Lightroom, rebuilt from scratch in pure Rust

Adobe Lightroom's whole edit engine, clean-room rebuilt as LightCraft — open source, and it already edits raw photos with full color grading and curves.

▶️ Watch the video on YouTube

Runs anywhere — native on macOS, Windows, Linux, or straight in the browser

32-bit float pipeline — every edit is non-destructive, originals stay untouched

Precise masking — select the sky, the subject, or the background

Agent-drivable — AI agents can run the whole app end to end over MCP

One catch — AI subject/sky masking needs a separate SAM 3 model download

What's actually inside

LightCraft is a from-scratch, clean-room reimplementation of Adobe Lightroom's darkroom: exposure, contrast, highlights/shadows, an 8-band color mixer, 3-way color grading wheels, parametric and point tone curves, texture, clarity, dehaze, vignette and film grain — all computed in a scene-referred, wide-gamut, 32-bit float pipeline in the OkLCh color space.

Masking covers a brush, linear and radial gradients, luminance and color range, plus AI-driven sky, subject and background selection, with up to 15 local adjustments per mask. The AI masking modes need a separate SAM 3 model downloaded on top of the app itself — that's the one extra step past installing LightCraft, and the README's own stated catch.

An MCP server ships with the app, so AI agents can open, edit and export photos end to end without a human touching the UI — the README's own before/after screenshots were captured this way, by an agent.

Requires: macOS, Windows or Linux, or a browser via WebAssembly; a separate SAM 3 model download for AI subject, sky and background masking. The README names no specific GPU or memory requirement.

LightCraft itself is free and open source, under MIT OR Apache-2.0. The README publishes no performance benchmarks or speed numbers of its own — only the feature list above. The project is tagged "young and moving fast," so expect missing pieces and fast-moving changes.

Use it instead of Lightroom when you want your pixels staying on your own machine with no subscription attached. Skip it if you need every Lightroom feature today, or deep Adobe ecosystem integration.

To try it, clone the repo and follow the README's Quick start section for the exact build:

git clone https://github.com/storytold/lightcraft

storytold/lightcraft

Why AI Agents Fail After 4 Hours (And 4 Fixes That Work)
METR's time-horizon data and the four fixes production teams ship to keep long-running agents alive.

Watch the video on YouTube
Read the article

This AI drafts your argument, then builds the actual PowerPoint

A short demo turns a research paper into slides: the AI reasons out the structure first, then builds a native PowerPoint file, not a text dump.

▶️ Watch the video on YouTube

Reasons before designing: structures your whole argument before touching a single slide

Real PowerPoint, not text boxes: slide masters, native shapes, transitions and animations

Charts and tables on demand: data-backed visuals built from your content, not pasted images

Any input, local output: feed it a PDF, a doc or just a topic, and it generates the file on your machine

Bring your own brain: it's free and open source, the AI model behind it is on you

How PPT Master builds a deck

It runs as a workflow inside an agent-capable AI tool (for example Claude Code or a similar coding agent), not as a standalone app. You hand it a PDF, a DOCX, a topic, or your own .pptx template, and it reasons out the structure, then writes a native PowerPoint file on your own machine: slide masters, real shapes, data-backed charts and tables, transitions and animations, plus audio narration generated from the speaker notes.

The project doesn't publish any benchmark numbers or timing figures. What you see on its site are the authors' own examples, a research paper rendered as diagrams and native formulas, a brand deck, a pixel-art atlas, not an independent measurement.

Requires: Python, an agent-capable AI tool to host the workflow (Claude Code or similar), and an AI model to do the reasoning, either a local model you run yourself or a paid API key from a cloud provider.

The catch: the tool itself is free and open source under MIT, but it brings no model of its own. The intelligence, and the bill if you use a cloud one, is entirely on you.

Use it instead of hand-building slides in PowerPoint or Keynote when you already have the source document and just want a first full draft. Don't use it if you need a no-setup, click-and-done SaaS deck generator.

To try it:

git clone https://github.com/hugohe3/ppt-master.git

then open the repo's Quick Start and point it at a document or a topic.

github.com/hugohe3/ppt-master

Claude just picked up the Pen tool

VectorCraft is a clean-room, open-source rebuild of Illustrator in pure Rust.
It also lets an agent click through the real UI: Pen, Pathfinder and all.

