schrute

Schrute

Teach your AI a website once. After that, it can repeat the job much faster.

Schrute is for repeated website tasks.

It supports both paths: you can teach it by having it watch what happens in a real browser, or you can let it discover useful backend structure on its own when a site exposes it. Schrute can do either well, and it is up to you which path to use for a given site or task. Either way, it turns what it finds into reusable tools. That means the first run can happen in the browser, or you can start from direct discovery when that fits better, and later runs can often skip the UI and go straight to the site's backend.

If you keep asking an AI to do the same website task over and over, Schrute is the layer that helps it stop starting from scratch every time.

• Learn from a real browser session
• Reuse your logged-in state
• Replay repeatable tasks without brittle click scripts
• Fall back to the browser when direct replay is not possible
• Use it from MCP, CLI, REST, Python, or TypeScript

Why People Use It

Without Schrute:

• An agent opens the site again
• Clicks through the UI again
• Waits for the page again
• Pays the same latency again

With Schrute:

• You teach it the task once
• Schrute learns the request pattern behind the page
• The next run can often call the learned action directly

That is especially useful for things like:

• pulling the same dashboard data every day
• checking prices or market pages repeatedly
• searching a site with the same flow many times
• reusing internal tools that only work when you are already logged in

Quick Start

npm install -g schrute
schrute setup

Install

Choose the install path that fits your environment:

• npm CLI (recommended)

  npm install -g schrute
  schrute setup
  

• Homebrew

  brew install sheeki03/tap/schrute
  schrute setup
  

• Docker

  Pull the image:

  docker pull ghcr.io/sheeki03/schrute:latest
  

  Then run Schrute with persistent data and an auth token:

  docker run --rm \
    -p 3000:3000 \
    -p 3001:3001 \
    -e SCHRUTE_AUTH_TOKEN=my-secret \
    -v schrute-data:/data \
    ghcr.io/sheeki03/schrute:latest
  

  A Docker Hub mirror can also be published when enabled for the repository.

• Standalone binaries

  Download the latest archive for Linux, macOS, or Windows from GitHub Releases, unpack it, and run schrute.

If you want to use Schrute from an AI client over MCP:

{
  "mcpServers": {
    "schrute": {
      "command": "npx",
      "args": ["-y", "schrute", "serve"]
    }
  }
}

Ways To Use Schrute

You can use the same learned skills in different ways depending on your workflow:

• MCP Best when you want Claude Code, Cursor, Cline, Windsurf, or another MCP client to call learned website actions as tools.

• CLI Best when you want to explore, record, inspect, and run skills manually from the terminal.

• REST API Best when you want another app, script, or backend service to call Schrute over HTTP.

• Python and TypeScript clients Best when you want a lightweight client package instead of calling raw HTTP endpoints yourself.

So Schrute is not tied to one interface. You can teach it a task once, then reuse that same learned task from the interface that fits your workflow.

First Run In 2 Minutes

# 1. Start Schrute
schrute serve

# 2. Open a site in a browser session
schrute explore https://httpbin.org

# 3. Start recording a task
schrute record --name get_ip

# 4. In the opened browser, go to:
#    https://httpbin.org/ip

# 5. Stop recording
schrute stop

# 6. Poll the background pipeline job until skill generation completes
schrute pipeline <job-id>

# 7. Run the learned skill
schrute execute httpbin_org.get_ip.v1 --yes

What just happened:

1. Schrute watched the browser traffic for that action.
2. It found the real request behind the page.
3. It saved that request as a reusable skill.
4. You can now run that learned action again without manually driving the page.

Commands Most People Will Use

schrute explore https://example.com
schrute record --name my_action
schrute stop
schrute pipeline <job-id>
schrute execute my_skill.v1

schrute skills list --status active
schrute skills search "bitcoin price"
schrute skills show <skill-id>

schrute workflow create --site example.com --name summary --spec '{"steps":[...]}'
schrute workflow run example_com.summary.v1

schrute discover https://api.example.com
schrute doctor
schrute trust

What Schrute Can Do Today

Schrute is no longer just "record and replay." Here is what the current product does, in practical terms: