Mixpanel Analytics connector for AI agents

Tools your analytics agent reaches for on Mixpanel Analytics, scoped per tenant.

CALL ANY TOOL

Product analytics across the Query API, Lexicon schemas, and warehouse connectors: segmentation, funnels, retention, Insights, profiles, and custom JQL.

mixpanelanalytics_activity_stream

Activity stream

Get the raw event stream (activity feed) for one or more specific users over a date range — every event each user did, in order. Useful for inspecting an individual user's journey rather than aggregate analytics. Rate limited to 60 queries/hour and 5 concurrent queries.

Parameters

Name

Type

Required

Description

distinct_ids

array

Required

The user distinct_ids to fetch activity for, as a JSON array of strings.

from_date

string

Required

Start date of the query range, in yyyy-mm-dd format (inclusive).

project_id

integer

Required

Your Mixpanel numeric Project ID (Project Settings > Overview). Required when authenticating with a Service Account; not needed with a legacy Project Secret.

to_date

string

Required

End date of the query range, in yyyy-mm-dd format (inclusive).

workspace_id

integer

Optional

The numeric ID of the Mixpanel workspace to query, for projects with multiple workspaces. Omit to use the project's default workspace.

mixpanelanalytics_annotation_create

Create annotation

mixpanelanalytics_annotation_delete

Delete annotation

mixpanelanalytics_annotation_get

Get annotation

mixpanelanalytics_annotation_tag_create

Create annotation tag

mixpanelanalytics_annotation_tags_list

List annotation tags

mixpanelanalytics_annotation_update

Update annotation

mixpanelanalytics_annotations_list

List annotations

mixpanelanalytics_event_properties

Event properties

mixpanelanalytics_event_top_property_values

Event top property values

mixpanelanalytics_events_query

Query events

mixpanelanalytics_events_top_today

Events top today

mixpanelanalytics_feature_flag_definitions_get

Get feature flag definitions

mixpanelanalytics_feature_flag_delete

Delete feature flag

mixpanelanalytics_feature_flag_update

Update feature flag

Replace an existing feature flag's full configuration by id — this is a full update, not a partial patch, so provide the complete desired 'name', 'key', 'tags', 'context', 'serving_method', and 'ruleset' (not just the fields you're changing). Use 'mixpanelanalytics_feature_flag_get' first to see the flag's current configuration. Use 'mixpanelanalytics_feature_flags_list' to find the flag_id.

Parameters

Name

Type

Required

Description

context

string

Required

A short label (1-255 characters) identifying the context this flag belongs to. Mixpanel's API reference does not fully define the allowed values or meaning beyond the length constraint — check the flag's current value (via 'mixpanelanalytics_feature_flag_get') for the convention your team uses.

flag_id

string

Required

The id of the feature flag to update. Find it via 'mixpanelanalytics_feature_flags_list'.

key

string

Required

The unique programmatic key SDKs use to reference this flag, e.g. "new-checkout-flow".

name

string

Required

The human-readable display name for this flag, shown in the Mixpanel UI.

project_id

string

Required

Your numeric Mixpanel Project ID, required to authenticate Service Account requests. Find it in Mixpanel under Project Settings > Overview.

ruleset

object

Required

A JSON object defining this flag's variants and rollout rules. Shape: {"variants": [{"key": "control", "value": true, "is_control": true, "split": 0.5}, {"key": "treatment", "value": true, "is_control": false, "split": 0.5}] (max 11 variants), "rollout": [{"rollout_percentage": 1.0, "variant_splits": {"control": 0.5, "treatment": 0.5}}]}. This replaces the flag's entire ruleset — fetch the current one with 'mixpanelanalytics_feature_flag_get' and modify it rather than writing a new one from scratch, since Mixpanel supports additional optional targeting fields (cohort-based or runtime-event-based triggers) not fully reproduced here.

