Tootfinder

Cursor's recent experiment involved running hundreds of AI agents for nearly a week to build a web browser, writing 1M lines of code across 1,000 files (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2026/Jan/19/scaling-long-running-a…

Scaling long-running autonomous coding
Wilson Lin at Cursor has been doing some experiments to see how far you can push a large fleet of "autonomous" coding agents: This post describes what we've learned from …

I laughed for a second. Then I got very sad.
(link: https://bsky.app/profile/abstracttesseract.bsky.social/post/3mewrojqa422b )

Hands-on with Claude Cowork: looks well positioned to bring the powerful capabilities of Claude Code to a wider audience, but risks of prompt injections remain (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2026/Jan/12/claude-cowork/

First impressions of Claude Cowork, Anthropic’s general agent
New from Anthropic today is Claude Cowork, a “research preview” that they describe as “Claude Code for the rest of your work”. It’s currently available only to Max subscribers ($100 …

Spotlighting The World Factbook as We Bid a Fond Farewell (via) Somewhat devastating news today from CIA: ... (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2026/Feb/5/the-world-factbook/
http://www.memeorandum.com/260204/p136#a260204p136

OpenAI quietly adopted Anthropic's "skills" mechanism in ChatGPT and Codex; ChatGPT's skills include creating and modifying spreadsheets, docx files, and PDFs (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2025/Dec/12/openai-skills/

OpenAI are quietly adopting skills, now available in ChatGPT and Codex CLI
One of the things that most excited me about Anthropic’s new Skills mechanism back in October is how easy it looked for other platforms to implement. A skill is just …

GPT-5.2 models match GPT-5 and 5.1 with a 400K context window and 128K max output tokens, but have a newer knowledge cutoff of Aug. 31, 2025 vs. Sept. 30, 2024 (Simon Willison/Simon Willison's Newsletter)
https://simonw.substack.com/p/gpt-52-and-useful-patterns-for-…

GPT 5.2 and useful patterns for building HTML tools
Plus a YouTube video, a podcast appearance and more

Anthropic rolls out a fast mode for Claude Opus 4.6 in research preview, saying it offers the same model quality 2.5 times faster but costs six times more (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2026/Feb/7/claude-fast-mode/

A Claude user gets Claude 4.5 Opus to generate a 14,000-token document that Claude calls its "Soul overview"; an Anthropic staffer confirms its authenticity (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2025/Dec/2/claude-soul-document/…

Some 2025 takeaways in LLMs: reasoning as a signature feature, coding agents were useful, subscriptions hit $200/month, and Chinese open-weight models impressed (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2025/Dec/31/the-year-in-llms/

2025: The year in LLMs
This is the third in my annual series reviewing everything that happened in the LLM space over the past 12 months. For previous years see Stuff we figured out about …

A look at Moltbook, a social network where OpenClaw assistants interact autonomously, as they discuss consciousness and identity, technical tips, and more (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2026/Jan/30/moltbook/

Moltbook is the most interesting place on the internet right now
The hottest project in AI right now is Clawdbot, renamed to Moltbot, renamed to OpenClaw. It’s an open source implementation of the digital personal assistant pattern, built by Peter Steinberger …

Anthropic prices Claude Opus 4.5 at $5/1M input and $25/1M output tokens, much cheaper than Opus 4.1 at $15/$75 but still pricier than GPT-5.1 and Gemini 3 Pro (Simon Willison/Simon Willison's Weblog)
https://simonwillison.net/2025/Nov/24/claude-opus/

Claude Opus 4.5, and why evaluating new LLMs is increasingly difficult
Anthropic released Claude Opus 4.5 this morning, which they call “best model in the world for coding, agents, and computer use”. This is their attempt to retake the crown for …