ZeroToken is a local, open source MCP server that turns an AI agent's browser actions into reusable scripts. It's for people automating recurring web tasks, such as daily logins, report collection or browser tests, through OpenClaw, Cursor or another MCP client. The Python project uses Playwright and Chromium and carries the MIT license.
Its main distinction is script replay: after an agent works through a task, later runs can follow the recorded actions without asking an LLM to reason through every step again. Scripts support conditions, loops and variables, so repeated jobs can use different inputs. OpenClaw can trigger saved scripts through its scheduled jobs, while a persistent HTTP service preserves browser state across agent calls.
The browser tools handle multiple tabs and embedded frames, plus file uploads and downloads. Recordings retain page state, screenshots and results for review, and an exploration mode keeps unsuccessful experiments out of the final recording. ZeroToken stores scripts and execution records locally in SQLite.
When a step fails or needs judgment, execution pauses so the agent can retry, skip, amend or stop it. Alternative selectors and automatic waits help with changing pages; saved element fingerprints can help find an element after its original selector stops working. Page summaries and reduced screenshot payloads limit the context sent to the connected agent. Script health tracking records success rates and consecutive failures, and users can retire or restore saved scripts.
Claim this page and we'll verify you by hand. ZeroToken gets the verified badge, and you can upgrade the listing to be featured on localhosted. Proud to be listed? Put our badge on your site.
Want more people to find ZeroToken?Promote it
Something wrong or outdated on this page?
6.2KUpdated 4 months agoMIT
macOS · Docker#Agent Skills#Browser automation#Code execution
bb-browser lets AI agents work through your local Chrome browser using the website accounts you're already signed into. It's for people building agents that need to research across websites or interact with authenticated pages without a separate API integration for each service. The project is open source under the MIT license.
701Updated 2 days ago
#Browser automation#Human approval#MCP
Fortress is a Chromium fork for developers whose scrapers and AI agents encounter browser fingerprint checks. It runs as a local browser binary and changes the fingerprint inside Chromium's C++ code, rather than applying JavaScript overrides that pages can inspect.
28.2KUpdated 1 day agoApache-2.0
Linux · Docker#Browser automation#Code execution#MCP
893Updated 6 days agoMIT
Docker · Web#Agent Skills#Guardrails#Human approval
433Updated 4 days agoMIT
Browser Extension#Agent Skills#Browser automation#Code execution
52.8KUpdated 1 day agoApache-2.0
#Browser automation#MCP#Tool calling
Obscura is a headless browser engine for developers building AI agents and web scraping systems that need many independent browser sessions. Each session starts with fresh state, so cookies, logins, and cached data don't carry between tasks. It's written in Rust and runs without Chromium, with a smaller memory footprint intended to support more concurrent sessions per machine.
Auto Browser lets AI agents and human operators share a live Chromium browser, with a person able to take over when a login or fragile page needs attention. It's a self-hosted tool for internal dashboards, account workflows, and browser QA. The full stack runs on your own hardware through Docker Compose; Codespaces provides a hosted demo. It's open source under the MIT license.
Browser Control gives trusted coding agents access to the Chromium browser you already use, with its signed-in accounts and installed extensions. It's for developers who want agents to work in an existing browser session rather than a separate headless browser. The driver runs locally through a browser extension and local relay. It contains no LLM.
Chrome DevTools MCP gives AI coding assistants access to a live Chrome browser through a locally running MCP server. It's for developers who want their agent to inspect how a web app behaves, investigate errors and measure performance alongside its coding work. It works with clients such as Claude Code, Cursor, Copilot and Antigravity.