serving_method

string

Required

How this flag is served: 'client' (evaluated in client SDKs), 'server' (evaluated in server SDKs), 'remote_or_local' (either), or 'remote_only' (must be evaluated via a live call to Mixpanel, not cached locally).

tags

array

Required

Tags to categorize this flag, as a JSON array of strings.

workspace_id

string

Required

The numeric ID of the Mixpanel workspace this flag belongs to.

data_group_id

string

Optional

The group key this flag is evaluated against for group-level (rather than user-level) targeting, if this flag uses Group Analytics. Omit for user-level flags.

description

string

Optional

A human-readable description of what this flag controls.

experiment_id

string

Optional

The UUID of a Mixpanel experiment to link this flag to, if it powers an A/B experiment. Omit for a plain rollout/release flag with no linked experiment.

hash_salt

string

Optional

A 32-character string used to deterministically hash users into variants for this flag. Omit to leave the current salt unchanged.

is_experiment_active

boolean

Optional

Whether the linked experiment (see 'experiment_id') is actively running. Irrelevant if this flag has no linked experiment.

reset_hash_salt

boolean

Optional

If true, generate a new random hash salt for this flag instead of using 'hash_salt', re-randomizing which variant every user is bucketed into.

status

string

Optional

The flag's status: 'enabled' to serve it, 'disabled' to turn it off, or 'archived' to retire it.

mixpanelanalytics_feature_flag_variant_assignments_get

Get feature flag variant assignments

mixpanelanalytics_feature_flags_list

List feature flags

mixpanelanalytics_funnels_list_saved

Funnels list saved

mixpanelanalytics_insights_query

Query insights

mixpanelanalytics_profiles_query

Query profiles

mixpanelanalytics_retention_query

Query retention

mixpanelanalytics_schema_get

Get schema

mixpanelanalytics_schema_upload

Upload schema

mixpanelanalytics_schemas_delete_all

Schemas delete all

mixpanelanalytics_schemas_delete_by_entity

Schemas delete by entity

mixpanelanalytics_schemas_list_by_entity

Schemas list by entity

mixpanelanalytics_segmentation_numeric_query

Query segmentation numeric

mixpanelanalytics_warehouse_import_create_event_stream

Warehouse import create event stream

Create a new warehouse import that streams warehouse rows into Mixpanel as events. Maps a warehouse table to Mixpanel's Events dataset using 'table_params' to select the source table/columns and 'time_column_name' (plus 'event_name' or 'event_column_name') to derive each event's timestamp and name. Requires a warehouse source to already be configured in Mixpanel (Lexicon > Warehouse Sources) — pass its id as 'warehouse_source_id'. Requires Warehouse Connectors to be enabled on your Mixpanel plan; without it, calls to this endpoint return a 403 Forbidden regardless of input.

Parameters

Name

Type

Required

Description

project_id

string

Required

Your numeric Mixpanel Project ID, required to authenticate Service Account requests. Find it in Mixpanel under Project Settings > Overview.

sync_mode

string

Required

How this import stays in sync with the warehouse table over time, e.g. a full re-sync mode or an incremental mode based on a time column. Mixpanel's Warehouse Connectors API does not fully document every valid value for this import type in one place — the sibling lookup-table import accepts 'full_sync' or 'one_time', but event/people/group streaming imports may support additional incremental modes. Check your warehouse source configuration in the Mixpanel dashboard, or current Mixpanel documentation, to confirm the exact values accepted before relying on a value here.

table_params

object

Required

JSON object describing which warehouse table (and columns) this import reads from, e.g. the database, schema, and table name. The exact shape depends on your warehouse type (Snowflake, BigQuery, Databricks, or Redshift) and is passed through to Mixpanel as-is — check an existing import's configuration (via 'mixpanelanalytics_warehouse_import_get') or current Mixpanel documentation for the precise fields your warehouse type expects. Example: {"schema": "public", "table_name": "events"}.

time_column_name

string

Required

The name of the warehouse column containing each row's event timestamp. Mixpanel uses this column to set the imported event's time.