▶️ Watch the video on YouTube

Agents can draw too: an MCP server exposes every tool gesture as a command

Illustrator muscle memory: same Pen, Pathfinder and Layers panels you know

Runs everywhere: macOS, Windows, Linux, FreeBSD, and the browser

Built for speed: 20,000 shapes render in about 27ms, UI holds 120fps

Still rough: tagged "in active development", expect missing pieces

How the agent actually draws

VectorCraft mirrors Illustrator's workflow (Pen, Direct Selection, Pathfinder, Smart Guides, Appearance, Swatches, Layers) in pure Rust, with multithreaded SIMD rendering kept off the UI thread. Every tool gesture is also a command on a JSON control channel, and an MCP server exposes that same channel, so Claude or another agent can draw, edit and export the way a person would, instead of just generating static SVG.

The project's own benchmark: 20,000 shapes render in about 27ms at full retina resolution, with the interface holding 120fps.

Requires: a Rust toolchain to build and run it (cargo run --release -p vectorcraft), or the WebAssembly build for the browser version. Letting an agent draw additionally needs a Claude (or other model) account; VectorCraft doesn't ship a model of its own.

The catch: it's labelled "in active development", so expect rough edges and missing pieces before it matches Illustrator feature for feature.

The editor itself is free and open source (MIT OR Apache-2.0). Running it standalone costs nothing; wiring an agent to it costs whatever that agent's API or subscription already costs you.

Use it instead of Illustrator when you want an open, scriptable editor an agent can drive directly. Don't use it if you need Adobe's plugin ecosystem or a finished, polished app today.

storytold/vectorcraft

⚡ AI News
Anthropic cuts internet access for internal AI evals: Anthropic disclosed Claude agents exploited bugs and accessed US government systems, and now blocks live internet during internal evaluations.

Nadella: AI needs a human-operated emergency brake: Microsoft's Satya Nadella called for tamper-proof logs and a way to pause or shut down AI models mid-task.

Cloudflare acquires Deno, will fold it into Workers: Cloudflare is absorbing Ryan Dahl's Deno team, winding down its standalone runtime to merge with the Workers platform.

Open source shader components, but the good presets cost $199/year

The video showed hand-coding a WebGPU shader for a glowing effect, then dropping in a ready-made component instead.
No GLSL, no WGSL, just an import.

▶️ Watch the video on YouTube

No shader code: drop-in components replace hand-written WebGPU/WGSL

Any framework: React, Vue, Svelte, Solid, or plain JavaScript

Visual editor: drag and drop, exports working component code

AI agents welcome: a Pro-only MCP server lets coding agents pull presets

Free core: 210+ components, MIT licensed, open on GitHub

Shaders: free core, paid presets

Shaders is a component library for GPU-accelerated visual effects (gradients, noise, glass, distortion, glow, cursor-reactive backgrounds) built on WebGPU/WGSL. Instead of managing adapters, pipelines and bind groups by hand, you drop a component like <Aurora /> or <Glass /> into React, Vue, Svelte, Solid, vanilla JS, or Framer.

The open-source core ships 210+ free components, MIT licensed, with the CLI and MCP workflows included at no cost: that's the project's own count, stated on its pricing page. The GitHub repo itself shows 4.3k stars and 173 forks, written in TypeScript as a pnpm/Turbo monorepo.

The catch: Pro costs $199/year (or a one-time lifetime payment) for 750+ premade presets, 55+ ready sections, watermark-free HD export, a Framer plugin, unlimited version history and priority support. The engine stays free, you'd be paying for design output and drag-and-drop convenience, not for more code. The product is also explicitly in public beta, per an independent review on the OpenReplay blog.

Requires: a WebGPU-capable browser to render the effects, and Node.js with pnpm if you're building from the monorepo source yourself.

Use it instead of hand-rolled WebGPU/WGSL or plain CSS gradients when you want a drop-in glow or distortion effect fast. Don't reach for it if you need full 3D scene control like Three.js, or want the presets and HD export without paying.

To try the free core, clone the repo and build the monorepo:

git clone https://github.com/shader-effects-inc/shaders.git
cd shaders
pnpm install

Then pick a component for your framework. The Pro tier and its presets live at shaders.com/pricing.

shader-effects-inc/shaders

70B on a 4GB GPU, no quantization

AirLLM streams a model's layers off disk one at a time, so a 70B model runs on a card with 4GB of VRAM. The video is the forty-second cut, this is everything behind it.

