TideSurf
Browser use agents tend to prefer the models' native multimodality over concrete source, and, even if they do, they still tend to take too much context to even barely function. I was running into this problem when using LLM Agents; Then I came up with an idea. What if I can just... send the rendered DOM to the agent, but with markdown-like compression? Turns out, it works! It reduces token consumption by thirty-two times on GitHub (vs. raw DOM), at least according to my experiments, while only taking ~30ms to parse. Also, it comes with 18 tools for LLMs to work interactively with pages, and…
TideSurf launched on March 19, 2026 with 3 votes, #3,303 of 23,380 launches that month and more than 50% of that year's launches.
What it does
Browser use agents tend to prefer the models' native multimodality over concrete source, and, even if they do, they still tend to take too much context to even barely function. Then I came up with an idea. What if I can just... send the rendered DOM to the agent, but compressed? And, that's what TideSurf is. It reduces token consumption by thirty-two times on GitHub (vs. raw DOM), at least according to my experiments, while only taking ~30ms to parse. Also comes w/ 18 interactive tools.
TideSurf gives agents a stateful Chromium CLI, compact page text, and snapshot-scoped action IDs.
Run browser tools from the shell. TideSurf keeps Chromium state between commands and turns the live DOM into compact text with current action handles. Actual TideSurf output: wrapper noise disappears. Navigation, form controls, visible copy, and action handles stay. <div class="header"> <nav role="navigation"> <ul class="nav-list"> <a href="/">Home</a> <a href="/about">About</a> </ul> </nav> </div> <form action="/search"> <div class="input-wrapper"> <input placeholder="Search..." /> <button>Search</button> </div> </form> 5,348 separate fixture benchmark 1,218 e-commerce source HTML → TideSurf text TideSurf text after # Example Search > example.com/search | 0/1200 800vh NAV [L1] (/) Home…from tidesurf.org
In the maker’s words, at launch
Browser use agents tend to prefer the models' native multimodality over concrete source, and, even if they do, they still tend to take too much context to even barely function. I was running into this problem when using LLM Agents; Then I came up with an idea. What if I can just... send the rendered DOM to the agent…
Does a similar job
all alternatives →
Convert HTML DOM to semantic markdown for use in LLMs2024 · github.com · ▲146DOM to Semantic-Markdown for use with LLMs. Contribute to romansky/dom-to-semantic-markdown development by creating an account on GitHub.
- DRDontBeEvil.rip: Search, for developers (API, expressions, CLI)2022 · ▲269
I'd like to invite everyone to try out DontBeEvil.rip, an experimental search engine for developers. tl;dr $ alias rip="curl -G -H 'Accept: text/plain' --url https://dontbeevil.rip/search --data-urlencode " $ rip 'q=Heartbleed bug' DontBeEvil.rip is a year long experiment to see if a small team can build a developer-focused search engine that is self-sustaining on $10 monthly subscriptions. It works by only indexing high-quality resources that are relevant to developers. You won't get useless listicles because we'll never crawl them. Relevant urls are harvested from HN,…
Code ModeNov 2025 · github.com · ▲144I’ve been testing a simple idea inspired by Apple/Cloudflare/Anthropic: LLMs are far better at writing a small program than coordinating multiple tool calls. So instead of giving the model 10+ tools, I exposed one: a TypeScript sandbox with access to the same interfaces. The model writes a script → it runs once → done. What changed - +68% reduction in token use - No multi-step drift or retries - Local models (Llama 3.1 8B / Phi-3) became much more reliable Repo: https://github.com/universal-tool-calling-protocol/code-mode Curious whether others have seen…
AgentReadyFeb 2026 · agentready.cloud · ▲118Compression API that reduces prompt size by 40-60% before sending to your LLM. Your API key never leaves your environment.
Context GatewayMar 2026 · github.com · ▲220We built an open-source proxy that sits between coding agents (Claude Code, OpenClaw, etc.) and the LLM, compressing tool outputs before they enter the context window. Demo: https://www.youtube.com/watch?v=-vFZ6MPrwjw#t=9s. Motivation: Agents are terrible at managing context. A single file read or grep can dump thousands of tokens into the window, most of it noise. This isn't just expensive — it actively degrades quality. Long-context benchmarks consistently show steep accuracy drops as context grows (OpenAI's GPT-5.4 eval goes from 97.2% at 32k to 36.6% at 1M…
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 22d ago · stillwet.art
FastRouter.ai▲364Route requests to the right LLM for cost, latency & quality
AI · 23d ago · fastrouter.ai
Source: Product Hunt launch ↗ · launch text quoted in part
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
Littlebird▲709Littlebird remembers everything across your meetings, messages, docs, and apps, then takes action so you can recall any detail and never lose context.
AI · Mar 2026 · littlebird.ai
