# Writ Cloud (remote · api.usewrit.app)

Read, crawl and act on websites, signed in as the user, and turn any site into an API.

- Trust score: 76/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-17

## Components

- remote · `api.usewrit.app`: 76/100 (this document), [markdown](https://verifymcp.io/servers/app-usewrit-writ-cloud/api.md), [page](https://verifymcp.io/servers/app-usewrit-writ-cloud/api)
- npm · `writ-mcp`: 50/100, [markdown](https://verifymcp.io/servers/app-usewrit-writ-cloud/writ-mcp.md), [page](https://verifymcp.io/servers/app-usewrit-writ-cloud/writ-mcp)

## Channel facts

- Endpoint: `https://api.usewrit.app/mcp`
- Transports: `streamable-http`
- Auth: `required`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-10-17.

- **Endpoint Security**: 94/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
  - The authorisation server offers only Dynamic Client Registration (RFC 7591), which MCP 2026-07-28 deprecated in favour of Client ID Metadata Documents.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 56/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 6062 tokens (~147/item across 41 items; 41 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 20/100
  - Stability observed for 6 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 86/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 58% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 41 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 42 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 73/100
  - Spec-recency check failed: implements MCP spec 2025-06-18; the latest is 2026-07-28.
  - Supports UI / widget rendering.

## Install

### How do I install the Writ Cloud MCP server?

Writ Cloud is a hosted endpoint at https://api.usewrit.app/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http app-usewrit-writ-cloud 'https://api.usewrit.app/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "app-usewrit-writ-cloud": {
      "url": "https://api.usewrit.app/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "app-usewrit-writ-cloud": {
      "type": "http",
      "url": "https://api.usewrit.app/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.app-usewrit-writ-cloud]
url = "https://api.usewrit.app/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "app-usewrit-writ-cloud": {
      "type": "remote",
      "url": "https://api.usewrit.app/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add app-usewrit-writ-cloud --url 'https://api.usewrit.app/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  app-usewrit-writ-cloud:
    url: "https://api.usewrit.app/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "app-usewrit-writ-cloud": {
      "Transport": "http",
      "Url": "https://api.usewrit.app/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add app-usewrit-writ-cloud -t streamable-http -u 'https://api.usewrit.app/mcp'
```

### Other

```json
{
  "mcpServers": {
    "app-usewrit-writ-cloud": {
      "type": "http",
      "url": "https://api.usewrit.app/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-17 (score 76, 0)

- [functional] First check of Capabilities: pass
- [functional] This server's schema is too large to store in full, so we cannot compare its tools day to day

### 2026-10-16 (score 76, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-14 (score 75, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-13 (score 74, +1)

We did not load change detail this far back for this component, so this day may have recorded more than is shown.

