Connect
Optimize
Secure
PostHog MCP Server
for AI Agents
Production-ready PostHog MCP server with 63 extensible actions — plus built-in authentication, security, and optimized execution.
PostHog MCP Server
Built by 
StackOne
StackOne 
Coverage
63 Agent Actions
Create, read, update, and delete across PostHog — and extend your agent's capabilities with custom actions.
Authentication
Agent Tool Authentication
Per-user OAuth in one call. Your PostHog MCP server gets session-scoped tokens with zero credentials stored on your infra.
Agent Auth →
Security
Agent Protection
Every PostHog tool response scanned for prompt injection in milliseconds — 88.7% accuracy, all running on CPU.
Prompt Injection Defense →
Performance
Max Agent Context. Min Cost.
Free up to 96% of your agent's context window to enhance reasoning and reduce cost, on every PostHog call.
Tools Discovery →What is the PostHog MCP Server?
A PostHog MCP server lets AI agents read and write PostHog data through the Model Context Protocol — Anthropic's open standard for connecting LLMs to external tools. StackOne's PostHog MCP server ships with 63 pre-built actions, fully extensible via the Connector Builder — plus managed authentication, prompt injection defense, and optimized agent context. Connect it from MCP clients like Claude Desktop, Cursor, and VS Code, or from agent frameworks like OpenAI Agents SDK, LangChain, and Vercel AI SDK.
All PostHog MCP Tools and Actions
Every action from PostHog's API, ready for your agent. Create, read, update, and delete — scoped to exactly what you need.
An Actions
- Retrieve An Action
Retrieve details of a specific action
- Update An Action
Update an action's configuration
- Delete An Action
Delete an action (soft delete)
A Dashboards
- Retrieve A Dashboard
Retrieve details of a specific dashboard
- Update A Dashboard
Update a dashboard's configuration
A Feature Flags
- Retrieve A Feature Flag
Retrieve details of a specific feature flag
- Update A Feature Flag
Update a feature flag's configuration
Groups Types Metrics
- Create Groups Types Metrics
Create a new metric for a group type
- Retrieve Groups Types Metrics
Retrieve details of a specific group type metric
- Update Groups Types Metrics
Update a group type metric's configuration
- Delete Groups Types Metrics
Delete a group type metric
An Insights
- Retrieve An Insight
Retrieve details of a specific insight
- Update An Insight
Update an insight's configuration
- Delete An Insight
Delete an insight (soft delete)
Members
- Update Members
Update a member's details
- Delete Members
Remove a member from the organization
Organizations
- Create Organizations
Create a new organization
- Retrieve Organizations
Retrieve a list of organizations
- Delete Organizations
Delete an organization
Projects
- Create Projects
Create a new project within an organization
- List Projects
Retrieve a list of all projects in the organization
- Delete Projects
Delete a project
Roles
- Create Role
Create a new role
- Retrieve Role
Retrieve details of a specific role
- Update Role
Update a role's details
- Delete Role
Delete a role
Role Memberships
- Create Role Membership
Assign a user to a role
- Retrieve Role Membership
Retrieve details of a specific role membership
- Delete Role Membership
Remove a user from a role
A Session Recordings
- Update A Session Recording
Update a session recording's metadata
- Delete A Session Recording
Delete a session recording
Users
- Retrieve Users
Retrieve details of a specific user
- Update Users
Update a user's details
- Delete Users
Delete a user
Other (29)
- Create Actions
Create a new action
- Create Dashboards
Create a new dashboard
- Create Feature Flags
Create a new feature flag
- Create Insights
Create a new insight
- List All Actions
Retrieve a list of all actions
- List All Activity Log
Retrieve activity log entries for a project
- List All Alerts
Retrieve a list of all alerts
- List All Dashboards
Retrieve a list of all dashboards
- List All Feature Flags
Retrieve a list of all feature flags
- List All Groups Types
Retrieve a list of all group types
- List All Groups Types Metrics
Retrieve metrics for a specific group type
- List All Insights
Retrieve a list of all insights
- List All Members
Retrieve a list of organization members
- Retrieve An Organization
Retrieve details of a specific organization
- List All Integrations
List all organization-level integrations
- Retrieve List
List all projects in an organization
- Retrieve An Organization Project
Retrieve details of a specific project
- List All Roles
Retrieve a list of all roles
- List All Role Memberships
List all memberships for a specific role
- List All Session Recordings
Retrieve a list of all session recordings
- List All Users
Retrieve a list of all users
- Update Groups Types Update Metadata
Update group types metadata
- Update Partial Organization
Update an organization's settings
- Update Partial Organization Project
Update a project's configuration
- Update Complete Product Onboarding
Mark product onboarding as complete
- Delete Groups Types
Delete a group type
- Capture Event
Send a single event to PostHog
- Capture Batch Events
Send multiple events to PostHog in a single request
- Change Organization
Move a project to a different organization
Set Up Your PostHog MCP Server in Minutes
One endpoint. Any framework. Your agent is talking to PostHog in under 10 lines of code.
