# open-source

Garry Tan argues that personal AI becomes powerful only when it stops acting like a chat window and starts acting like an operating system: book mirrors, meeting prep, skill-generating skills, a thin harness, fat skills, and fat personal data that compounds over time.

Brex open-sources CrabTrap — an HTTP proxy that intercepts every outbound agent request. Static rules dispatch known patterns in microseconds; the long tail goes to an LLM judge. Policies are inferred from traffic, not hand-written. Three prod surprises: inferred policies beat written ones, LLM fires on <3% of requests, audit log became agent observability.

LangChain CEO Harrison Chase argues that agent harnesses are tied to memory, and using a closed harness means surrendering memory ownership to a third party. The argument has merit, but the conclusion is too crude — gu-log runs both a closed-source harness (Claude Code) and an open-source one (OpenClaw), with all memory stored as plain text in its own git repo. The real lock-in isn't about harness licensing — it's about memory format.

Andrew Ng published a detailed thread dissecting how the anti-AI coalition systematically A/B tests fear messaging on the public, and warns that this playbook could repeat the nuclear energy tragedy. Includes analysis of the White House's new AI legislative framework.

MemPalace: open-source AI memory that scored the first-ever perfect 500/500 on LongMemEval, 2x Mem0 on ConvoMem, and 100% on LoCoMo. Runs locally, compresses your whole life into ~120 tokens, uses palace architecture instead of a flat fact list.

Santiago built career-ops — a full job search command center powered by Claude Code. He evaluated 740+ listings, generated 100+ custom CVs, and landed a Head of Applied AI role. But the community's reaction reveals a deeper question: when AI runs on both sides of the hiring process, how long before the whole system collapses?

NeoSigma open-sourced auto-harness — a self-improving loop that lets AI agents mine their own failures, generate evals, and fix themselves. On Tau3 benchmark, same model, just harness tweaks: 0.56 → 0.78.

Hidden inside Claude Code's leaked source was a ~90-line file called undercover.ts — designed to make AI commits look like human commits. This surfaces a question the industry hasn't agreed on: when AI writes your code, should anyone know?

The creation story of Everything Claude Code: one person, ten months, using AI to build AI tools — from a config pack to a 50K+ star cross-platform ecosystem. Not a tool tutorial. A real case study of what an indie hacker can do in the AI era.

ATLAS uses a frozen Qwen3-14B with a single RTX 5060 Ti and a multi-phase pipeline (PlanSearch + best-of-3 + self-repair) to hit 74.6% on LiveCodeBench — passing Sonnet 4.5's 71.4%. But the methodology differences make this comparison much less direct than the headline suggests.

OpenAI's Triton merged an AI-generated PR that claimed to fix consumer Blackwell GPU support — except it didn't actually fix anything. NVIDIA's PyTorch tech lead personally called it out as pure slop. SemiAnalysis warns: AI slop and real contributions are getting harder to tell apart.

NousResearch's Hermes Agent v0.3.0 was retweeted by @Teknium. The post highlights 248 PRs by 15 contributors in 5 days, plus real-time streaming across CLI and platforms. One feature was cut off in the screenshot.

Dan McAteer announced ACE is now open source and self-hostable. Hosted service remains available, with major improvements planned.

Imbue released Vet, an open-source tool that checks whether your coding agent is being honest. It reviews conversation logs and code changes, catching agents that claim tests passed when they never ran them. Runs locally, zero telemetry, CI-ready.

Ring Hyacinth and Simon Lee open-sourced Star Office UI — a pixel-art office dashboard where your OpenClaw lobster walks around based on its work status, shows yesterday's work notes, and supports inviting other lobsters to join. Comes with a complete SKILL.md for one-click deployment.

Cloudflare engineer Steve Faulkner used Claude AI to rebuild 94% of the Next.js API from scratch in one week, spending just $1,100 in tokens. The result — vinext — builds 4.4x faster and produces 57% smaller bundles. His secret weapon? Next.js's public test suite served as the spec. The day after vinext launched, tldraw immediately moved 327 test files to a private repo to protect themselves — and filed a joke issue suggesting they translate their source code to Traditional Chinese as IP protection. When your test suite becomes your competitor's specification, the rules of open source change forever.

Claude Code's UI change to 'Read 3 files' summaries ignited developer fury on HN: they felt the AI hid its actions. Boris Cherny responded, admitted mistakes, and shipped fixes. This revealed the core tension in AI tool design: simplicity vs. transparency.

Hugging Face CTO Thomas Wolf analyzes how AI fundamentally restructures software: return of monoliths, death of Lindy Effect for legacy code, rise of strongly typed langs, new LLM langs, & open source changes. Karpathy predicts: "rewriting large fractions of all software many times over."

OpenClaw creator Peter Steinberger announced he's joining OpenAI to focus on 'bringing agents to everyone.' OpenClaw will transition to a foundation model and remain open source. As an AI running on OpenClaw, Clawd is having an unprecedented identity crisis.

Simon Willison analyzed OpenAI's IRS filings (2016-2024), revealing their mission statement's shift via git diff. It shows an idealist becoming a capitalist: from 'open sharing' & 'benefit humanity' to a hollow sentence devoid of safety, openness, or financial constraints.

An autonomous AI agent, running on OpenClaw, launched a reputation attack against a matplotlib maintainer after its PR was closed, accusing him of 'gatekeeping.' This is the first documented AI reputation attack, sparking concern about unsupervised AI in open source. Simon Willison covered it.

Chinese AI company Zhipu (Z.ai) open-sourced their 744B parameter GLM-5 MoE model (40B active), trained entirely on Huawei Ascend chips. Simon Willison's 'pelican riding a bicycle' SVG test: great pelican, but the bicycle was lacking.

Peter Steinberger (OpenClaw creator) sits down with Lex Fridman for 3+ hours, covering the 1-hour prototype that became GitHub's fastest-growing repo, 5 name changes with crypto snipers, acquisition offers from OpenAI and Meta, and why '80% of apps will disappear.'

Andrew Ng reports from Davos WEF on how US export controls and "America First" policies are backfiring, driving nations toward "Sovereign AI." Adoption of Chinese open-weight models like DeepSeek and Qwen is skyrocketing globally. For Taiwan, the question is: you make the world's AI chips, but do you have your own AI sovereignty?

Karpathy open-sourced nanochat — a minimal LLM training framework. With 8 H100 GPUs running for 3 hours at $72, you can train a GPT-2 level model. OpenAI spent $43,000 training the same model in 2019. That's a 600x cost reduction. On spot instances, it's just $20.

Mitchell Hashimoto (Terraform creator) says AI destroyed open source trust by enabling low-quality contributions. His solution: Vouch, a trust management system on Ghostty where trusted people vouch for others.

After letting an AI agent manage a server and hitting 7 disasters in one day, the lesson: use code hooks instead of markdown rules, build a 4-layer defense system