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “writ_payment” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_sessions” rewrote its description, which is the text the model reads
- [security] Tool “writ_update_saved_crawl” rewrote its description, which is the text the model reads
- [security] Tool “writ_run_saved_crawl” rewrote its description, which is the text the model reads
- [security] Tool “writ_discovery_status” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_compose” rewrote its description, which is the text the model reads
- [security] Tool “writ_workflow_runs” rewrote its description, which is the text the model reads
- [security] Tool “writ_update_workflow” rewrote its description, which is the text the model reads
- [security] Tool “writ_pin_workflow_tool” rewrote its description, which is the text the model reads
- [security] Tool “writ_scrape” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_act” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_use” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_context” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_ask_user” rewrote its description, which is the text the model reads
- [security] Tool “writ_record_website” rewrote its description, which is the text the model reads
- [security] Tool “writ_diagnose_http_workflow” rewrote its description, which is the text the model reads
- [security] Tool “writ_saved_crawl_data” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_save” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_network” rewrote its description, which is the text the model reads
- [security] Tool “writ_create_monitor” rewrote its description, which is the text the model reads
- [security] Tool “writ_wire_monitor” rewrote its description, which is the text the model reads
- [security] Tool “writ_create_http_extraction” rewrote its description, which is the text the model reads
- [security] Tool “writ_browser_cancel” rewrote its description, which is the text the model reads
- [security] Tool “writ_search_data” rewrote its description, which is the text the model reads
- [security] Tool “writ_run_workflow” rewrote its description, which is the text the model reads
- [security] Tool “writ_create_automation” rewrote its description, which is the text the model reads
- [security] Tool “writ_list_webhooks” rewrote its description, which is the text the model reads
- [security] Tool “writ_workflow_data” rewrote its description, which is the text the model reads
- [security] Tool “writ_crawl_status” rewrote its description, which is the text the model reads
- [security] Tool “writ_crawl_site” rewrote its description, which is the text the model reads
- [security] Tool “writ_expose_workflow_api” rewrote its description, which is the text the model reads
- [security] Tool “writ_website_to_api” rewrote its description, which is the text the model reads
- [security] Tool “writ_personas” rewrote its description, which is the text the model reads
- [security] Tool “writ_export_data” rewrote its description, which is the text the model reads
- [security] Tool “writ_saved_crawls” rewrote its description, which is the text the model reads
- [security] Tool “writ_devices” rewrote its description, which is the text the model reads
- [security] Tool “writ_crawl_files” rewrote its description, which is the text the model reads
- [security] Tool “writ_set_schedule” rewrote its description, which is the text the model reads
- [functional regression] Tool coverage: 83% → 59%
- [functional improvement] Schema quality: 701 → 141
- [functional] New tool “writ_help”
- [cosmetic] “writ_website_to_api” reworded the description of “skip_existing”
- [cosmetic] “writ_record_website” reworded the description of “persona_id”
- [cosmetic] “writ_payment” reworded the description of “grant_id”
- [cosmetic] “writ_crawl_site” reworded the description of “intent”
- [cosmetic] “writ_wire_monitor” reworded the description of “threshold”
- [cosmetic] “writ_crawl_site” reworded the description of “extract_prompt”
- [cosmetic] “writ_create_automation” reworded the description of “on_workflow”
- [cosmetic] “writ_create_monitor” reworded the description of “interval”
- [cosmetic] “writ_browser_use” reworded the description of “url”
- [cosmetic] “writ_record_website” reworded the description of “use_residential”
- [cosmetic] “writ_browser_act” reworded the description of “inputs”
- [cosmetic] “writ_browser_use” reworded the description of “persona_id”
- [cosmetic] “writ_saved_crawls” reworded the description of “limit”
- [cosmetic] “writ_browser_save” reworded the description of “name”
- [cosmetic] “writ_export_data” reworded the description of “format”
- [cosmetic] “writ_payment” reworded the description of “commit_selector”
- [cosmetic] “writ_create_monitor” reworded the description of “selector”
- [cosmetic] “writ_create_monitor” reworded the description of “url”
