In short:
ak-agent-browserlets the AI drive a separate browser through theagent-browserCLI: open pages, click, fill forms, take screenshots, hunt bugs. It does not use your everyday Chrome, so work that needs cookies or a signed-in account goes toak-chrome-profile.
Disclosure: I maintain AgentKit and have an affiliate relationship with it: the AgentKit link in this post is an affiliate link; buying through it gets you 30% off.
Summary
Use it when you want the AI to walk through a flow on the web or in an Electron app, capture the result, or hunt bugs before a release, even in CI with no browser installed. Do not use it when the work needs a session already signed in to your real Chrome.
flowchart LR
A["Work that needs a browser"] --> B{"Need real signed-in Chrome?"}
B -->|"No"| C["ak-agent-browser"]
B -->|"Yes"| D["ak-chrome-profile"]
C --> E["Screenshots, data, bugs found"]Four things are enough to get going:
| Concept | Question | Example |
|---|---|---|
| Separate browser or real Chrome | Does this need a session already signed in to your Chrome? | Testing a sign-up form on localhost uses this skill. Opening an admin page with the company Google account in Chrome uses ak-chrome-profile. |
| Snapshot and element refs | How does the AI click the right button? | agent-browser snapshot -i says the "Send" button is @e3, then the AI runs agent-browser click @e3. If the page changes, take a new snapshot: old refs go stale. |
| Instructions from the installed version | Which commands are right for the agent-browser on your machine? | Before the first command, the AI runs agent-browser skills get core. For special work it also loads skills get electron, slack or dogfood (QA-style bug hunting). |
| Local machine or cloud browser | Does the place it runs have a browser? | In CI without Chromium, set BROWSERBASE_API_KEY, BROWSERBASE_PROJECT_ID and run with -p browserbase. Each open is a new cloud session, close ends it. |
This skill is only a guide layer: the command details live in the CLI and change between versions, so the skill does not copy them and always reads them from the installed version.
Sample prompt (checking a form):
/ak-agent-browser Open http://localhost:3000/signup for <app name>.
Fill the sign-up form with the email test@example.com and any password, click Sign up.
Done when: there is a screenshot of the page shown after the click, and you report whether that page has any error message.This skill has no flags.
Below is the detailed version: the run flow, 5 situations with prompts, the change history by ak release, and common pitfalls. The detailed version and the interactive page are for subscribers. Your agent can also search them through the portal's MCP.
Opens when your account has subscriber access
This guide is for subscribers
Once your account has access, the full piece, the interactive visual and the agent call all open.
- Sign in or create a modernEngineer accountReading access belongs to your account email.
- Choose a plan on the pricing page1, 3 or 12 months, paid by VietQR bank transfer. Access opens about a minute after the money arrives.
- Or redeem a Facebook subscriber codePeople supporting the page on Facebook get an activation code by message each month.
Already a subscriber? Sign in to read.
Use this piece in your agent
Get an API keyCopy the text below and paste it into Claude Code, Codex or whichever agent you use. The agent reads the piece and checks it against the repo you have open.
Read "ak-agent-browser: drive a separate browser to test and take screenshots" on modernEngineer and apply it to this repo.
1. Call the content_get tool of the modernengineer MCP server with locale "en", slug "ak-agent-browser". If the modernengineer MCP server is not connected, stop and walk me through connecting it from https://modernengineer.dev/en/mcp.
2. Summarize the main steps of the piece and point out which apply to this repo.
3. Propose concrete changes and ask me before editing any file.Connect MCP in Claude Code (once)
claude mcp add --transport http modernengineer \
https://api.modernengineer.dev/mcp \
--header "Authorization: Bearer <API key>"Reading the full piece over MCP needs an API key from a subscriber account or one that bought this guide. Without access, the agent only gets the opening.
- Version
- v1.0.0
- Updated
- 22/10/2026
- Licence
- All rights reserved
- Slug
- ak-agent-browser
