frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Show HN: Channel Surfer – Watch YouTube like it’s cable TV

https://channelsurfer.tv
243•kilroy123•2d ago•102 comments

Show HN: Context Gateway – Compress agent context before it hits the LLM

https://github.com/Compresr-ai/Context-Gateway
36•ivzak•3h ago•25 comments

Show HN: ShellSelf – A Developer Portfolio That Feels Like Home

https://www.shellself.com/
5•truetaurus•24m ago•1 comments

Show HN: Anthrology – Time-Traveling Radio

https://anthrology.site/
3•airstrike•45m ago•0 comments

Show HN: Mutate – free inline text replacement for Mac

https://github.com/robert-v/Mutate-public
2•rob3rth•50m ago•1 comments

Show HN: Svglib a SVG parser and renderer for Windows

https://github.com/bibhas2/svglib
8•leopoldj•3d ago•1 comments

Show HN: What was the world listening to? Music charts, 20 countries (1940–2025)

https://88mph.fm/
96•matteocantiello•3d ago•43 comments

Show HN: DJX – Convention over Configuration for Django (Rails-Inspired CLI)

3•RedsonNgwira•1h ago•1 comments

Show HN: Tiny macOS app that adds a facecam bubble to screen recordings

https://github.com/backnotprop/CamBubble
5•ramoz•1h ago•0 comments

Show HN: Mjmx – render mjml using JSX

https://mjmx.dev/
3•skwee357•2h ago•0 comments

Show HN: An addendum to the Agile Manifesto for the AI era

https://github.com/brackishman/Agile-Manifesto-AI-Addendum
2•brackishman•10m ago•0 comments

Show HN: AgentLog – a lightweight event bus for AI agents using JSONL logs

https://github.com/sumant1122/agentlog
4•paperplaneflyr•2h ago•0 comments

Show HN: Execute local LLM prompts in remote SSH shell sessions

3•smudgy3746•2h ago•2 comments

Show HN: Axe – A 12MB binary that replaces your AI framework

https://github.com/jrswab/axe
206•jrswab•1d ago•118 comments

Show HN: AI milestone verification for construction using AWS

https://builder.aws.com
2•eugenelotsu•3h ago•0 comments

Show HN: RepoCrunch – CLI to analyze GitHub repos

https://github.com/kimwwk/repocrunch
2•chillkim•3h ago•1 comments

Show HN: OpenClaw docs in Japanese, now open source

https://openclawdoc.org
2•mixfox•4h ago•0 comments

Show HN: OneCLI – Vault for AI Agents in Rust

https://github.com/onecli/onecli
155•guyb3•1d ago•49 comments

Show HN: Rudel – Claude Code Session Analytics

https://github.com/obsessiondb/rudel
139•keks0r•1d ago•83 comments

Show HN: Understudy – Teach a desktop agent by demonstrating a task once

https://github.com/understudy-ai/understudy
110•bayes-song•1d ago•41 comments

Show HN: Mesa – A collaborative canvas IDE built for agent-first development

https://www.getmesa.dev/
14•visekr•5h ago•0 comments

Show HN: s@: decentralized social networking over static sites

http://satproto.org/
408•remywang•1d ago•218 comments

Show HN: OpenClaw-class agents on ESP32 (and the IDE that makes it possible)

https://pycoclaw.com/
27•pycoclaw•23h ago•1 comments

Show HN: Web-based ANSI art viewer

https://sure.is/ansi/
27•lubujackson•3d ago•7 comments

Show HN: Open-source browser for AI agents

https://github.com/theredsix/agent-browser-protocol
151•theredsix•2d ago•52 comments

Show HN: 724claw.icu – Anonymous vent wall for "shrimp workers" grinding 7×24

2•867762462f•3h ago•0 comments

Show HN: Vanilla JavaScript refinery simulator built to explain job to my kids

https://fuelingcuriosity.com/game.html
125•fuelingcurious•2d ago•48 comments

Show HN: I built a tool that watches webpages and exposes changes as RSS

https://sitespy.app
313•vkuprin•2d ago•76 comments

Show HN: Autoresearch@home

https://www.ensue-network.ai/autoresearch
76•austinbaggio•1d ago•19 comments

Show HN: Global Maritime Chokepoints

https://ryanshook.org/chokepoints/
15•RyanShook•16h ago•5 comments
Open in hackernews

Show HN: Context Gateway – Compress agent context before it hits the LLM

https://github.com/Compresr-ai/Context-Gateway
36•ivzak•3h ago
We 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 https://openai.com/index/introducing-gpt-5-4/).

