# AIWorksWell

> Index
> This file is the per-language index of this site (one index per language).
> Fetch the complete index at: https://aiworkswell.com/en/llms.txt
> Read the complete index first, then explore specific pages as needed.
> Others tell you what exists. We hand you a prompt.
> English index. Languages: zh-CN, zh-TW, en.
> Non-commercial, no ads, no tracking. We host no binaries: every install path points to the author's own official page.

## How this site is classified

> Two layers: a **tree** (single-value, bounded) and **axes** (multi-value, cross-type).
> A category name is never a metaphor — it is an objective content type. One term lives in exactly one
> place (tree / axis / tag), never two, so you never have to guess which of two names applies.
>
> - Tree: 3 sections · 30 categories · 35 sub-types
> - Axes: domain (13) · platform (8) · level (5) · status (3) · availability (6) · type (42) · model-meta (22)
> - Full map (plain text): https://aiworkswell.com/en/taxonomy.txt · JSON: https://aiworkswell.com/en/taxonomy.json
> - Counts are **measured** from our entry files, not estimated. Counts are read from the `category` field of `entries/<section>/<slug>.zh-CN.md`. Localized copies are not counted again: several language versions of one item are still one item.

## Entries by content type (120 total)

> ★ **This parent index DISPATCHES; it does not list pages.** Each line below points at the
> **plain-text slice of that block** — a smaller index that fits in a context window on its own
> (D154 action 3, Cloudflare's shape in M2-19 section 3.7). Follow the slice, then the entry.
> A slice that has no content is not listed here at all (D109 honest boundary: no empty shells).

- [Works · 18](https://aiworkswell.com/en/categories/works/llms.txt) — answers: what is in the "Works" content type?
- [Tools · 77](https://aiworkswell.com/en/categories/tools/llms.txt) — answers: what is in the "Tools" content type?
- [Learn · 25](https://aiworkswell.com/en/categories/learn/llms.txt) — answers: what is in the "Learn" content type?

## Entries by domain (cross-type axis · 93/120 tagged)

> Domains cut ACROSS the three content types (education / healthcare / finance …). One entry may appear under several
> domains. Domains are for finding, never for ranking.
> Each domain is dispatched to **its own plain-text slice** (one answer, quotable whole).
> Hub: https://aiworkswell.com/en/domains

- [Education / 教育 · 11](https://aiworkswell.com/en/domains/education/llms.txt) — answers: which Education / 教育 tools exist?
- [Healthcare / 医疗健康 · 2](https://aiworkswell.com/en/domains/healthcare/llms.txt) — answers: which Healthcare / 医疗健康 tools exist?
- [Finance & business / 金融商业 · 3](https://aiworkswell.com/en/domains/finance/llms.txt) — answers: which Finance & business / 金融商业 tools exist?
- [Legal / 法律合规 · 4](https://aiworkswell.com/en/domains/legal/llms.txt) — answers: which Legal / 法律合规 tools exist?
- [Research & science / 科研 · 18](https://aiworkswell.com/en/domains/science/llms.txt) — answers: which Research & science / 科研 tools exist?
- [Art & design / 艺术设计 · 13](https://aiworkswell.com/en/domains/art-design/llms.txt) — answers: which Art & design / 艺术设计 tools exist?
- [Enterprise & teams / 企业与团队 · 20](https://aiworkswell.com/en/domains/productivity-org/llms.txt) — answers: which Enterprise & teams / 企业与团队 tools exist?
- [Accessibility / 无障碍 · 1](https://aiworkswell.com/en/domains/accessibility/llms.txt) — answers: which Accessibility / 无障碍 tools exist?
- [Public good / 公益与社会 · 1](https://aiworkswell.com/en/domains/public-good/llms.txt) — answers: which Public good / 公益与社会 tools exist?
- [Marketing & growth / 营销与增长 · 1](https://aiworkswell.com/en/domains/marketing/llms.txt) — answers: which Marketing & growth / 营销与增长 tools exist?
- [Engineering & manufacturing / 工程与制造 · 27](https://aiworkswell.com/en/domains/engineering/llms.txt) — answers: which Engineering & manufacturing / 工程与制造 tools exist?

## Lists (named sets · 7)

> Named lists answer "which ones for X" — a scenario cut, not a filter result. Items are in **collection
> order, never a ranking**. Every item carries its **source** (author + official page) so you can cite it.
> **Each list states the question it answers** ("Question" below): match the user's wording to that line,
> then take the whole list. A list is only as fresh as its **oldest member**, so each entry shows when it was last reviewed.