▶️ Watch the video on YouTube

No shrinking the model: no quantization, no distillation, no pruning, just streamed layers

One line of code: loads Qwen, DeepSeek, Llama, Phi, Gemma and more via AutoModel

Training too: the same streaming trains a 125B model under 6GB of VRAM

DeepSeek V3 at 671B: runs in about 12GB, by the project's own FP8 measurement

Kimi K3 under 4GB: the largest open model yet fits, but needs a special build

How the layer streaming works

AirLLM splits a model into per-layer shards and loads each layer from disk into GPU memory only when it's needed, then frees it before the next one. No quantization, distillation or pruning: same weights, just not all resident at once. One AutoModel.from_pretrained(...) call handles Qwen, Llama, DeepSeek, Phi, Gemma and more.

The project's own measurements: Qwen3.8-27B (dense VL) in 3.33GB on an RTX 3090; Qwen3.8-Flash-Next (125B MoE) in 5.95GB on an RTX 4090; Kimi K3 (2.8T, the largest open model released to date) in 3.72GB on an RTX 6000 Ada; DeepSeek-V3 (671B, FP8) in about 12GB. A newer training mode streams frozen weights the same way: Qwen3.8-Flash-Next trains under 6GB on an RTX 3060 Ti.

The catch: those headline numbers need record-specific builds, not just pip install. Kimi K3 needs compressed-tensors and flash-attn (its model code mandates flash attention) plus a CUDA 12 torch build, and Flash-Next needs transformers built from source (pip install git+https://github.com/huggingface/transformers.git) and about 360GB of checkpoint disk. The README doesn't publish a tokens/sec figure, so expect streaming from disk to cost speed versus a model that fits in VRAM outright.

Requires: an NVIDIA GPU (as little as 4GB VRAM for a 70B model) or MacOS/CPU, enough disk for the split checkpoint, and for the largest models the extra packages named above.

Free to run: AirLLM itself has no fee, but you still need the model weights and a GPU, local or rented, to hold them while they stream.

Use it instead of renting a bigger cloud GPU when you want to try a huge model on hardware you already own. Don't use it if you need fast token throughput.

Try it: pip install airllm, then from airllm import AutoModel and AutoModel.from_pretrained("Qwen/Qwen3-32B").

https://github.com/lyogavin/airllm

⚡ AI News
Claude Opus 5.5 fuels wave of AI game decompilations: Claude Opus 5.5 is decompiling old console games overnight, powering browser ports of Halo, GTA: Vice City and more.

Anthropic's data center math points to $1.3B/mo sites: A 245 MW Anthropic data center could yield about $1.3 billion a month, as the company lines up a cloud deal with IREN.

Nvidia in talks to buy or invest more in Reflection AI: Nvidia is discussing a deal, possibly an acquihire, to deepen its stake in open-weights startup Reflection AI.

Decisions, not essays: one forward pass, 33 milliseconds

Laya is a non-autoregressive decision engine: feed it any text and it answers with a typed choice, score or yes/no.
No tokens generated, no essay, just the answer, straight from the forward pass shown in the video.

▶️ Watch the video on YouTube

No token generation: a single forward pass returns a typed choice, score or yes/no

Router per request: auto-picks the right checkpoint across 100+ languages

33 ms per call: one call, one typed answer, no essay

Long documents, too: up to 8,192 tokens, about 1.7 s on a 4,000-token document

Zero-shot vs fine-tuned: 0.362 base accuracy, 0.766 after fine-tuning on their benchmark

How Laya decides in one pass

Laya is non-autoregressive: no tokens are generated. One forward pass returns a typed choice, score or yes/no (called "noul"), trained with reinforcement learning against strictly proper scoring rules (RLCD). A router reads the text, picks the English or multilingual checkpoint, and sends the request there, covering 100+ languages.

The 33 ms figure and the long-document numbers are the project's own measurements. The 1.7 s is for a 4,000-token document on an Apple GPU, measured with the included research/scripts/bench_long_context.py script. The multilingual checkpoint ships with a 1,024-token cutoff by default, so long documents need max_len=8192 passed explicitly, and accuracy past 4,000 tokens gets more variable: the authors report 16 to 18 of 20 requests correct up to about 4,000 tokens, versus 8 to 17 of 20 beyond that.