Our solution uses small language models (SLMs): we look at model internals and train classifiers to detect which parts of the context carry the most signal. When a tool returns output, we compress it conditioned on the intent of the tool call—so if the agent called grep looking for error handling patterns, the SLM keeps the relevant matches and strips the rest.

If the model later needs something we removed, it calls expand() to fetch the original output. We also do background compaction at 85% window capacity and lazy-load tool descriptions so the model only sees tools relevant to the current step.

The proxy also gives you spending caps, a dashboard for tracking running and past sessions, and Slack pings when an agent is sitting there waiting on you.

Repo is here: https://github.com/Compresr-ai/Context-Gateway. You can try it with:

  curl -fsSL https://compresr.ai/api/install | sh
Happy to go deep on any of it: the compression model, how the lazy tool loading works, or anything else about the gateway. Try it out and let us know how you like it!

Comments

verdverm•2h ago
I don't want some other tooling messing with my context. It's too important to leave to something that needs to optimize across many users, there by not being the best for my specifics.

The framework I use (ADK) already handles this, very low hanging fruit that should be a part of any framework, not something external. In ADK, this is a boolean you can turn on per tool or subagent, you can even decide turn by turn or based on any context you see fit by supplying a function.

YC over indexed on AI startups too early, not realizing how trivial these startup "products" are, more of a line item in the feature list of a mature agent framework.

I've also seen dozens of this same project submitted by the claws the led to our new rule addition this week. If your project can be vibe coded by dozens of people in mere hours...

thesiti92•2h ago
do you guys have any stats on how much faster this is than claude or codex's compression? claudes is super super slow, but codex feels like an acceptable amount of time? looks cool tho, ill have to try it out and see if it messes with outputs or not.
uaghazade•2h ago
ok, its great
esafak•2h ago
I can already prevent context pollution with subagents. How is this better?
root_axis•1h ago
Funny enough, Anthropic just went GA with 1m context claude that has supposedly solved the lost-in-the-middle problem.
SyneRyder•1h ago
Just for anyone else who hadn't seen the announcement yet, this Anthropic 1M context is now the same price as the previous 256K context - not the beta where Anthropic charged extra for the 1M window:

https://x.com/claudeai/status/2032509548297343196

As for retrieval, the post shows Opus 4.6 at 78.3% needle retrieval success in 1M window (compared with 91.9% in 256K), and Sonnet 4.6 at 65.1% needle retrieval in 1M (compared with 90.6% in 256K).

siva7•1h ago
now that's major news
BloondAndDoom•58m ago
In addition to context rot, cost matters, I think lots of people use toke compression tools for that not because of context rot
hinkley•32m ago
From a determinism standpoint it might be better for the rot to occur at ingest rather than arbitrarily five questions later.
kuboble•1h ago
I wonder what is the business model.

It seems like the tool to solve the problem that won't last longer than couple of months and is something that e.g. claude code can and probably will tackle themselves soon.

kennywinker•34m ago
Business model is: Get acquired
teaearlgraycold•15m ago
Could also be selling data to model distillers.
tontinton•1h ago
Is it similar to rtk? Where the output of tool calls is compressed? Or does it actively compress your history once in a while?

If it's the latter, then users will pay for the entire history of tokens since the change uncached: https://platform.claude.com/docs/en/build-with-claude/prompt...

How is this better?

BloondAndDoom•59m ago
This is a bit more akin to distill - https://github.com/samuelfaj/distill

Advantage of SML in between some outputs cannot be compressed without losing context, so a small model does that job. It works but most of these solutions still have some tradeoff in real world applications.

thebeas•35m ago
We do both:

We compress tool outputs at each step, so the cache isn't broken during the run. Once we hit the 85% context-window limit, we preemptively trigger a summarization step and load that when the context-window fills up.

lambdaone•1h ago
This company sounds like it has months to live, or until the VC money runs out at most. If this idea is good, Anthropic et. al. will roll it into their own product, eliminating any purpose for it to exist as an independent product. And if it isn't any good, the company won't get traction.
sethcronin•50m ago
I guess I'm skeptical that this actually improves performance. I'm worried that the middle man, the tool outputs, can strip useful context that the agent actually needs to diagnose.
thebeas•40m ago
That's why give the chance to the model to call expand() in case if it needs more context. We know it's counterintuitive, so we will add the benchmarks to the repo soon.

Given our observations, the performance depends on the task and the model itself, most visible on long-running tasks

fcarraldo•25m ago
How does the model know it needs more context?