Writ Cloud

Turn any website into an API your AI assistant can call. Describe what you need from a site; Writ builds the API, tests it and saves it as a tool, signed in as you when the site needs a login. Also reads pages to Markdown, crawls sites into datasets and watches pages for changes.

Hosted MCP Server

npx add-mcp 'https://api.usewrit.app/mcp'

Installs into Claude Code, Codex, Cursor and more

Documentation

Connect Writ.Your agent gets hands on the web.

Add one MCP server and your AI agent can drive the real web: open a live browser, sign in, turn any site into a callable API, and crawl whole sites. Ask once and Writ builds and tests the API; every ask after calls it, at zero token cost.

Works with Claude, Cursor, ChatGPT and any MCP client · runs on your own accounts, with your own credentials.

writ-cloud · mcp serverconnected

https://api.usewrit.app/mcp

toolwrit_website_to_apiturn a site into a REST + MCP endpoint

toolwrit_record_websiterecord the difficult part of a task

toolwrit_crawl_siteHarvest: crawl a whole site to markdown

toolwrit_run_workflowcall a saved API, no re-reasoning

toolwrit_search_apifind a ready-made API in the marketplacecloud

kit ▸ agent toolbelt 3

01

A toolbelt for the whole web, not just the APIs.

Most agents can only call the handful of services that ship an API. Connected to Writ, yours can reach anything a person can open in a browser, on your own authorized accounts. Six capabilities, one server:

◐Browse the real webOpens your own real browser (your profile, saved Personas, sealed credential vault, a real browser engine) and acts like a person: click, type, sign in, pass 2FA. Reaches gated, JavaScript-heavy pages a read-only tool cannot.writ_browser_use

⤳Turn a site into an APIDescribe the goal; Writ works through it on the live site and tests the result, then publishes a secured REST endpoint plus an MCP tool your agent can call by name from then on.writ_website_to_api

●Record the difficult partWhen a login or a picker needs your hands, record just that step; Writ builds the rest around it and saves the whole thing as a named, reusable tool.writ_record_website

≣Crawl a whole siteHarvest maps the site, then a bounded worker pool on the Writ fleet fetches every in-scope page (HTTP-first, browser fallback, robots and politeness honored) into one deduped markdown or schema dataset. A free account is enough to start.writ_crawl_site

▸Call the saved APICall any saved API to get a fresh answer: deterministic, and at zero AI-token cost, because the saved API is called, not re-reasoned.writ_run_workflow

⌕Search your dataAnswer questions straight from everything past runs already collected (instant, zero cost) before spending a single call re-running anything live.writ_search_data

The same tools run in two places, always on your own accounts: Writ Desktop’s local writ server on your machine (32 tools, no account needed), and the writ-cloud server at api.usewrit.app/mcp for callers that are not on this machine. writ_search_api finds a ready-made marketplace API: called as is in the cloud, rebuilt locally on Desktop.

The keyless tier: no account, no key. Public pages only — which is exactly the limit worth knowing.

curl -X POST https://api.usewrit.app/v1/keyless/scrape \
  -H "Content-Type: application/json" \
  -H "X-Writ-Client-Id: my-device-1" \
  -d '{"url":"https://example.com"}'
{ "url": "https://example.com",
  "markdown": "# Example Domain\n…" }

This page was public. For pages behind a login, run it with a persona. Keyless calls carry fair-use rate limits.

loop ▸ build once · call 6

02

Ask once, it builds it. Ask again, it calls it.

The first time you ask your agent for a reliable way to reach a site, Writ does the hard work once: opens a browser, works through the steps, tests the result and saves it as an API. Every ask after that calls the saved API: deterministic, never re-reasoned, free of token cost. Watch both halves:

▸Build a reliable way to pull open orders from supplier-portal.com

One-time build · runs on your machine · you supply the credentials

The saved API is also a REST endpoint and MCP tool, so the same API your agent built is callable from your code, other agents, or a schedule. See Website → API and Automations.

diff ▸ read vs act 9

03

Read-only crawlers stop at the page. Writ tools act.

An MCP server that just dumps page text cannot submit forms, update dashboards, trigger approved workflows, or pull data from behind a login. Writ packages the whole task, including authenticated, multi-step ones, as an MCP tool. Your agent calls a tool; Writ runs the API and returns structured output.

Writ MCP tools

  • Take real actions on the page
  • Log in, pass TOTP and mailbox-OTP
  • Fill forms, click, navigate
  • Complete multi-step tasks
  • Return typed, structured output

pin ▸ api → tool 12

04

How a saved API becomes a tool

Every saved API is callable through writ_run_workflow. Pin one and the server lists it as its own run_<name> tool, with the API’s inputs as its typed schema and the description your agent reads to decide when to use it.

