awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
eze-is avatar

eze-is/web-access

0
View on GitHub↗

Web Access

Web Access is a skill for local AI coding agents that gives them web access by driving a real Chromium-based browser. It connects to the browser the user already uses every day through its debugging port and a local proxy, so agents inherit existing logins and can work on dynamic pages, open background tabs, click, fill forms, scroll, upload files, and capture screenshots.

Beyond browser control, it routes each task to the cheapest sufficient channel: search, plain page fetch, raw retrieval, Markdown conversion through a third-party converter (which loses some information), or full browser control when login or dynamic rendering is needed. Multi-target research can be split across sub-agents that each work in their own isolated tab over one shared connection, keeping fetched content out of the main context. The skill also reads and clicks inside cross-site and cross-process iframes, and it keeps per-site notes such as URL patterns and known pitfalls so later sessions can reuse them.

The local proxy reuses a healthy existing connection and reconnects after disconnects, which avoids repeated browser authorization prompts. It closes tabs it created after an idle period and leaves the user's own tabs alone. It supports Chrome, Edge, Brave, Arc, and other Chromium browsers via a configured data directory, and it will not silently fall back to a different browser. It can also search local bookmarks and history to find previously visited pages, including internal systems. Preferences and site notes are stored in a home-directory folder outside the install, with a one-time non-destructive migration from older locations, so upgrades do not overwrite them.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI

Features

  • Browser Automation Agents - Gives local AI coding agents web access by driving a real Chromium browser with reused logins for dynamic pages and interactive actions.
  • Local Browser Proxies - Connects through a local proxy to the user's everyday Chromium browser, inheriting logged-in sessions and using background tabs that close when finished.
  • Browser Tab Control Proxies - Provides a local HTTP proxy over the Chrome DevTools Protocol so agents can open, navigate, script, and screenshot tabs through simple requests.
  • Browser Automation - Connects to the user's existing Chromium-based browser through its debugging port so tasks run in the real logged-in browser.
  • API and Browser Channel Routing - Routes each web task to the cheapest sufficient channel, from search and plain fetch up to full browser control when login or rendering is needed.
  • Browser Form Automation - Automates the user's daily browser with existing logins to open tabs, click, fill forms, scroll, upload files, and take screenshots.
  • Browser Control APIs - Exposes a local HTTP interface over a browser WebSocket connection to open, navigate, and manage tabs, closing idle tabs automatically.
  • Chrome DevTools Protocols - Connects to Chromium browsers through their debugging port and a local proxy so agents drive tabs, frames, and scripts over CDP.
  • Browser Interaction Actions - Drives the user's everyday logged-in Chromium browser through a local proxy to handle dynamic pages and interactive actions.
  • Multi-Channel Retrieval Routers - Routes each web task to the cheapest sufficient channel, from search and plain fetch to Markdown conversion or full logged-in browser control.
  • Web Research Agents - Splits multi-target research across sub-agents that each work in isolated tabs over one shared connection, then merges findings.
  • Site Experience Stores - Keeps per-site notes such as URL patterns and known pitfalls in a persistent home-directory folder that later sessions reuse and upgrades never overwrite.
  • Web Browsing Tools - Picks among search, page fetch, curl, Markdown conversion, and browser control per task, combining them to reach web content efficiently.
  • Parallel Research Orchestration - Splits multi-target research across sub-agents that each use their own isolated browser tab over one shared connection, keeping fetched content out of the main context.
  • Search and Web - Selects automatically among search, fetch, curl, reader proxies, and browser control to complete a web task.
  • Site-Specific Memory - Stores per-domain notes such as URL patterns and known pitfalls in a persistent folder that later sessions reuse and upgrades preserve.
  • Sub-Problem Parallelization - Splits independent research targets across sub-agents that each use their own browser tabs, keeping fetched content out of the main context.
  • Session-Sharing HTTP Daemons - Runs a long-lived local daemon that holds one authorized browser connection and exposes it as a simple HTTP API for many callers.
  • Iframe Navigation and Targeting - Reads and clicks inside cross-site and cross-process iframes by attaching to child sessions and descending nested frames by coordinates.
  • Page Interaction Actions - Queries elements, clicks, scrolls, fills forms, and uploads files, choosing scripted or real mouse actions to avoid anti-bot blocks.
  • Cross-Frame Interaction Synchronization - Reads and clicks content inside cross-site, nested, and cross-process iframes that ordinary page scripts cannot reach.
  • Auto-Reconnecting Connections - Runs a background proxy that reuses a healthy browser connection and reconnects after disconnects, avoiding repeated browser authorization prompts.
  • Per-Agent Tab Isolation - Gives each worker agent its own background tab over one shared connection, keeping parallel tasks separated and fetched content out of the main context.
  • Web Page Retrievers - Chooses between search, targeted page fetch, raw HTML retrieval, and a live browser depending on URL knowledge and login needs.
9,358 stars·659 forks·JavaScript·11 views

Star history

Star history chart for eze-is/web-accessStar history chart for eze-is/web-access

How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

Frequently asked questions

What does eze-is/web-access do?