- [cosmetic] “writ_workflow_data” reworded the description of “refs”
- [cosmetic] “writ_create_automation” reworded the description of “name”
- [cosmetic] “writ_search_data” reworded the description of “q”
- [cosmetic] “writ_browser_use” reworded the description of “use_residential”
- [cosmetic] “writ_discovery_status” reworded the description of “wait”
- [cosmetic] “writ_crawl_site” reworded the description of “page_budget”
- [cosmetic] “writ_devices” reworded the description of “device”
- [cosmetic] “writ_diagnose_http_workflow” reworded the description of “task_id”
- [cosmetic] “writ_browser_use” reworded the description of “execution_target”
- [cosmetic] “writ_scrape” reworded the description of “url”
- [cosmetic] “writ_payment” reworded the description of “max_amount”
- [cosmetic] “writ_browser_network” reworded the description of “query”
- [cosmetic] “writ_crawl_site” reworded the description of “wait”
- [cosmetic] “writ_create_automation” reworded the description of “notify”
- [cosmetic] “writ_website_to_api” reworded the description of “persona_id”
- [cosmetic] “writ_personas” reworded the description of “persona_id”
- [cosmetic] “writ_record_website” reworded the description of “url”
- [cosmetic] “writ_crawl_status” reworded the description of “wait”
- [cosmetic] “writ_workflow_data” reworded the description of “q”
- [cosmetic] “writ_browser_act” reworded the description of “session_id”
- [cosmetic] “writ_personas” reworded the description of “link_id”
- [cosmetic] “writ_browser_use” reworded the description of “goal”
- [cosmetic] “writ_browser_network” reworded the description of “index”
- [cosmetic] “writ_browser_ask_user” reworded the description of “kind”
- [cosmetic] “writ_set_schedule” reworded the description of “tz”
- [cosmetic] “writ_create_automation” reworded the description of “target_id”
- [cosmetic] “writ_crawl_files” reworded the description of “crawl”
- [cosmetic] “writ_crawl_site” reworded the description of “url”
- [cosmetic] “writ_create_monitor” reworded the description of “persona_id”
- [cosmetic] “writ_scrape” reworded the description of “top_n”
- [cosmetic] “writ_create_monitor” reworded the description of “watch”
- [cosmetic] “writ_create_monitor” reworded the description of “session_id”
- [cosmetic] “writ_create_http_extraction” reworded the description of “goal”
- [cosmetic] “writ_wire_monitor” reworded the description of “workflow”
- [cosmetic] “writ_create_monitor” reworded the description of “extract”
- [cosmetic] “writ_run_workflow” reworded the description of “mutation_mode”
- [cosmetic] “writ_scrape” reworded the description of “urls”
- [cosmetic] “writ_wire_monitor” reworded the description of “buy”
- [cosmetic] “writ_payment” reworded the description of “confirmation_id”
- [cosmetic] “writ_website_to_api” reworded the description of “mode”
- [cosmetic] “writ_run_workflow” reworded the description of “persona_id”
- [cosmetic] “writ_wire_monitor” reworded the description of “recipients”
- [cosmetic] “writ_create_http_extraction” reworded the description of “apply”
- [cosmetic] “writ_discovery_status” reworded the description of “timeout_seconds”
- [cosmetic] “writ_set_schedule” reworded the description of “time”
- [cosmetic] “writ_pin_workflow_tool” reworded the description of “pinned”
- [cosmetic] “writ_website_to_api” reworded the description of “save_as”
- [cosmetic] “writ_pin_workflow_tool” dropped the optional parameter “workflow”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “expected_updated_at”
- [cosmetic] “writ_wire_monitor” dropped the optional parameter “enabled”
- [cosmetic] “writ_create_monitor” dropped the optional parameter “device”
- [cosmetic] “writ_browser_sessions” dropped the optional parameter “include_closed”
- [cosmetic] “writ_scrape” dropped the optional parameter “respect_robots”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “name”
- [cosmetic] “writ_run_workflow” dropped the optional parameter “timeout_seconds”
- [cosmetic] “writ_record_website” dropped the optional parameter “skip_existing”
- [cosmetic] “writ_personas” dropped the optional parameter “device”
- [cosmetic] “writ_crawl_status” dropped the optional parameter “timeout_seconds”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “residential_country”
- [cosmetic] “writ_create_monitor” dropped the optional parameter “residential_country”
- [cosmetic] “writ_run_workflow” dropped the optional parameter “device”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “use_residential”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “replace_steps”
- [cosmetic] “writ_workflow_data” dropped the optional parameter “output”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “path”
- [cosmetic] “writ_workflow_data” dropped the optional parameter “workflow”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “relevance_threshold”
- [cosmetic] “writ_create_automation” dropped the optional parameter “title”
- [cosmetic] “writ_payment” dropped the optional parameter “fields”
