This project addresses a core limitation of AI agents interacting with web browsers: state synchronization. By forking Chromium and freezing execution after each agent action, it provides a real-time, accurate state (screenshot, events) back to the agent, enabling more reliable multimodal interaction. The maker claims high benchmark scores with Anthropic's Opus 4.6.
Reach out to AI engineers and solo founders building agents through communities focused on MCP (Multi-Modal Chat Protocol) and agent development, highlighting how ABP can significantly improve agent reliability and testability by providing accurate real-time browser state.
Hi HN, I forked chromium and built agent-browser-protocol (ABP) after noticing that most browser-agent failures aren’t really about the model misunderstanding the page. Instead, the problem is that the model is reasoning from a stale state. ABP is designed to keep the acting agent synchronized with the browser at every step. After each action (click, type, etc), it freezes JavaScript execution and rendering, then captures the resulting state. It also compiles the notable events that occurred during that action loop, such as navigation, file pickers, permission prompts, alerts, and downloads, and sends that along with a screenshot of the frozen page state back to the agent. The result is that browser interaction starts to feel more like a multimodal chat loop. The agent takes an action, gets back a fresh visual state and a structured summary of what happened, then decides what to do next from there. That fits much better with how LLMs already work. A few common browser-use failures ABP helps eliminate: * A modal appears after the last Playwright screenshot and blocks the input the agent was about to use * Dynamic filters cause the page to reflow between steps * An autocomplete dropd