warehouse_source_id

integer

Required

The numeric ID of the warehouse source (the connected Snowflake, BigQuery, Databricks, or Redshift warehouse) this import reads from. Warehouse sources are configured separately in Mixpanel under Lexicon > Warehouse Sources.

company_column_name

string

Optional

The name of the warehouse column containing a company/account identifier to associate with each imported event, for B2B group analytics.

databricks_params

object

Optional

Databricks-specific cluster configuration for this import, passed through to Mixpanel as-is. Only relevant when the underlying warehouse source is Databricks — omit for Snowflake, BigQuery, or Redshift sources. Check current Mixpanel documentation or an existing Databricks import (via 'mixpanelanalytics_warehouse_import_get') for the exact fields expected, such as a cluster id or SQL warehouse id.

device_column_name

string

Optional

The name of the warehouse column containing a device identifier to associate with each imported event.

event_column_name

string

Optional

The name of the warehouse column whose value supplies each row's event name, when different rows should become different event types. Provide this or a fixed 'event_name'.

event_name

string

Optional

A fixed event name to apply to every row imported from this table. Provide this or 'event_column_name' (a column whose value supplies the event name per-row) — check Mixpanel's current Warehouse Connectors documentation for which one your use case requires.

insert_time_column_name

string

Optional

The name of a warehouse column marking when each row was inserted into the warehouse, used by Mixpanel to incrementally pick up only new rows on subsequent syncs instead of rescanning the whole table.

json_properties_column_name

string

Optional

The name of a warehouse column containing a JSON blob of additional event properties to merge into each imported event, for schemas that store extra attributes as a single JSON column rather than one column per property.

property_mappings

object

Optional

JSON object remapping warehouse column names to different Mixpanel property names on import, e.g. {"warehouse_col": "Mixpanel Property Name"}. Omit to import columns using their original warehouse column names.

run_every

integer

Optional

How often this warehouse import syncs, in NANOSECONDS — not seconds or milliseconds. This unit is easy to get wrong: 86400000000000 nanoseconds = 1 day (86400 seconds × 1,000,000,000). Common values: 3600000000000 = hourly, 86400000000000 = daily, 604800000000000 = weekly. Omit to keep Mixpanel's default sync frequency for this import type.

user_column_name

string

Optional

The name of the warehouse column containing the user identifier (Mixpanel distinct_id) to associate with each imported event.

mixpanelanalytics_warehouse_import_create_lookup_table

Warehouse import create lookup table

Create a new warehouse import that syncs a warehouse table into Mixpanel as a Lookup Table, joining additional properties onto events or user profiles by a shared key (similar in effect to 'mixpanelingestion_lookup_table_replace', but kept continuously in sync from the warehouse instead of being replaced manually via CSV). 'mixpanel_property' identifies which Mixpanel property this table enriches, and 'property_key_column_name' is the warehouse column holding the join key. Requires a warehouse source to already be configured in Mixpanel (Lexicon > Warehouse Sources) — pass its id as 'warehouse_source_id'. Requires Warehouse Connectors to be enabled on your Mixpanel plan; without it, calls to this endpoint return a 403 Forbidden regardless of input.

Parameters

Name

Type

Required

Description

mixpanel_property

object

Required

JSON object identifying which Mixpanel property this lookup table enriches. Shape: {"value": , "resourceType": <"event" or "user">, "customPropertyId": }. Example: {"value": "Total Revenue", "resourceType": "event", "customPropertyId": null} enriches the event property "Total Revenue". Find the exact property identifiers in Lexicon.

project_id

string

Required

Your numeric Mixpanel Project ID, required to authenticate Service Account requests. Find it in Mixpanel under Project Settings > Overview.

property_key_column_name

string

Required

The name of the warehouse column containing the lookup key — the value that matches this property's values on incoming events or profiles.

sync_mode

string

Required

How this lookup table import stays in sync with the warehouse table. 'full_sync' re-imports the entire table on every scheduled run (see 'run_every'), keeping it continuously up to date. 'one_time' imports the table once and does not sync again automatically.

table_params

object

Required

warehouse_source_id

integer

Required

databricks_params

object

Optional

run_every

integer

Optional

mixpanelanalytics_warehouse_import_run_sync

Sync warehouse import run

For more tools, view docs.