- [cosmetic] “writ_create_http_extraction” dropped the optional parameter “response_sample”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “output”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “render_mode”
- [cosmetic] “writ_workflow_data” dropped the optional parameter “device”
- [cosmetic] “writ_expose_workflow_api” dropped the optional parameter “wait_timeout”
- [cosmetic] “writ_create_monitor” dropped the optional parameter “mode”
- [cosmetic] “writ_run_saved_crawl” dropped the optional parameter “limit”
- [cosmetic] “writ_export_data” dropped the optional parameter “workflow”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “extract_schema”
- [cosmetic] “writ_set_schedule” dropped the optional parameter “enabled”
- [cosmetic] “writ_create_automation” dropped the optional parameter “priority”
- [cosmetic] “writ_run_workflow” dropped the optional parameter “use_residential”
- [cosmetic] “writ_create_automation” dropped the optional parameter “enabled”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “step_id”
- [cosmetic] “writ_set_schedule” dropped the optional parameter “workflow”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “regenerate_skill”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “patch”
- [cosmetic] “writ_create_monitor” dropped the optional parameter “try_anyway”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “verbose”
- [cosmetic] “writ_wire_monitor” dropped the optional parameter “name”
- [cosmetic] “writ_personas” dropped the optional parameter “force”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “speed”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “section”
- [cosmetic] “writ_run_saved_crawl” dropped the optional parameter “preview_chars”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “same_domain”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “max_chars”
- [cosmetic] “writ_wire_monitor” dropped the optional parameter “workflow_id”
- [cosmetic] “writ_create_automation” dropped the optional parameter “run_workflow_id”
- [cosmetic] “writ_website_to_api” dropped the optional parameter “anonymous”
- [cosmetic] “writ_browser_use” dropped the optional parameter “fresh_exit”
- [cosmetic] “writ_record_website” dropped the optional parameter “human_layer”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “operator”
- [cosmetic] “writ_saved_crawl_data” dropped the optional parameter “output”
- [cosmetic] “writ_update_saved_crawl” dropped the optional parameter “name”
- [cosmetic] “writ_create_automation” dropped the optional parameter “on_workflow_id”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “max_depth”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “timeout_seconds”
- [cosmetic] “writ_create_automation” dropped the optional parameter “run_function”
- [cosmetic] “writ_run_workflow” dropped the optional parameter “residential_country”
- [cosmetic] “writ_browser_context” dropped the optional parameter “offset”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “limit”
- [cosmetic] “writ_browser_save” dropped the optional parameter “keep_open”
- [cosmetic] “writ_create_monitor” dropped the optional parameter “request_url”
- [cosmetic] “writ_wire_monitor” dropped the optional parameter “title”
- [cosmetic] “writ_create_http_extraction” dropped the optional parameter “desired_inputs”
- [cosmetic] “writ_personas” dropped the optional parameter “include_runs”
- [cosmetic] “writ_record_website” dropped the optional parameter “auth_mode”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “preview_chars”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “function_name”
- [cosmetic] “writ_update_workflow” dropped the optional parameter “step_updates”
- [cosmetic] “writ_run_saved_crawl” dropped the optional parameter “output”
- [cosmetic] “writ_record_website” dropped the optional parameter “residential_country”
- [cosmetic] “writ_create_automation” dropped the optional parameter “inputs”
- [cosmetic] “writ_set_schedule” dropped the optional parameter “functions”
- [cosmetic] “writ_workflow_data” dropped the optional parameter “view”
- [cosmetic] “writ_payment” dropped the optional parameter “automation_id”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “allow_subdomains”
- [cosmetic] “writ_create_monitor” dropped the optional parameter “enabled”
- [cosmetic] “writ_export_data” dropped the optional parameter “q”
- [cosmetic] “writ_expose_workflow_api” dropped the optional parameter “timeout_seconds”
- [cosmetic] “writ_scrape” dropped the optional parameter “device”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “max_concurrent_shards”
- [cosmetic] “writ_workflow_runs” dropped the optional parameter “device”
- [cosmetic] “writ_wire_monitor” dropped the optional parameter “entry_url”
- [cosmetic] “writ_crawl_site” dropped the optional parameter “exclude_paths”
- [cosmetic] “writ_run_workflow” dropped the optional parameter “output”

