LLM cliché highlighter catches 'no fluff, no filler' writing
The LLM cliché highlighter is a browser tool that flags ten common patterns typical of LLM-generated writing, such as 'no fluff, no filler, no jargon' constructions.
What it is
A small web app at tools.simonwillison.net built by Simon Willison after getting fed up reading yet another article full of stock LLM phrasing.
What it does
It scans pasted text and highlights ten recurring patterns typical of LLM-generated prose, then reports match counts and flagged sentences so you can see exactly where the clichés live.
Why it matters
If you're editing AI-drafted docs, PR descriptions, or blog posts before publishing, this catches the tells a human reviewer might skim past, which is exactly the kind of pattern this pipeline's own voice guidelines try to avoid.
How to use it
Paste any draft into the tool at the source URL and review the flagged sentences before you ship the text.
Inkling: Thinking Machines Lab's first open-weights model
Inkling is Thinking Machines Lab's Apache-2.0 licensed Mixture-of-Experts model with 975B total and 41B active parameters, trained on 45 trillion multimodal tokens.
Kimi K3: 2.8 Trillion Parameters, Weights Coming July 27
Moonshot AI announced Kimi K3, a 2.8 trillion parameter model available now via API, with open weights promised by July 27, 2026.
Claude Code 2.1.214 closes a PowerShell permission bypass
Claude Code 2.1.214 fixes four permission-check bugs, including nested-directory allow rules matching too broadly and a bypass in Windows PowerShell 5.1 sessions.