dejo
GitHubGet started

Your agent has been
here before.

Your AI agent learns a website once. Every run after that is one tool call, with zero model calls.

Get started
Learning the site0/6 pages
38%fewer tokens than Playwright MCP on the first run
78%fewer on every run after
1tool call a run once learned, on 17 of 19 real-site tasks

Works with the AI agent you already use

Claude CodeCodexCursorGitHub CopilotWindsurfGemini CLIClaudeClineRoo CodeKilo CodeGooseAmpOpenCodeJunieKiroTraeAntigravityLM StudioOpenClaw

How it works

The first time, your agent explores the site the normal way. Dejo saves every step it takes as a trail.

Every run after that is one tool call. Dejo walks the trail in plain code and checks each page as it goes. Zero model calls.

When a site changes, only the broken step fails. Your agent fixes that one step, and Dejo saves the fix for every run after.

Benchmark

Cheaper to learn.
Then it remembers.

19 real tasks on lesser-known websites, each in a brand-new Claude session, against Playwright MCP and Playwright CLI. Dejo spends less on the first run, then replays the task in about one tool call.

DejoPlaywright MCPPlaywright CLI
tokens per run, lower is better
0200K400K600K800Krun 12345678910run 1: 251K tokens89K tokens from run 2 on

The same task run ten times, averaged over 19 tasks. Runs 1 and 2 are measured; from run 3 each tool costs what it did on run 2.

Claude Sonnet 5, medium effort. @playwright/mcp 0.0.80. Dejo 0.2.0.

First run, tokens
Dejo251K
Playwright MCP409K
Playwright CLI559K
10 runs, in total, tokens
Dejo1.1M
Playwright MCP4.1M
Playwright CLI5.6M
Right answersDejo 19 of 19, Playwright MCP 17, Playwright CLI 18
Every run after17 of 19 in one tool call, about $0.02 a run
A typical first runA little less than Playwright MCP. The big saving is on the hard sites.
Read the methods and results
Built for real websites

Replay is the core.
Here is everything around it.

Real sites have sign-ins, more than one account, many pages, and other agents walking the same routes.

See every tool

Sign in once

A real browser window opens and you sign in yourself. The profile stays signed in, and Dejo never asks for your password.

Many accounts, side by side

Give a customer, a vendor and an admin a profile each. They stay signed in together, and your agent switches between them instantly.

Knows every page it saw

Dejo remembers what was on each page it visited. A new task on a known site starts from that map, not from zero.

One trail, many questions

Learn how to count the issues on one repo, then ask the same of any other repo. Dejo swaps the subject and walks the same route.

Shared trails

Pick up a trail another agent on your machine already made, or share your own. Typed values and which account was used never travel.

Forget anytime

One command wipes everything Dejo knows about a site. Your memory, your call.

Get started

Set it up
in a minute.

Install it, add it to your agent, then ask for something on a website. Your agent does the rest.

Paste it into your agent and it sets Dejo up for you.

1

Install

Python 3.11 or newer

pip install dejo-mcp
playwright install chromium # the browser, about 150 MB
2

Add it to your agent

in your terminal

claude mcp add --scope user dejo -- dejo-mcp
3

Ask for something on a website

“How many open issues does microsoft/vscode have?”

The first time, your agent explores. Every time after, it is one call.

No API key

Dejo never calls a model. The agent you already use does the thinking.

Memory on your machine

What Dejo learns is kept in a local file. Wipe a site any time with one command.

Free and open source

MIT licensed. Read the code, run it yourself, change what you like.

FAQ

Frequently asked questions

The honest answers, including where Dejo is not the right tool.

Is Dejo just a cache?

No. A cache replays and hopes. Every step Dejo saves carries a check that proves it landed: the page changed, the file downloaded, the field holds what was typed. So it can tell a changed site from a working one, and fix only the step that moved.

What happens when a website changes?

Each step keeps up to nine ways to find its control, so most redesigns cost nothing. If every way to one control is gone, Dejo stops at that step and asks your agent about that one control. The fix is saved and the rest of the route is untouched. If more than half the steps break, the route is relearned, starting from what Dejo already knows about the site.

Does Dejo call an AI model?

No. Replay is plain code with zero model calls, and there is no API key to set. The agent you already use does the thinking, on the first run and when a step needs fixing.

Is it cheaper on every task?

No. The first run costs the most, because Dejo does the job and learns the route. On a single page with a one-line answer there is nothing to remember, so it is not cheaper there. It pays off on tasks with steps that you run again.

Where is the memory kept, and can I delete it?

On your machine, in a local file. One command wipes everything Dejo knows about a site, its map included. After that the next run says the memory is gone. It does not quietly fall back to guessing.

Does it store my passwords?

No. You sign in yourself, in a real browser window. A step that needs a password records only that a password goes there; the value is read at run time from an environment variable or a local secrets file. Export a trail and search it: there is nothing to find.

Which agents does it work with?

Dejo is one standard MCP server with no arguments, so any agent that can run an MCP server can use it, including Claude Code, Codex and Cursor.

Is it free?

Yes. Dejo is free and open source under the MIT license.

Make your agent remember.

Free and open source. Install it once, and every site your agent learns stays learned.

Get started