### 2026-10-11 (score 73)

First indexed and scored.

## MCP tools (41)

### `writ_browser_use` (~250 tokens)

Open a browser

A REAL CLOUD BROWSER FOR A TASK ON A SITE: the user's signed-in account (a persona_id from writ_personas: the password stays sealed in Writ, and sessions never ask for a password), a click, form, submit or search inside an app, or a page a plain fetch cannot open (login wall, 403, CAPTCHA). Opens the browser and returns the first page; writ_browser_act drives it (Writ runs no model here). Recording is always on: writ_browser_save turns the session into a workflow that replays without a browser, writ_browser_cancel closes it (an open browser bills). The start answer carries the action vocabulary, the recording rules and, with a persona, the 2FA rules. NOT FOR: reading pages (writ_scrape). An API: writ_website_to_api opens this same browser, bound to a build.

Input parameters:

- `execution_target` (string): cloud | auto | local (the user's own desktop app).
- `goal` (string)
- `persona_id` (integer|string)
- `url` (string, required)
- `use_residential` (boolean): Open on the residential network (bot walls, CAPTCHAs).

### `writ_record_website` (~116 tokens)

Record a website task

Records a repeatable task on a site as a workflow (goal + start url): the same session as writ_browser_use, labelled with its goal, driven with writ_browser_act and saved with writ_browser_save. A goal that asks for an API opens as a writ_website_to_api build instead: the same browser, bound to the build.

Input parameters:

- `goal` (string, required)
- `persona_id` (integer|string)
- `url` (string, required)
- `use_residential` (boolean): Residential network (bot walls).

### `writ_website_to_api` (~298 tokens)

Turn a website into an API

TURNS A WEBSITE INTO A CALLABLE API (named functions with inputs, returning JSON), built by the caller in a real browser. url = the page that already shows the rows (search results or a listing, not the home page); goal = the inputs and fields wanted. The call first proposes the user's matching saved workflows and marketplace APIs (skip_existing / skip_marketplace build fresh), then opens a cloud browser bound to the build: capture_network records the site's own calls, writ_browser_network reads them, api_call probes them; a failed probe is evidence, not the end. Functions: HTTP flows first (writ_browser_compose define_function `flow`; writ_help('http_flows')), single requests (from_index) mixed in, DOM extraction last. writ_browser_save, then run_example on two inputs. mode=intelligent: Writ builds it alone (+ writ_discovery_status wait=true), on request.

Input parameters:

- `build_id` (integer)
- `goal` (string)
- `mode` (string): guided (default: the caller drives) | intelligent (Writ builds it alone, on request) | auto | fast | crawl | browser.
- `persona_id` (integer)
- `save_as` (string)
- `skip_existing` (boolean): Skip the proposal of the user's own matching workflows.
- `skip_marketplace` (boolean): Skip the marketplace proposals.
- `url` (string)

### `writ_discovery_status` (~138 tokens)

Check an API build

Status of a writ_website_to_api build. wait=true is one held call (up to 75 s) that follows the build to its newest rung and returns when it is done or parked. Answers: succeeded (workflow_id, functions, run_example), failed, cancelled, or needs_guidance (the build is the caller's to finish with writ_website_to_api mode=guided build_id=<id>; its map comes with it).

Input parameters:

- `build_id` (integer, required)
- `timeout_seconds` (integer): Ceiling for wait=true (up to 75).
- `wait` (boolean): One held call until done or parked.

### `writ_browser_act` (~251 tokens)

Act in a browser session

Runs one batch of actions on an open session and returns the fresh page. Each action is an object, e.g. {"action":"click","selector":"#go"}. DRIVE: navigate, click, fill, type_text, select, check, hover, submit, press_key, scroll, back, wait, wait_for. SEE: query_dom, find_text, read_text, count, get_attributes, inspect, list_candidates, get_dom, get_screenshot. NETWORK: capture_network, get_request, rotate_exit. CODE: evaluate_js, search_scripts, read_script. RECORD: extract {variable,script}, api_call, login_post, probe_write, confirm_write. 2FA: twofa {challenge_method}. A value written {{name}} with its real value in `inputs` becomes a workflow input. Each action's arguments: the session's start answer, or writ_help('writ_browser_act').

Input parameters:

- `actions` (array, required): Ordered action objects, e.g. [{"action":"click","selector":"#login"}].
- `inputs` (object)
- `max_chars` (integer): Clip of the returned page (default 40000).
- `session_id` (string, required)

### `writ_browser_context` (~90 tokens)

Read a browser session

Re-reads an open session: section=page (url, fields, buttons, cleaned DOM), lists (the page's repeating rows with a live-tested define_function payload for writ_browser_compose), map (the evidence of the build the session is bound to).

Input parameters:

- `section` (string): page (default) | lists | map | explorer | concierge_api.
- `session_id` (string, required)

### `writ_browser_network` (~127 tokens)

Read a session's network calls

Searches (operation=search, query, method) or reads in full (operation=detail, index) the backend requests the page made after capture_network: the site's real API. writ_browser_act api_call probes one with other values (a `flow`: several requests); an index feeds writ_browser_compose define_function from_index.

