

    Claude Code · about 2.8K tokens · pinned 6 days ago

  • mixpanelyst

    The skill analyzes Mixpanel data by running Python code with the mixpanel_headless library and pandas.

    Claude Code · about 4.3K tokens · pinned 6 days ago

  • session-replay

    This skill retrieves and analyzes Mixpanel session replay recordings to show what a specific user did on screen, action by action.

    Claude Code · about 2.6K tokens · pinned 6 days ago

  • setup

    This skill creates an isolated Python environment for Mixpanel analysis, installs mixpanel-headless version 0.4.0 or newer along with data science libraries…

    Contains scripts

    Claude Code · about 1.1K tokens · pinned 6 days ago

  • analyze-report

    Interprets existing Mixpanel reports and charts by re-running queries, identifying patterns, and summarizing what the data shows and what stands out.

    Claude Code, Codex, Cursor +4 · about 1.9K tokens · pinned 6 days ago

  • pinecone:quickstart

    This skill guides new developers through Pinecone with an interactive quickstart offering two paths: a Database path that creates an integrated index, uploads…

    by PineconeOfficial publisher

    Contains scripts

    Claude Code · about 2.2K tokens · pinned 6 days ago

  • learn-mcp

    This skill guides users through interactive modules on how to use the Mixpanel MCP server for product analytics.

    by MixpanelOfficial publisher

    Claude Code, Codex, Cursor +4 · about 3.6K tokens · pinned 6 days ago

  • pinecone-quickstart

    This skill guides developers through their first Pinecone experience by choosing between two paths: a Database path that creates a vector search index…

    by PineconeOfficial publisher

    Contains scripts

    Cursor · about 2.2K tokens · pinned 6 days ago

  • algolia-automation

    This skill automates Algolia search and indexing operations through the Rube MCP interface.

    by ComposioOfficial publisher

    Claude Code, Codex, Cursor +3 · about 690 tokens · pinned 6 days ago

  • building-a-dashboard

    This skill constructs or modifies PostHog dashboards by selecting and arranging insight tiles.

    by PostHogOfficial publisher

    Claude Code, Codex, Cursor +4 · about 2.1K tokens · pinned 6 days ago

  • signals-scout-product-analytics

    This skill monitors product analytics behavioral flows (funnels, retention, lifecycle, stickiness, paths) for regressions in conversion or retention rates…

    by PostHogOfficial publisher

    Claude Code, Codex, Cursor +3 · about 6.2K tokens · pinned 6 days ago

Open roles at Mixpanel

3 open roles on the jobs board

The skills listed here are third-party software, collected from public GitHub repositories as a catalog of recent skills and provided as is. Agent Native does not write or control them and is not responsible for any harm, loss or damage caused by installing or using them. Labels and checks describe what we saw at one commit and are not a guarantee. Read a skill's files before you run it. Terms of use