MCP Clients
Agent Frameworks
JSON{
"mcpServers": {
"stackone": {
"command": "npx",
"args": [
"-y",
"mcp-remote@latest",
"https://api.stackone.com/mcp?x-account-id=<account_id>",
"--header",
"Authorization: Basic <YOUR_BASE64_TOKEN>"
]
}
}
}More Data & Analytics MCP Servers
Tableau
114+ actions
SurveyMonkey
104+ actions
Microsoft Excel
101+ actions
Qlik
100+ actions
Snowflake
80+ actions
Amplitude
53+ actions
Gong
34+ actions
PostHog MCP Server Resources
MCP Code Mode: Keeping Tool Responses Out of Agent Context
Anthropic's code_execution processes data already in context. Custom MCP code mode keeps raw tool responses in a sandbox. 14K tokens vs 500.
11 min
Comparing BM25, TF-IDF, and Hybrid Search for MCP Tool Discovery
Benchmarking BM25, TF-IDF, and hybrid search for MCP tool discovery across 916 tools. The 80/20 TF-IDF/BM25 hybrid hits 21% Top-1 accuracy in under 1ms.
10 min
Indirect Prompt Injection Defense for MCP Tools: A Technical Guide
MCP tools that read emails, CRM records, and tickets are indirect prompt injection vectors. Here's how we built a two-tier defense that scans tool results in ~11ms.
12 min
PostHog MCP Server FAQ
PostHog MCP server vs direct API integration — what's the difference?
A PostHog MCP server and direct API integration serve different use cases. Direct API integration is for software-to-software — backend code calling PostHog. A PostHog MCP server is for AI agents — MCP clients like Claude and Cursor, plus framework agents built with OpenAI, LangChain, or Vercel AI — discovering and calling PostHog at runtime. StackOne provides both.
How does PostHog authentication work for AI agents?
PostHog authentication for AI agents works through a StackOne Connect Session. Create one via the dashboard or the SDK — you get an auth link and ready-to-paste config for Claude Desktop, Cursor, and other MCP clients. Your user authenticates their own PostHog account; StackOne handles token exchange, storage, and refresh. Credentials never reach the LLM, and each user is isolated via
origin_owner_id.Are PostHog MCP tools vulnerable to prompt injection?
Yes — PostHog MCP tools can be vulnerable to indirect prompt injection. Any tool that reads user-written content — documents, messages, tickets, records, or free-text fields — is a potential vector. StackOne Defender scans every tool response before it enters the agent's context — regex patterns in ~1ms, then a MiniLM classifier in ~4ms. 88.7% accuracy, CPU-only.
What is the context bloat of a PostHog agent and how do I avoid it?
Context bloat happens when PostHog tool schemas and API responses eat your PostHog agent's memory, preventing it from reasoning effectively. A single PostHog query can return a massive JSON response, and connecting multiple tools compounds the problem. Tools Discovery and Code Mode reduce context bloat — loading only relevant tools per query and keeping raw responses out of the agent's context.
Can I limit which actions my PostHog agent can access?
Yes — you can limit which actions your PostHog agent can access directly from the StackOne dashboard. Toggle actions on or off, or restrict them to specific accounts, with no code changes to your agent. Session tokens can be scoped to exact actions so if one leaks, exposure stays contained.
Can I create custom agent actions for my PostHog MCP server?
Yes — you can create custom agent actions for your PostHog MCP server using Connector Builder. It's an integration agent your coding assistant (Claude Code, Cursor, or Copilot) can invoke to research PostHog's API, generate production-ready connector YAML, test against the live API, and validate before you ship.
When should I NOT use a PostHog MCP server?
Skip a PostHog MCP server if your integration is purely software-to-software — direct PostHog API integration is simpler when no AI agent is involved. For deterministic, compliance-critical operations (financial transactions, regulatory reporting), direct API gives you predictable behavior without agent-driven decision-making. MCP shines when AI agents need to dynamically discover and call PostHog actions at runtime.
What AI frameworks and AI clients does the StackOne PostHog MCP server support?
The StackOne PostHog MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.
Put your AI agents to work
All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.