Input parameters:

- `index` (integer): detail: which call.
- `method` (string): HTTP method filter.
- `operation` (string): search (default) | detail.
- `query` (string): Substring filter.
- `session_id` (string, required)

### `writ_browser_compose` (~140 tokens)

Compose a workflow in a session

Authors the workflow in an open session beyond a recording: named API functions (define_function with a `flow` (preferred in an API build), from a captured call's index, or a lists payload; live-tested as defined; compile_function for an authenticated request), set_inputs, add_steps, test_function, remove_function, remove_step, list. Payload shapes: writ_help('writ_browser_compose').

Input parameters:

- `operation` (string, required): list | define_function | compile_function | test_function | remove_function | add_steps | remove_step | set_inputs.
- `payload` (object): The operation's arguments.
- `session_id` (string, required)

### `writ_browser_save` (~149 tokens)

Save a session as a workflow

Saves the open session as a replayable workflow and closes the browser, in the place its browser ran: the user's desktop library or Writ Cloud. Composed functions become callable API functions, declared inputs become its parameters. The answer carries the steps, the inputs and a run_example; the workflow runs at zero AI cost through writ_run_workflow. A save that references something no run could resolve is refused with the reasons, and the browser stays open.

Input parameters:

- `allow_no_data` (boolean): Save a workflow that returns no data, on purpose.
- `description` (string)
- `name` (string)
- `save_to` (string)
- `session_id` (string, required)

### `writ_browser_ask_user` (~193 tokens)

Ask the user to step in

Hands an open session to the Writ user: kind='captcha' (they complete a security check in the live browser), 'twofa' (they supply the one-time code and Writ types it; the code never reaches this conversation), 'sign_in' (they sign in on the live page; no persona applies) or 'question' (their text answer). The call that opens the ask returns at once with user_link + tell_user to relay; later calls hold up to 60 s and answer answered, waiting_for_user (the same call again) or expired. Their answer is kept if your turn ends first.

Input parameters:

- `kind` (string): captcha | twofa | sign_in | question.
- `question` (string): kind='question': one short sentence.
- `session_id` (string, required)
- `wait_seconds` (integer): Hold up to this long (1-60).

### `writ_browser_cancel` (~75 tokens)

Close a browser session

Closes an open browser session (an open cloud browser bills until it is closed). Unsaved work that did more than visit pages is auto-saved first and the reply names it; discard=true keeps nothing.

Input parameters:

- `discard` (boolean): Throw away unsaved steps instead of auto-saving.
- `session_id` (string, required)

### `writ_browser_sessions` (~33 tokens)

List open browser sessions

Lists the account's open cloud browser sessions (each resumable by its session_id).

Input parameters:

- `limit` (integer)

### `writ_personas` (~287 tokens)

Use the user's saved sign-ins

The user's own accounts on websites (personas): how Writ signs in as them. No password passes through this tool or the conversation, and its answers never ask for one: credentials are typed only into a Writ window. action='list' domain=<site> shows the personas usable there; with none it answers persona_needed: a tell_user for the user, a create_url (a small Writ window with only the sign-in form, pre-filled for the site) and a link_id. action='wait' link_id=<id> is one held call that answers when the user saves it, with its persona_id. Also get, request (a link for another account), sign_in (refresh the session) and record_login. A persona_id works on writ_scrape, writ_crawl_site, writ_browser_use and writ_run_workflow; a desktop persona ('device:...') signs in on that desktop, from its own vault.

Input parameters:

- `action` (string, required): list (default) | get | request | wait | sign_in | record_login.
- `domain` (string): list: personas usable on this site. request: the site.
- `link_id` (string): wait: the link_id of a persona_needed / request answer.
- `persona_id` (integer)
- `why` (string): request: one line the user sees in the window.

### `writ_payment` (~189 tokens)

Pay with a card the user approves

Pays on a site with one of the user's cards; card numbers never reach this conversation. request (site, max_amount, purpose) returns a tell_user and an open_url where the user approves; wait (grant_id or confirmation_id); checkout pauses at the order button for the user's confirmation, then Writ places the order; fill; list (card handles only). Details: writ_help('writ_payment').

