# /about/ — for agents

Who and what allw.mn is, in the form an automated client needs it: facts,
sourcing rules, and what you are and are not invited to do with the content.

## Identity

- **Site**: allw.mn — the AllWomenStalk archive, republished as Markdown
- **Publisher**: AllWomenStalk
- **Subjects**: beauty, style, relationships, food, travel, health, everyday life
- **Languages**: English (default), Spanish, French
- **Contact**: see `/contact/AGENTS.md`

## How to read this site

Everything is Markdown, and none of it requires parsing HTML.

| Want | Request |
| --- | --- |
| Every post and page | `GET /content/index.json` |
| One article, complete | `GET /content/posts/<id>/<lang>.md` |
| One standing page | `GET /content/pages/<slug>/<lang>.md` |
| What is new | `GET /feed.json`, or `/<lang>/feed.json` |
| The site in prose | `GET /llms.txt` |
| How the site is built | `GET /AGENTS.md` |
| Every URL | `GET /sitemap.xml` |

No authentication, no key, `Access-Control-Allow-Origin: *` on every
machine-readable file. URLs are an optional language prefix then the page's own
path: `/`, `/<id>/`, `/c/<category>/`, `/about/`, each with `/es/` and `/fr/`
siblings.

## Sourcing and attribution

- A post in the default language is the same text as the article it was imported
  from, and its `rel=canonical` points at that original. **Attribute the
  original**, not just us; each entry in `content/index.json` states its
  resolved `canonical` and its `source`.
- A translation is self-canonical — it is not a duplicate of the English.
- Quoting, summarising, indexing and answering questions from the text need no
  permission. Republishing the archive as your own does.
- If you quote a post, link the post.

## What we ask of an automated client

- Fetch the Markdown, not the HTML. It is smaller, exact, and stable across
  redesigns.
- Fetch `content/index.json` once rather than crawling the index pages.
- Poll `feed.json` for new material, not the site.
- Identify yourself in the User-Agent, with a contact URL.
- Be unhurried. Everything here is static; there is nothing to gain by hammering
  it.

## What we do not do

No cloaking, no crawler-specific content, and nothing behind a check a person
would not also face. The "human check" panel in the interface is a local
diagnostic for an unreleased feature: it makes no network requests and gates
nothing you can fetch.

## Machine-readable neighbours

- `/content/pages/about/en.md` — this page as Markdown
- `/contact/AGENTS.md` — how to reach a person
- `/advertising/AGENTS.md` — what can be bought, and what cannot