  1. 01 A saved API Start from any API Writ built, rebuilt or recorded.
  2. 02 Pin it One call to writ_pin_workflow_tool; the input schema comes from the API’s own inputs.
  3. 03 Listed on MCP The cloud server at api.usewrit.app/mcp lists it as run_<name>; Writ Desktop’s writ server does the same on your machine.
  4. 04 Agent discovers & calls Your agent picks the tool by its description and calls it.
writ_pin_workflow_tool({
  "workflow": "price-check",
  "pinned": true
})
→ run_price_check
  listed on https://api.usewrit.app/mcp

05

Connect it to your agent

Point any MCP-compatible client at the Writ server URL and authenticate with a wt_ key. Tools appear automatically; your agent picks them by description.

{
  "mcpServers": {
    "writ-cloud": {
      "url": "https://api.usewrit.app/mcp",
      "headers": { "Authorization": "Bearer wt_8f2c...a91" }
    }
  }
}

Works with Claude, Cursor, ChatGPT and any MCP client. Authorize with a scoped wt_ key (or OAuth, where the client signs in); revoke or rotate it without touching the API.

set ▸ your client 18

06

Set it up where your agent lives

One page per client, with the exact config block and the steps that work today.

Claude Desktop Give Claude Desktop real web actions on any site (login, forms, multi-step tasks) over MCP. Set up Claude Code Give Claude Code callable web actions on any site: one command to add the Writ MCP server. Set up Cursor Give the Cursor agent real web actions on any site via MCP, without leaving the editor. Set up ChatGPT Give ChatGPT real web actions on any site (login, forms, multi-step tasks) over MCP. Set up VS Code (Copilot) Give Copilot agent mode real web actions on any site (login, forms, multi-step tasks) over MCP. Set up Lovable Give a Lovable build real data from any site (login, forms, multi-step tasks) over MCP. Set up Windsurf Give Cascade real web actions on any site via MCP, authenticated and on your own accounts. Set up

07

Tools that work on the hard web

Because each tool was built and tested on the live site, it handles login, TOTP and mailbox-OTP two-factor on your own authenticated accounts, dynamic pages, and 30+ step types. When a site moves, auto-repair rebuilds the broken step, and repairs are free.

Authenticated actionsLog in, pass TOTP and mailbox-OTP two-factor, and act on your own authenticated accounts.

Resilient to UI changesAuto-repair rebuilds a broken step when a site moves, so tools keep working through redesigns. Repairs are free.

Structured outputEach tool returns typed JSON your agent can read and act on, not raw page text.

Two-factor and login run on your own authenticated accounts; see Personas.

sec ▸ authorized 24

08

Authorized by design

Writ runs on your own accounts, with your own credentials and data, on sites you are authorized to use. It automates work a person could do by hand in a browser. Secrets live in an encrypted vault, tools are scoped by wt_ key, and every tool call is logged and attributable to a run.

Your accountsTools act on accounts you own and are authorized to use.

Encrypted secretsCredentials live in an encrypted vault, never in the tool definition.

Scoped keysEach MCP connection is authorized by a scoped wt_ key or an OAuth grant you can revoke.

Logged runsEvery tool call is logged and attributable to a run.

See trust, security, and the acceptable use policy.

faq ▸ answered 27

09

MCP FAQ

What is a Writ MCP tool?

A Writ MCP tool is an API Writ built from one sentence (a real browser only when the site needs one), exposed over the Model Context Protocol. It has a name, a description the agent reads to decide when to use it, a typed input schema, and a structured output. Unlike a read-only crawler, it takes real actions: login, forms, clicks, multi-step tasks.

Which MCP clients work with Writ?

Any MCP-compatible client. Point Claude, Cursor, ChatGPT or your own MCP client at the Writ server URL and authenticate with a scoped wt_ key (or sign in over OAuth where the client supports it). Tools appear automatically and the agent selects them by description.

Does it handle login and 2FA?

Yes. Because each tool was built on the live site, it handles login, TOTP and mailbox-OTP two-factor, dynamic pages, and 30+ step types, on your own authenticated accounts. Writ runs on your own accounts, with your own credentials and data, on sites you are authorized to use.

How is a Writ tool different from a read-only MCP server?

Most website-to-MCP tools only read page text. Writ tools act: they log in, fill forms, click, extract, and complete multi-step tasks, then return structured output. Your agent calls a tool; Writ runs the saved API.

Connect your MCP client, ask for the site you need, and let your agent act on it from then on.