Statsig MCP server: 67 tools, 27 for reviews

A feature-flag connector whose largest single surface is an approval workflow, plus the separate documentation server people confuse it with.

Last verified 27 August 2026 · from Statsig's own MCP documentation

This page is one of 90 in a directory of monitoring and developer tools. Each page checks what an AI coding agent can get out of the tool through its MCP server, the connector that lets the agent query the tool directly.

What is the Statsig MCP server?

It is Statsig's server for the Model Context Protocol. Statsig is an experimentation and feature-flag platform, so a connected agent reaches gates, experiments and the metrics behind them.

The surface is broader than flags alone. Layers, which group experiments so that no user is in more than one of them, parameter stores, segments and metric sources all have their own tools, and Query_Logs_Explorer and Cluster_Log_Patterns reach observability data.

27 of the 67 tools are review tools. An agent can create a gate review, approve it, and commit it, which is a governance workflow rather than a data query. Gates, experiments and autotunes each carry a set of them; an autotune is Statsig's multi-armed bandit, which shifts traffic towards the best-performing variant for a single metric.

Statsig sits in CI and release, reaches production observability through its logs explorer, and reaches nothing in agent observability or cost.

For an agent, Statsig answers what is gated, what is being tested and who approved it. The review tools also record decisions. A gate change can be proposed, approved and committed through this connection. It has no application metrics or traces, so it pairs with a telemetry connector.

What can an AI agent do with Statsig?

The review tools come in a fixed sequence: create, edit, approve or reject, then commit or cancel. Reading Get_List_of_Gate_Reviews first shows what is already open.

What connecting costs before the first question

A context window is the amount of text a model can hold at once. Sixty-seven tools cost 3,669 to 4,619 tokens on Claude, which is modest for a surface that covers gates, experiments, layers, metrics and three separate review workflows.

The review tools are the reason the count is high without the cost following. Nine near-identical tools each for gates, experiments and autotunes share a description shape, so they add names more than they add text.

Encodings: GPT-5.6, GPT-5.5 o200k_base; GPT-4, GPT-3.5 cl100k_base; Claude Haiku 4.5 to Opus 5 count_tokens. Counted from the 67 tool entries in the vendor's own documentation. That is the only public inventory.

The server could not be run for a live tools/list. So no input schemas are available. These figures cover each tool's name and description only. They are a floor, not the figure.

The Claude columns are measured through the API's count_tokens endpoint, with an empty schema per tool. So they cover the same content as the GPT columns. The range there is two tokenizer generations. Haiku 4.5 and Opus 4.6 sit at the low end. Opus 5 and Sonnet 5 sit at the high end. The gap between them is about a quarter.

Statsig MCP server limits: where answers come back incomplete

Two facts set the limits below. The same connection can propose a change and approve it. A second Statsig server shares the brand and reaches none of your data.

How to configure Statsig MCP for agents

1. 1

   Decide whether an agent may approve

   The review category includes approve, reject and commit tools for gates, experiments and autotunes.

2. 2

   Know which server you are configuring

   The Docs MCP is read-only over public documentation; the product endpoint at api.statsig.com/v1/mcp needs a Console key and reaches your data.

3. 3

   Understand the update semantics

   Several tools are named Update_..._Entirely, and they replace an object rather than patching it.

4. 4

   Scope the Console key

   Statsig states that the authenticated server accesses product resources according to the account's permissions.

Paste this into your agent’s instructions

The same connector can propose a change and approve it, and several updates replace rather than patch. The response must disclose each behavior.

When you answer using Statsig MCP tools, state:
- The project and object the answer covers.
- Whether you used the Docs MCP or the authenticated product
  endpoint. They return different things.
- Whether you created, approved, committed or canceled a
  review, or replaced an object with an Update_..._Entirely tool.
Never state that data does not exist. State that no objects
matched the filters above.
Write your answer in ASD-STE100 Simplified Technical
English. Use short sentences with one idea in each.

Do you need the MCP server at all?