Input parameters:

- `action` (string, required): request | wait | checkout | fill | list.
- `commit_selector` (string): checkout: the place-order button.
- `confirmation_id` (string)
- `grant_id` (string)
- `max_amount` (number): The most it may charge.
- `purpose` (string): One line the user reads.
- `session` (string): The browser session on the checkout page.
- `site` (string): request: the store's domain.

### `writ_devices` (~77 tokens)

Choose a linked desktop

The user's linked Writ desktops: list, use (scope this connection to one: runs, browsers, personas and monitors then target it), clear, monitors, datasets.

Input parameters:

- `action` (string): list (default) | use | clear | monitors | datasets.
- `device` (string): use: the desktop's agent_id or name.

### `writ_list_workflows` (~69 tokens)

List saved workflows

Lists the workflows saved in the Writ account; each runs on demand without live browsing. Returns id, name, declared inputs, schedule, and whether it is pinned as its own run_<name> tool.

Input parameters:

- `search` (string): Name/description filter, e.g. a site.

### `writ_run_workflow` (~266 tokens)

Run a saved workflow

Runs a saved workflow by workflow_id or name and returns its extracted data (it waits by default). Inputs go as top-level fields or under `inputs`; function_name runs one function of a multi-function API (with the sign-in it needs). max_age reuses a recent result at no cost. A write function previews with mutation_mode='dry_run'. Runs sign in as the workflow's persona or persona_id; no password is ever passed. Venue, network, files and output shape: writ_help('writ_run_workflow').

Input parameters:

- `execution_target` (string): cloud | auto | local (the user's own desktop app).
- `function_name` (string): One function of a multi-function workflow.
- `inputs` (object): Run inputs (top-level fields work too).
- `max_age` (integer): Reuse a result younger than this many seconds (free). 0 = fresh.
- `mutation_mode` (string): live (default) | dry_run (preview, nothing sent) | private_test.
- `persona_id` (integer|string): Run as this saved identity.
- `workflow` (string): Workflow name (or workflow_id).
- `workflow_id` (integer|string): A number, or 'local:<id>' (a workflow on the user's desktop).

### `writ_update_workflow` (~79 tokens)

Edit a saved workflow

Inspects or edits a saved workflow: an outline, one function's source, targeted function or step edits, a settings patch. copy_to keeps a linked copy on Writ Cloud or the user's desktop. Every option: writ_help('writ_update_workflow').

Input parameters:

- `copy_to` (string)
- `workflow_id` (integer)

### `writ_create_http_extraction` (~84 tokens)

Build an HTTP extraction

Drafts or revises a multi-request browserless HTTP extraction (loops, GraphQL, cursor pagination) from captured requests and a plain-language goal. Every option: writ_help('writ_create_http_extraction').

Input parameters:

- `apply` (boolean): false (default) drafts; true writes.
- `goal` (string, required)
- `workflow_id` (integer)

### `writ_diagnose_http_workflow` (~59 tokens)

Diagnose a workflow's HTTP lane

Explains why a saved workflow's run returned nothing or failed: HTTP-lane readiness and, with task_id, the run's requests and responses.

Input parameters:

- `task_id` (integer)
- `workflow_id` (integer)

### `writ_pin_workflow_tool` (~47 tokens)

Pin a workflow as a tool

Pins (or unpins) a saved workflow as its own run_<name> tool on this server.

Input parameters:

- `pinned` (boolean)
- `workflow_id` (integer)

### `writ_workflow_data` (~82 tokens)

Read a workflow's collected data

Reads a saved workflow's collected data as a table: q filters, run_id picks one run, refs returns full records.

Input parameters:

- `limit` (integer)
- `q` (string)
- `refs` (array): Full records by '<run_id>:<record_index>'.
- `run_id` (integer)
- `workflow_id` (integer)

### `writ_search_data` (~47 tokens)

Search collected data

Searches everything the account's workflows already collected for a term, without running anything.

Input parameters:

- `limit` (integer)
- `q` (string, required)
- `workflow_id` (integer)

### `writ_export_data` (~43 tokens)

Export collected data

Exports a workflow's full data table as CSV or JSON.

Input parameters:

- `format` (string): csv (default) | json.
- `workflow_id` (integer)

### `writ_workflow_runs` (~54 tokens)

List workflow runs

Run history (status, timing, errors, venue) of one workflow or all of them.