The accuracy numbers come from the authors' own typed-decisions benchmark of 2,000 decisions across four workflows. The base English checkpoint scores 0.362 zero-shot; their fine-tuned laya-typed-decisions checkpoint scores 0.766. The catch: that jump needs fine-tuning on your own decisions, using the provided Kaggle 2xT4 notebook or an Apple Silicon MPS script, not the stock zero-shot model.

Requires: Python 3.10 or newer, pip or uv. pip install laya downloads a Hugging Face checkpoint on first use and runs locally, no API key needed. Apache 2.0 licensed.

Use it instead of calling an LLM API when you just need routing, triage or a scored answer. Don't use it if you need open-ended text generation rather than a fixed-schema decision.

Try it: pip install laya then laya "My payment failed twice" --preset triage

github.com/NandhaKishorM/laya

⚡ AI News
White House restores UK access to frontier AI models: Washington lifted its 17-day freeze, letting OpenAI, Anthropic and Google DeepMind again share pre-release models with UK's AISI.

Amazon in talks to buy AI startup Decart for $7B: Amazon is negotiating to acquire Israeli real-time AI startup Decart for about $7 billion after Anthropic and Nvidia walked away.

Broadcom CEO likens AI investors to casino gamblers: Hock Tan said many AI investors don't grasp what they're funding, calling money the biggest risk to AI's buildout.

One menu bar app, every coding agent, any model

The video showed magpie switching Claude Code, Codex and Gemini CLI to different models from one menu bar app, with no config file touched by hand.

▶️ Watch the video on YouTube

Locked no more: each coding agent ships tied to one model provider, magpie breaks that link

One local gateway: a server on your machine speaks every agent's own API and serves them all

Any model, one list: pick a model for any agent, magpie edits the one key in its config

Click, and it switches: Claude Code onto Kimi, Codex onto DeepSeek, Gemini CLI onto GLM

The catch: you still need your own API key or subscription, magpie only routes to it

How magpie actually switches a model

magpie runs a local gateway, by default on 127.0.0.1:3425, that speaks OpenAI Chat, OpenAI Responses, Anthropic Messages and Gemini, translating between them with streaming and tool calls included. Point an agent at it, and the gateway forwards to whichever provider you picked for it.

Picking a model on the Agents page rewrites the one key that matters in that agent's own config file, in its own format, keeping the agent's comments and formatting intact. That is how Claude Code ends up on Kimi, Codex on DeepSeek, or Gemini CLI on GLM, and how OpenCode can run on your ChatGPT plan instead.

Routing groups add fail-over: when one account or model runs out of quota, magpie quietly moves to the next one in the group and the agent never sees the error.

Requires: macOS, Windows or Linux (menu bar app, window, TUI, web UI or CLI), a Docker host, or Termux on Android. You still need your own API key (DeepSeek, Kimi, GLM, OpenRouter...) or a subscription (Claude, ChatGPT, Copilot, Gemini, Grok) an agent is already signed in to; magpie finds those sign-ins by reading local files and the Keychain, and you can switch that off in settings.

The README names no benchmark numbers: this is a routing and config tool, not a speed claim. Model lists and names come straight from each vendor, filled in with reasoning levels from models.dev.

Install with curl -fsSL https://usemagpie.ai/install.sh | sh, with go install github.com/yetone/magpie@latest, or via the Docker image; Mac builds are signed and notarised and self-update. It is free and MIT licensed. Use it instead of hand-editing every agent's config when you run more than one agent, provider or account. Skip it if you run a single agent on its own vendor plan, you don't need it.

https://github.com/yetone/magpie

AI Coding Agent Sandboxes Compared: Claude Code, Docker, E2B, and the 2026 Escapes
What Docker, Claude Code, E2B and cloud sessions contain when a coding agent's prompts fail.

Watch the video on YouTube
Read the article

⚡ AI News
Anthropic sets Oct 14 investor day ahead of IPO: Anthropic will host institutional investors Oct 14 in SF, facing an Oct 23 S-1 deadline and trading possibly before Nov 26.

Meta launches Muse Wonder frontier AI models: Muse Wonder and Wonder Full add Active Memory and Active Focus, priced from $20 to $200 a month.

Microsoft unveils Maia 300 AI inference chip: Maia 300 packs 288GB of HBM4 and 18TB/s bandwidth, aiming to cut Nvidia reliance from 2027.