Not for pulling results on a schedule. Statsig's Console API covers the same objects and is the direct route there.

The connector earns its place on the review workflow. Creating a gate review and finding its eligible reviewers from the editor is not something a fixed report does.

For flags without the experimentation surface, LaunchDarkly's connector is narrower and its tools are all described.

Connect it when an agent participates in flag and experiment decisions. Use the Console API for reporting.

Statsig MCP server setup

The server is hosted at api.statsig.com/v1/mcp and takes a Statsig Console API key. Nothing is installed.

Statsig publishes configuration guides for Codex, Cursor, Claude Code, a ChatGPT connector and a manual setup route.

claude mcp add --transport http statsig \
  https://api.statsig.com/v1/mcp \
  --header "STATSIG-API-KEY: <your-console-key>"

{"mcpServers":{"statsig":{
  "url":"https://api.statsig.com/v1/mcp",
  "headers":{"STATSIG-API-KEY":"<your-console-key>"}}}}

[mcp_servers.statsig]
url = "https://api.statsig.com/v1/mcp"
headers = { "STATSIG-API-KEY" = "<your-console-key>" }

Every other client

Each block below is the configuration for one client, with the file path and the key that client expects.

{"mcpServers":{"statsig":{"url":"https://api.statsig.com/v1/mcp",
  "headers":{"STATSIG-API-KEY":"<your-console-key>"}}}}

{"mcpServers":{"statsig":{"url":"https://api.statsig.com/v1/mcp",
  "headers":{"STATSIG-API-KEY":"<your-console-key>"}}}}

{"servers":{"statsig":{"url":"https://api.statsig.com/v1/mcp",
  "headers":{"STATSIG-API-KEY":"<your-console-key>"}}}}

{"mcpServers":{"statsig":{"url":"https://api.statsig.com/v1/mcp",
  "headers":{"STATSIG-API-KEY":"<your-console-key>"}}}}

{"mcpServers":{"statsig":{"url":"https://api.statsig.com/v1/mcp",
  "headers":{"STATSIG-API-KEY":"<your-console-key>"}}}}

# Statsig's Docs MCP, for public documentation only.
# Configure it separately from the product endpoint above.

Read-only access, permissions and security

• 27 tools drive approvals

  Gates, experiments and autotunes each have create, edit, approve, reject, commit and cancel review tools.

• Approval and commit are reachable from the same connection

  Nothing in the tool list separates proposing a change from approving it.

• Several updates replace rather than patch

  Tools named Update_..._Entirely overwrite the whole object.

• One tool deletes

  Delete_Param_Store removes a parameter store outright.

• Permissions follow the key

  Statsig documents that the authenticated server accesses resources according to the account's permissions.

Troubleshooting

The endpoint returns 401

It accepts an active Console key only, and says so in the error. Other Statsig key types are rejected.

An update wiped fields you did not touch

Tools named Update_..._Entirely replace the object rather than patching it. Read the object first and send it back complete.

The agent answered from documentation, not your data

It may have used the Docs MCP, which is a separate read-only server over public documentation.

A review will not commit

Check eligibility. Get_Gate_Eligible_Reviewers and its experiment equivalent return who may approve.

Statsig MCP server: Reference

What engineers report

Statsig documents its tools in per-capability tables, so the counts below are its own.

Tool inventory read from Statsig's own MCP overview on 24 August 2026. An unauthenticated tools/list against api.statsig.com/v1/mcp returned 401 on the same date, with a message stating that the endpoint accepts an active Console key only.

Should you connect an agent to Statsig?

Yes. Best for a Console key scoped to what the agent should decide, with review tools treated as a deliberate choice.

Strongest fit

Statsig teams who want an agent inside the gate and experiment review process, not just reading it.

Main advantage

A wide surface with a full approval workflow, plus logs tooling most flag connectors do not carry.

Main weakness

Two servers share the brand, and only the authenticated one reaches your project.

Operational risk

Approve and commit sit in the same tool list as create, and four tools replace objects entirely.

Frequently asked questions