Web Access is a skill for local AI coding agents that gives them web access by driving a real Chromium-based browser. It connects to the browser the user already uses every day through its debugging port and a local proxy, so agents inherit existing logins and can work on dynamic pages, open background tabs, click, fill forms, scroll, upload files, and capture screenshots.

What are the main features of eze-is/web-access?

The main features of eze-is/web-access are: Browser Automation Agents, Local Browser Proxies, Browser Tab Control Proxies, Browser Automation, API and Browser Channel Routing, Browser Form Automation, Browser Control APIs, Chrome DevTools Protocols.

Which projects share features with eze-is/web-access?

Projects with overlapping indexed features include: garrytan/gstack — gstack is an AI agent framework and development workflow system designed to automate the software development… browsermcp/mcp — BrowserMCP is a browser automation bridge that connects AI tools to a live browser session through a local proxy… browser-use/browser-harness — This project is an automation framework that connects large language models to web browsers via the Chrome DevTools… segmentio/nightmare — Nightmare is an Electron-based browser automation library and headless browser controller. It provides the… mastra-ai/mastra — Mastra is an orchestration framework designed for building, deploying, and managing autonomous AI agents and… knockoutez/wigolo — Wigolo is a local-first web research toolkit for AI agents. It runs as a Model Context Protocol server that gives…

Projects sharing features with Web Access

These projects share indexed features with Web Access. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
  • garrytan/gstackgarrytan avatar

    garrytan/gstack

    140,647View on GitHub↗

    gstack is an AI agent framework and development workflow system designed to automate the software development lifecycle. It coordinates specialized AI personas to manage tasks across product design, engineering management, and quality assurance, transforming product intent into technical specifications and final releases. The project is distinguished by its deep integration of headless browser automation and semantic code memory. It utilizes a persistent Chromium daemon for web scraping and visual auditing, and implements a searchable knowledge base that logs architectural decisions and repos

    TypeScript
    View on GitHub↗140,647
  • browsermcp/mcpBrowserMCP avatar

    BrowserMCP/mcp

    5,817View on GitHub↗

    BrowserMCP is a browser automation bridge that connects AI tools to a live browser session through a local proxy server. It implements a standardized protocol for sending commands like click, type, and navigate to a real browser instance running on the user's machine, while keeping all browsing data on the device. The project distinguishes itself by preserving user sessions and fingerprints across automation tasks. It attaches to the user's existing browser profile to maintain cookies, logins, and authentication state, and uses the real browser's user agent, viewport, and extension context to

    TypeScriptbrowser-automationbrowser-extensionclaude
    View on GitHub↗5,817
  • browser-use/browser-harnessbrowser-use avatar

    browser-use/browser-harness

    15,265View on GitHub↗

    This project is an automation framework that connects large language models to web browsers via the Chrome DevTools Protocol for autonomous task execution. It functions as a bridge between intelligent agents and browser engines, allowing for the direct control of browser sessions and profiles. The framework features a self-healing agent capable of generating and executing custom scripts during runtime to resolve failures and optimize browser tasks. It supports stealthy deployment through the use of integrated proxies and captcha solvers to bypass bot detection and security mitigations. The s

    Pythonai-agentbrowser-agentbrowser-automation
    View on GitHub↗15,265
  • segmentio/nightmaresegmentio avatar

    segmentio/nightmare

    19,775View on GitHub↗

    Nightmare is an Electron-based browser automation library and headless browser controller. It provides the infrastructure to programmatically navigate web pages, interact with DOM elements, and execute JavaScript within a background browser instance. The project distinguishes itself by integrating a full Chromium instance within an Electron shell, allowing for the management of browser sessions, network proxy settings, and persistent storage partitions. It enables the capture of page states as PNG screenshots, PDF documents, or HTML files. The tool covers a broad range of capabilities includ

    JavaScript
    View on GitHub↗19,775
Compare all 30 related projects→