Build your Agent

Drop the toolkit in, point it at the tenant, and your agent can run funnel and retention queries from the first run.

import { ScalekitClient } from "@scalekit-sdk/node";
import { DynamicStructuredTool } from "@langchain/core/tools";
import { createReactAgent } from "@langchain/langgraph/prebuilt";
import { z } from "zod";

const sk = new ScalekitClient(envUrl, clientId, clientSecret);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["mixpanelanalytics"], toolNames: ["mixpanelanalytics_events_query", "mixpanelanalytics_funnels_query", "mixpanelanalytics_retention_query"] },
  pageSize: 100,
});

const lcTools = tools.map((t) => new DynamicStructuredTool({
  name: t.tool.definition.name,
  description: t.tool.definition.description,
  schema: z.object({}).passthrough(),
  func: async (args) => {
    const { data } = await sk.tools.executeTool({
      toolName: t.tool.definition.name,
      identifier: "user_123",
      params: args,
    });
    return JSON.stringify(data);
  },
}));

const agent = createReactAgent({ llm, tools: lcTools });

import { ScalekitClient } from "@scalekit-sdk/node";
import OpenAI from "openai";

const sk = new ScalekitClient(envUrl, clientId, clientSecret);
const openai = new OpenAI();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["mixpanelanalytics"], toolNames: ["mixpanelanalytics_events_query", "mixpanelanalytics_funnels_query", "mixpanelanalytics_retention_query"] },
  pageSize: 100,
});

const llmTools = tools.map((t) => ({
  type: "function",
  function: {
    name: t.tool.definition.name,
    description: t.tool.definition.description,
    parameters: t.tool.definition.input_schema,
  },
}));

const resp = await openai.responses.create({
  model: "gpt-4o", input: prompt, tools: llmTools,
});

import { ScalekitClient } from "@scalekit-sdk/node";
import Anthropic from "@anthropic-ai/sdk";

const sk = new ScalekitClient(envUrl, clientId, clientSecret);
const anthropic = new Anthropic();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["mixpanelanalytics"], toolNames: ["mixpanelanalytics_events_query", "mixpanelanalytics_funnels_query", "mixpanelanalytics_retention_query"] },
  pageSize: 100,
});

const llmTools = tools.map((t) => ({
  name: t.tool.definition.name,
  description: t.tool.definition.description,
  input_schema: t.tool.definition.input_schema,
}));

const msg = await anthropic.messages.create({
  model: "claude-sonnet-4-6", max_tokens: 1024,
  tools: llmTools,
  messages: [{ role: "user", content: prompt }],
});

import { Agent } from "@google/adk/agents";
import {
  MCPToolset, StreamableHTTPConnectionParams,
} from "@google/adk/tools/mcp";

const toolset = new MCPToolset({
  connectionParams: new StreamableHTTPConnectionParams({
    url: "https://mcp.scalekit.com/mixpanelanalytics",
    headers: { Authorization: `Bearer ${userScopedToken}` },
  }),
});

const agent = new Agent({
  name: "agent", model: "gemini-2.0-flash",
  tools: await toolset.getTools(),
});

Try these prompts

Paste any prompt into your agent to start querying Mixpanel from your workflows.

Funnels and retention

Copy the prompt

Copied

Show conversion for saved funnel [funnel_id] over the last 14 days.

Copy the prompt

Copied

Compare week-one retention for users who did [event] against those who did not.

Copy the prompt

Copied

Which step in [funnel_id] lost the most users this month?

Event exploration

Copy the prompt

Copied

List the top events today with percent change against yesterday.

