Model Context Protocol

Overlayed provides two Model Context Protocol (MCP) servers:

Documentation MCP (https://docs.overlayed.gg/mcp): search the Overlayed docs from your AI tools. No authentication required.
API MCP (https://api.overlayed.gg/mcp): work with your Overlayed account (applications, releases, analytics, feature flags, notices, and more) using an API key. See Overlayed API MCP.

What is MCP?

The Model Context Protocol creates standardized connections between AI applications and external services like documentation. When connected, your AI tool can search the Overlayed documentation directly during response generation, giving you more accurate answers about building overlays.

Connect to the documentation MCP

Choose your preferred tool below to connect the Overlayed documentation MCP server:

Claude
Claude Code
Cursor
VS Code

Add the Overlayed MCP server to Claude

Navigate to the Connectors page in the Claude settings.
Select Add custom connector.
Add the Overlayed MCP server:

Name: Overlayed
URL: https://docs.overlayed.gg/mcp

Select Add.

Access the MCP server in your chat

When using Claude, select the attachments button (the plus icon).
Select the Overlayed MCP server.
Ask Claude a question about Overlayed.

See the Model Context Protocol documentation for more details.

To use the Overlayed MCP server with Claude Code, run the following command:

claude mcp add --transport http Overlayed https://docs.overlayed.gg/mcp

Test the connection by running:

claude mcp list

See the Claude Code documentation for more details.

Open MCP settings

Use Command + Shift + P (Ctrl + Shift + P on Windows) to open the command palette.
Search for “Open MCP settings”.
Select Add custom MCP. This opens the mcp.json file.

Configure the Overlayed MCP server

In mcp.json, add:

{
  "mcpServers": {
    "Overlayed": {
      "url": "https://docs.overlayed.gg/mcp"
    }
  }
}

Test the connection

In Cursor’s chat, ask “What tools do you have available?” Cursor should show the Overlayed MCP server as an available tool.

See Installing MCP servers in the Cursor documentation for more details.

Create a .vscode/mcp.json file in your project and add:

{
  "servers": {
    "Overlayed": {
      "type": "http",
      "url": "https://docs.overlayed.gg/mcp"
    }
  }
}

See the VS Code documentation for more details.

Overlayed API MCP

The API MCP server exposes the Overlayed API as tools, so your AI tool can list your applications, check releases, query analytics, manage feature flags and notices, and more.

Create an API key

Create an API key in the Overlayed Dashboard. The tools available to the MCP server match the permissions of the key, so grant only what you need.

Add the server to your AI tool

Replace YOUR_API_KEY with the key you created.

Claude Code
Codex

claude mcp add --transport http --scope user overlayed https://api.overlayed.gg/mcp --header "x-api-key: YOUR_API_KEY"

Test the connection by running:

claude mcp list

Add the following to ~/.codex/config.toml:

[mcp_servers.overlayed]
url = "https://api.overlayed.gg/mcp"
http_headers = { "x-api-key" = "YOUR_API_KEY" }

To keep the key out of the file, set an OVERLAYED_API_KEY environment variable and replace the http_headers line with env_http_headers = { "x-api-key" = "OVERLAYED_API_KEY" }.

Tools that write data act on your live production account. Grant only read permissions if you only need to query data, and never commit your API key to source control.

Example prompts

Once connected, ask your AI tool about your app in plain language. It picks the right tools and pulls the data for you.

Analytics insights

“How many daily active users did my app have over the last 30 days, and how does that compare to the 30 days before?”
“What does weekly retention look like for users who joined this month?”
“Which games are my users playing the most?”
“What percentage of users are still on an old app version?”
“Summarize this week’s sessions, session duration, and new users in a short report.”

Bug report diagnostics

“Did bug reports spike after the latest release? Show the daily trend around the release date.”
“List the bug reports from the past week and group them by common cause.”
“Download the logs attached to the most recent bug report and tell me what went wrong.”
“Which releases are the most recent bug reports tied to?”

User feedback

“What are users suggesting since the last release? Summarize the top themes.”
“Find feedback that mentions crashes or performance and update its status.”

Releases and channels

“What’s the latest release on each channel?”
“How quickly are users adopting the newest version?”

Ad revenue

“What was my ad revenue last month, broken down by country?”
“How have CPM and fill rate changed over the last 30 days?”
“Which hours of the day earn the most ad revenue?”

Feature flags and notices

“List my feature flags and which ones are currently enabled.”
“Enable the new-hud feature flag for staff accounts only.”
“Create a notice telling users about scheduled maintenance tomorrow at 10:00 UTC.”

Prompts that change data (updating feedback, feature flags, or notices) need a key with write permissions. Most AI tools ask you to confirm before running them.

Benefits

Once the documentation MCP is connected, you can:

Get accurate answers about Overlayed features and APIs without leaving your IDE
Have the AI proactively search Overlayed docs when relevant to your questions
Access up-to-date documentation during development

Get Started

Essentials

Windows

Keybinds

Notices

Feature Flags

Event & Log Tracking

Logging

Deployment

Ads

Advanced

Extras

Model Context Protocol

What is MCP?

Connect to the documentation MCP

Overlayed API MCP

Example prompts

Benefits

​What is MCP?

​Connect to the documentation MCP

​Overlayed API MCP

​Example prompts

​Benefits

What is MCP?

Connect to the documentation MCP

Overlayed API MCP

Example prompts

Benefits