GOOD CODE. GREAT DISCOVERIES.

Your next
rabbit hole.

Open-source projects. Clever little tools. Big ideas.
Unearthed from the Hacker News community.

YCommunity signal. Human curiosity.No paid placements.
2,872unique repositories
4,114Show HN stories¹
3,330GitHub links¹
July 2026 snapshot

AI & agents✳

950 repos
↵
Ranked by HN points, not GitHub stars
Show HN
valmishq /

valmis

I built an open-source alternative to Claude Cowork

AI & agentsLicense unchecked
30#091
Show HN
dogtorjonah /

context-warp-drive

Context Warp Drive – Deterministic context folding for AI agents

AI & agentsLicense unchecked
30#092
cloudflare /

agentic-inbox

Self-hosted email client with an AI agent, running on Cloudflare Workers

AI & agentsLicense unchecked
30#093
Show HN
ToolJet /

ActionRail

ActionRail, verify agent actions using live system of records

AI & agentsLicense unchecked
30#094
Show HN
innovatorved /

sayitdev

SayItDev LLM and Speech capabilities with 0 dependencies and no models

AI & agentsLicense unchecked
30#095
Robbyant /

lingbot-map

Lingbot-map: A 3D foundation model for reconstructing scenes from streaming data

AI & agentsLicense unchecked
30#096
Show HN
fresswolf /

Slopera

Slopera, a browser that hallucinates every page with an LLM

AI & agentsLicense unchecked
30#097
Show HN
carbon-re /

skeights

Skeights – Serialize sklearn models to safetensors and JSON, no pickle

AI & agentsLicense unchecked
30#098
Show HN
amanharshx /

ultralytics-mcp

Control YOLO Training and Datasets from Claude/Cursor via MCP

AI & agentsLicense unchecked
30#099
Show HN
mihirahuja1 /

agentnestOSS

AgentNest, self-hosted sandboxes for AI agents

AI & agentsLicense unchecked
30#100
Show HN
projectargus-cc /

libargus.cc

Low-latency local LLM runner via OpenJDK Panama FFM (Java 22)

AI & agentsLicense unchecked
30#101
Show HN
mazen160 /

backlog

Backlog – tasks and contexts manager for AI coding agents

AI & agentsLicense unchecked
30#102
tencentcloud /

CubeSandbox

Secure, Fast, and Extensible Sandbox Runtime for AI Agents

AI & agentsLicense unchecked
30#103
raiyanyahya /

llmaker

Llmaker – spin up a working LLM app from a single prompt, right in your terminal

AI & agentsLicense unchecked
30#104
Show HN
samvallad33 /

vestige

Vestige – A queryless memory layer for multi agent coding fleets

AI & agentsLicense unchecked
23#105
wladimiravila /

esp32s3-distributed-ai

HN: I ran a 56M parameter LLM across three ESP32-S3 boards via ESP-NOW

AI & agentsLicense unchecked
22#106
Show HN
theoriclabs /

agent-convert

You can continue a Claude session in Codex, and vice versa

AI & agentsLicense unchecked
22#107
wonsukchoi /

domain-experts

Domain experts: All human experts into AI agents

AI & agentsLicense unchecked
22#108

Less noise. More “oh, that’s cool.”

An independent index of things the Hacker News community shares.