Copy the prompt

Copied

Break [event] down by the values of [property] for last week.

Copy the prompt

Copied

Show the activity stream for distinct_id [id] over the last 7 days.

Schemas and profiles

Copy the prompt

Copied

List event properties with no Lexicon description.

Copy the prompt

Copied

Query profiles where plan is enterprise and last_seen is over 30 days ago.

Copy the prompt

Copied

Run this JQL script and summarize the output.

SEE HOW AUTH WORKS

Your users connect once. Their Mixpanel Analytics credentials stay vaulted, every call is scope-checked, and every action is logged.

Authorize

Your user connects

Mixpanel Analytics

once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access

Who:

user ‘A’

when:

Once per user

access:

Limited to user

Store

Their

Mixpanel Analytics

token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection

vault:

encrypted

scope:

per-user

tokens:

auto-refreshed

Resolve

When your agent calls a

Mixpanel Analytics

tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs

speed:

~40ms

check:

before every call

seen by:

nobody

Audit

Every

Mixpanel Analytics

tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it

history:

90 days

export:

SIEM-ready

logged:

every call

Test other agents

See the same per-user auth pattern across other connectors.

GTM and RevOps Teams

Revenue forecast commentary

Pulls open pipeline from Salesforce and HubSpot, calculates coverage against quota, flags at-risk stages, posts commentary to Slack, and logs every snapshot to Google Sheets.

GTM and RevOps Teams

Deal intelligence agent

Pulls recent Gong calls, scores deal risk with an LLM, cross-references the record in Attio, and DMs each owner their at-risk deals in Slack. Every read is scoped to that rep's own access.

Engineering Teams

Engineering standup agent

Pulls commits from GitHub and GitLab, tracks issue movement in Jira, and posts a per-engineer standup brief to Slack. Each engineer's activity is read on their own delegated OAuth.

GTM and RevOps Teams

Competitive intelligence briefing agent

Scans Gong calls for competitor mentions, matches each one to its Notion battlecard, and DMs every affected rep a single Slack digest per cycle. Every call runs as the PMM who owns the briefing, never a shared bot.

Why Scalekit

Secure your agent's access. Connectors ship in minutes

01.

Shared tokens break per-user analytics

A shared token looks fine in a demo. In production every call looks like a service account. Scalekit resolves the real user credential so attribution, audit, and scope stay accurate.

// shared token
audit → bot_service_account
user_filter → broken

// scalekit
audit → user_abc
scope → enforced ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

Mixpanel Analytics today. Others tomorrow.

“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”

Venu Madhav Kattagoni

Head of Engineering / Von

FAQs

Frequently Asked Questions

Does the agent access Mixpanel Analytics as the user or as a shared key?

As the user. Each workspace member authorizes once and Scalekit resolves their credential at request time. Audit logs attribute every action to that user, not a shared service account.

Where is the Mixpanel Analytics service account key stored?

In Scalekit's managed AES-256 token vault, namespaced per tenant. Refresh is automatic. Revocation is a single dashboard action. Tokens never appear in prompts, logs, or LLM context.

Can I limit what the agent is allowed to do in Mixpanel Analytics?

Yes. Pass a tool name filter to listScopedTools so the analytics agent only sees the subset you authorize. Pre-API-call scope checks block out-of-policy actions before the request reaches Mixpanel Analytics.

What happens when a user revokes Mixpanel Analytics access?

The connection is invalidated on the next tool call. Subsequent requests for that user fail closed with a clear error. Other users in the tenant remain unaffected. The event is logged for audit.

Can a query in one tenant read another tenant's Mixpanel project?

No. The service account you register is project scoped and lives in that tenant's namespace of the AES-256 vault. Scalekit resolves it per request, so a query issued for one tenant can only touch that tenant's project.

On this page

Tool and actions Framework Compatibility Starter Prompt Auth Lifecycle Try Other Agents Why Scalekit