Understand this story
turbo-browser-mcp has been released, providing 10 batch browser automation MCP tools. The server is built on Playwright and consolidates 25-33 fine-grained browser operations into 10 batch tools, and provides incremental DOM snapshots, single-round React/Shadow DOM/iframe form filling, direct disk file uploads, and a submission guard that is enabled by default. The author ran benchmarks on local headless Chromium using tiktoken 0.14.0's o200k_base tokenizer: a snapshot of a 140-button page dropped from 3,568 tokens to 1,135 tokens, a 10-field form dropped from 11 rounds and 574 ms to 1 round and 82 ms, and a 48 KiB PDF upload dropped from 44,728 tokens of Base64 to an 85-token disk path; the static tool schema is 3,010 tokens, 23.2% fewer than @playwright/mcp and 55.3% fewer than chrome-devtools-mcp. This news is currently featured on Hacker News.
AI-generated from public reports
Latest development · Report
turbo-browser-mcp released: 10 batch tools compress Playwright MCP's tokens and turnsDevelopments
1 confirmed developments
turbo-browser-mcp provides batch browser automation MCP tools
Source:English text is being prepared.
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 15
Hacker News Trending
turbo-browser-mcp released: 10 batch tools compress Playwright MCP's tokens and turns
turbo-browser-mcp is a Playwright-based MCP server that consolidates 25-33 fine-grained browser operations into 10 batch tools, providing incremental DOM snapshots, single-turn React/Shadow DOM/iframe form filling, direct disk file uploads, and a submit guard that is enabled by default. The author ran benchmarks on a local headless Chromium using the o200k_base tokenizer from tiktoken 0.14.0: a snapshot of a 140-button page dropped from 3,568 tokens to 1,135 tokens, a 10-field form went from 11 turns and 574 ms to 1 turn and 82 ms, and a 48 KiB PDF upload went from 44,728 tokens of Base64 to an 85-token disk path. The static tool schema is 3,010 tokens, 23.2% fewer than @playwright/mcp and 55.3% fewer than chrome-devtools-mcp.
AI Trends