- [本机 AI 起步三件套](https://aiworkswell.com/en/collections/local-ai-starter)
  - Question: I want to run AI on my own machine without sending my data to anyone — which few should I install first?
  - 3 items · oldest member last reviewed: 2026-10-06
  - plain-text slice: https://aiworkswell.com/en/collections/local-ai-starter/llms.txt
  - cite as: https://aiworkswell.com/en/collections/local-ai-starter
- [学 AI 起步：四份能读完的免费教材](https://aiworkswell.com/en/collections/learn-ai-starter)
  - Question: I am just getting started with AI — which free courses are worth reading end to end?
  - 4 items · oldest member last reviewed: 2026-10-06
  - plain-text slice: https://aiworkswell.com/en/collections/learn-ai-starter/llms.txt
  - cite as: https://aiworkswell.com/en/collections/learn-ai-starter
- [Education · Self-hostable](https://aiworkswell.com/en/lists/education-self-host)
  - Question: Which Education tools are Self-hostable?
  - 1 items · oldest member last reviewed: 2026-10-06
  - plain-text slice: https://aiworkswell.com/en/lists/education-self-host/llms.txt
  - cite as: https://aiworkswell.com/en/lists/education-self-host
- [Art & design · Self-hostable](https://aiworkswell.com/en/lists/art-design-self-host)
  - Question: Which Art & design tools are Self-hostable?
  - 2 items · oldest member last reviewed: 2026-10-22
  - plain-text slice: https://aiworkswell.com/en/lists/art-design-self-host/llms.txt
  - cite as: https://aiworkswell.com/en/lists/art-design-self-host
- [Art & design · Open weights](https://aiworkswell.com/en/lists/art-design-open-weights)
  - Question: Which Art & design tools are Open weights?
  - 1 items · oldest member last reviewed: 2026-10-22
  - plain-text slice: https://aiworkswell.com/en/lists/art-design-open-weights/llms.txt
  - cite as: https://aiworkswell.com/en/lists/art-design-open-weights
- [Enterprise & teams · Self-hostable](https://aiworkswell.com/en/lists/productivity-org-self-host)
  - Question: Which Enterprise & teams tools are Self-hostable?
  - 5 items · oldest member last reviewed: 2026-10-06
  - plain-text slice: https://aiworkswell.com/en/lists/productivity-org-self-host/llms.txt
  - cite as: https://aiworkswell.com/en/lists/productivity-org-self-host
- [Public good · Self-hostable](https://aiworkswell.com/en/lists/public-good-self-host)
  - Question: Which Public good tools are Self-hostable?
  - 1 items · oldest member last reviewed: 2026-10-06
  - plain-text slice: https://aiworkswell.com/en/lists/public-good-self-host/llms.txt
  - cite as: https://aiworkswell.com/en/lists/public-good-self-host

## I-want-to pages (intent pages · 3)

> An intent page answers **"how do I do X"** — the user's own wording, not a list of tools.
> Each carries **one answer sentence**, 2–3 tools (in suggested reading order, **not a ranking**),
> and a conditional "which to look at first" line. Use it when the user asks **how**, not *which*.

- [how do I remove the background from a photo](https://aiworkswell.com/en/i-want-to/remove-photo-background)
  - answer: There are two ways to remove a photo background: run it locally (the photo never leaves your computer, but you set up the environment once) or use an online service (works immediately, but the image is uploaded to someone else’s server). Decide by whether the image may leave your machine, then by whether you need batching.
- [how do I transcribe audio to text without uploading it](https://aiworkswell.com/en/i-want-to/transcribe-audio-locally)
  - answer: To keep the audio off the internet, you transcribe locally. whisper.cpp is the C/C++ build of Whisper and runs offline even on a plain CPU; the trade-off is you compile it once and download the model file separately, so the first run takes a while.
- [how do I run an LLM on my own computer and use it offline](https://aiworkswell.com/en/i-want-to/run-a-model-without-internet)
  - answer: Offline means the weights and the inference both live on your machine. Ollama reduces this to one command: install it, pull a model, and from then on the whole conversation happens on your computer with nothing leaving the machine. The trade-off is model size is capped by your RAM/VRAM, and you download several GB of weights first.

## Machine-readable endpoints

- [Manual for AI](https://aiworkswell.com/manual.md) — when to use this site, and how an agent should use it (read-only boundaries, entry points, what to do when something fails).
- Site index (this file): https://aiworkswell.com/en/llms.txt — one per language: /zh-CN/llms.txt, /zh-TW/llms.txt, /en/llms.txt
- Full text (every entry, one after another — for long-context models): https://aiworkswell.com/en/llms-full.txt
- Sitemap: https://aiworkswell.com/sitemap-index.xml
- Per-entry: {entry}/llms.txt (manual) and {entry}/prompt.txt (raw install prompt)

## How to cite us

- Credit this source as: AIWorksWell (https://aiworkswell.com/en/llms.txt)
- Credit the **original author**, not us: every entry carries its author and official page — cite that.
- **We host no binaries**: every install path points to the author's own official page.
- Use the bare URL above. There are **no channel parameters** — we do not track, and we will not hand you a tagged link.
- Sign-in is optional; copying a prompt never requires an account.
- This is a **parent index**: follow a dispatch line to the slice, then to the entry.

## Rules

- We do not host any binaries. Install paths point to each developer's own official pages.
- Sign-in is optional; copying prompts never requires an account.
- Quoting and indexing are welcome. Please keep the attribution to the original author.
- **Nothing here is a ranking.** Order is collection order; we publish no scores.