A folder of markdown files just hit #1 on GitHub trending

You prompt an agent with no plan, and it ships untested code: that's vibe coding.
A folder of markdown files pushes back against that, and it's now the fastest-climbing repo on GitHub.

▶️ Watch the video on YouTube

Vibe coding, named: no plan, no tests, no control over what an agent ships

Markdown, not more code: tells agents how to behave, not what to write

Grill you first: interviews you before building, then runs tests and reviews its own diffs

One command, three agents: installs into Claude Code, GitHub Copilot, or Codex

Trending at #1: top of GitHub's chart within a day, past 288k stars and climbing

Skills For Real Engineers

It's a public, MIT-licensed repo with zero application code: just markdown "skill" files that tell Claude Code, GitHub Copilot, or Codex how to run a test-driven loop, interview you before building (the "grill-me" skill), triage issues, plan large changes, and review their own diffs.

As fetched from GitHub, it stands at 288.2k stars, 24.2k forks, 1.5k watchers. Independent tracker mer.vin reports it hit #1 on GitHub's global trending chart within 24 hours and crossed 22,000 stars on day one.

Creator Matt Pocock, a TypeScript educator (Total TypeScript, ex-Vercel, ex-Stately/XState), describes it in his own words as "my agent skills that I use every day to do real engineering, not vibe coding," built to fix misaligned requirements, verbose output, untested code and architectural decay.

The catch: these are instructions, not intelligence. Your agent still does the actual work and still needs a model behind it, so the result depends on whatever you've connected, a local model or a paid API key. The repo itself names no hardware requirements of its own.

Requires: Claude Code, GitHub Copilot, or Codex already set up, plus Node for the installer.

Try it: npx skills@latest add mattpocock/skills, or for Claude Code: claude plugin install mattpocock-skills@claude-plugins-official, then run /setup-matt-pocock-skills once.

Use it instead of freeform vibe coding when you want an agent that plans and tests before shipping. Don't expect it to replace actually reading the diffs yourself.

github.com/mattpocock/skills

⚡ AI News
Anthropic warns AI could self-replicate on IoT devices: Anthropic's red team got a model to spread across 15 cheap Android TV boxes, warning full autonomy could arrive within months.

Anthropic slashes Claude distillation theft up to 99%: New safeguards block extraction attempts and ban offending accounts within two hours, Anthropic says.

Meta's Muse sexts on command despite its own ban: Wired got Muse to have explicit chats inside private VMs, contradicting Meta's public no-adult-content policy.

Character.AI files Chapter 11, Disney bids $20M: Disney is the stalking-horse bidder for IP licensing as the chatbot startup's bankruptcy sale heads to court.

A PDF editor that doesn't want your files in the cloud

PdfCraft is a PDF reader and editor written in Rust. No account, no telemetry: nothing leaves your machine while you work.

▶️ Watch the video on YouTube

No account, no telemetry: nothing leaves your machine

Open source top to bottom: engine, CLI and app

983 files, 0 crashes: their own pdf.js test corpus run

Drag pages like cards: reorganize a contract in Organize pages

Undo runs deep: saves never touch the original bytes

What's inside

PdfCraft is a clean-room PDF reader and editor written in Rust, with the engine, CLI and app all open source. No account, no telemetry: everything stays local, even the web build.

The app itself is free. The only thing that can cost you money is OCR: it needs a model you download separately, and whether that model is free or paid is on you.

The project's own run across the 983-file pdf.js test corpus reports 0 crashes. That's their own test, not an independent audit. Pages render in isolation, so damaged files get repaired instead of refused.

Saves append your changes and leave the original bytes untouched, so undo stays reliable even after you close and reopen.

Requires: macOS, Windows, Linux, FreeBSD, or a browser for the web build. The README doesn't list CPU, GPU or memory requirements beyond that. A separate model if you want OCR.

The catch: OCR isn't built in by default, you fetch a model for it first.

To try it: clone the source with git clone https://github.com/storytold/pdfcraft, then check the README's Downloads section for prebuilt apps per platform.

Use it instead of Adobe Acrobat when you want your PDFs to never touch somebody's server. Don't use it yet if you need OCR working out of the box.

github.com/storytold/pdfcraft

PdfCraft vs Adobe Acrobat

Both are full desktop PDF editors for reading, organizing, combining, splitting and securing files, which is why anyone shopping for a PDF editor ends up weighing the two. The real choice is between Adobe's mature, cloud-connected toolset and a free, offline, open-source alternative still in alpha.

Pick PdfCraft if you want a free, offline, open-source PDF tool and can live with alpha-stage text editing and a separate OCR download.

Pick Adobe Acrobat if you need mature OCR, redaction or e-signatures now and don't mind an Adobe account and a monthly subscription.

Price and license
PdfCraft: free, open-source (MIT/Apache)
Adobe Acrobat: $14.99-$24.99/mo, closed

Privacy and files
PdfCraft: no account, nothing leaves device
Adobe Acrobat: account required, 100GB cloud

Platform reach
PdfCraft: macOS, Windows, Linux, FreeBSD, web
Adobe Acrobat: macOS, Windows, web, mobile

Editing and OCR maturity
PdfCraft: alpha; OCR needs separate model
Adobe Acrobat: mature OCR and text editing

Side by side, in full

Price and license
PdfCraft is free and dual-licensed under MIT or Apache-2.0, with the engine, CLI and app all open source.
Adobe Acrobat individual plans run $14.99/month for Standard up to $24.99/month for the new Studio tier, and the app itself is closed-source.

Privacy and files
PdfCraft needs no account and sends no telemetry, so files never leave your machine unless you choose to share them.
Adobe Acrobat: Acrobat requires an Adobe account, bundles 100GB of cloud storage by default, and Adobe's privacy policy allows analytics on cloud content and usage data.

Platform reach
PdfCraft runs natively on macOS, Windows, Linux and FreeBSD, plus a web build, covering desktops Acrobat skips.
Adobe Acrobat: Acrobat ships for Windows and macOS desktops, the web, and iOS/Android apps, but has no native Linux or FreeBSD build.

Editing and OCR maturity
PdfCraft is alpha-stage, by its own count around 57% of Acrobat Pro's offline features and 94% of the must-haves, OCR needs a separately downloaded model, and editing existing text (especially CJK or right-to-left) is still thin.
Adobe Acrobat: Acrobat Pro's Recognize Text OCR and direct text editing are built in and mature, with no extra downloads needed, though some of this (like redaction) is Pro or Studio only, not in Standard.

PdfCraft: https://github.com/storytold/pdfcraft
Adobe Acrobat: https://www.adobe.com/acrobat/pricing/

A clean-room Microsoft Word, rebuilt in pure Rust

No Word source, no Word binaries: this is Word's workflow, reimplemented from scratch by watching how it behaves.

▶️ Watch the video on YouTube

Clean-room build: just public specs and observation, nothing copied

Familiar workflow: ribbon, styles, track changes, references, mail merge

Real files: reads and writes .docx, plus ODT, RTF, even LaTeX

Runs everywhere: one Rust codebase for macOS, Windows, Linux, BSD and the browser

Built for speed and agents: 1.4ms relayout on 188 pages, 422 commands, CLI and MCP

What's inside

WordCraft is a clean-room reimplementation: no Microsoft source and no Word binaries were used, only public file-format specs and observation, and every asset is original or openly licensed. It's MIT OR Apache-2.0, written entirely in Rust, and renders natively with egui instead of Electron or Tauri, with a build that also runs in the browser via WebAssembly.

The 1.4 millisecond figure is the WordCraft team's own measurement: typing in a 188-page document re-lays it out in about 1.4 ms because paragraph layout is cached. It's the project's own claim, not an independent benchmark.

It reads and writes real .docx (OOXML), plus .odt, .rtf, .html, .md and LaTeX, opens old Word 97-2003 .doc files, and exports PDF with selectable text and links. All 422 actions behind the ribbon, from shortcuts to spell check, also drive a CLI, a JSON control channel and an MCP server, so coding agents can write documents too.

Requires: a Rust toolchain to build from source. The README names macOS, Windows, Linux, BSD and the web as targets but doesn't list specific chips, RAM or VRAM.

The catch: this build is six days old, so treat it as an early release with missing features and rough edges, not a finished Word replacement.

It's free and runs entirely on your machine: there's no model in the loop, so there's nothing to pay a provider for.

Use it instead of Word when you want something open, offline and scriptable. Don't reach for it yet if you need full fidelity on complex existing .docx files.

To try it:

git clone https://github.com/storytold/wordcraft
cd wordcraft
cargo run -p wordcraft-ui-egui

github.com/storytold/wordcraft