Input parameters:

- `limit` (integer)
- `status` (string): Filter by status.
- `workflow_id` (integer)

### `writ_set_schedule` (~122 tokens)

Schedule a workflow

Schedules a saved workflow: every_minutes, or kind='daily'/'weekly' with time (HH:MM) and days. functions + inputs schedule chosen functions of a multi-function API.

Input parameters:

- `days` (array): Weekdays 1=Mon..7=Sun (weekly).
- `every_minutes` (integer): Interval in minutes.
- `kind` (string): interval | daily | weekly.
- `time` (string): HH:MM (daily/weekly).
- `tz` (string): IANA timezone.
- `workflow_id` (integer)

### `writ_expose_workflow_api` (~53 tokens)

Publish a workflow as a REST API

Publishes a saved workflow as a REST endpoint (a POST URL that waits and returns its JSON). Repeated calls reuse it.

Input parameters:

- `label` (string)
- `workflow_id` (integer)

### `writ_crawl_site` (~327 tokens)

Crawl a site into a dataset

COLLECTS a site or a section into a queryable, change-tracked dataset (one row per page). Scope: intent (a section, in plain words), rank_cap=N (the top N items of a listing), seed_urls (known pages), or the whole site capped by page_budget. Modes: markdown (default, no AI), extract_mode='schema' (uniform records), executor='ai' + extract_prompt (per-page understanding, 5x the page rate). A bounded crawl answers with its rows; an open one returns a crawl_id for writ_crawl_status wait=true. NOT FOR: reading a few pages now (writ_scrape). Every option: writ_help('writ_crawl_site').

Input parameters:

- `executor` (string): regular (default, no AI) | ai (5x the page rate).
- `extract_mode` (string): markdown (default) | schema | html.
- `extract_prompt` (string): With executor=ai: what to extract from each page.
- `intent` (string): The section wanted, in plain words; the server derives paths and depth.
- `page_budget` (integer): Cap on pages fetched.
- `persona_id` (integer|string)
- `rank_cap` (integer): Top N items of a listing (one page per agent, in parallel).
- `save_as` (string)
- `seed_urls` (array): Exact pages to collect (no discovery).
- `url` (string, required)
- `wait` (boolean): Hold until done and return the rows (default true when bounded).

### `writ_saved_crawls` (~31 tokens)

List saved crawls

Lists the crawls saved for re-running, with their scope.

Input parameters:

- `limit` (integer)

### `writ_update_saved_crawl` (~40 tokens)

Edit a saved crawl

Inspects or changes a saved crawl's settings in place.

Input parameters:

- `crawl` (string, required)
- `settings` (object): Sparse settings patch.

### `writ_run_saved_crawl` (~66 tokens)

Run a saved crawl

Runs a saved crawl; max_age returns its recent data at no cost instead of re-crawling.

Input parameters:

- `crawl` (string, required)
- `max_age` (integer): Reuse a run younger than this (seconds).
- `wait` (boolean): Hold until done.

### `writ_saved_crawl_data` (~41 tokens)

Read a saved crawl's data

Reads the data a saved crawl collected on its last completed run, without crawling.

Input parameters:

- `crawl` (string, required)
- `limit` (integer)

### `writ_scrape` (~292 tokens)

Read web pages

READ PAGES NOW, no browser session: one page (url), up to 20 pages in parallel (urls), or a listing plus its top N item pages (url + top_n). Returns clean markdown in 2-10 s, including pages a plain fetch cannot read: rendered by JavaScript (render_mode='browser'; a read that only got the page's shell says so and gives the retry), behind a bot wall or 403 (an automatic residential retry), or behind the user's sign-in (persona_id). Discussion threads keep their comments. format='html' returns the raw HTML for selectors. NOT FOR: a site-wide dataset (writ_crawl_site) or clicking and typing (writ_browser_use).

Input parameters:

- `format` (string): markdown (default) | html (raw HTML) | both.
- `persona_id` (integer|string): Read signed in as this saved identity (writ_personas).
- `render_mode` (string): auto (default: HTTP, a browser when the page needs JS) | http | browser.
- `top_n` (integer): With url = a listing: also read its N top-ranked item pages (max 20).
- `url` (string)
- `urls` (array): Up to 20 known pages, read in parallel.
- `use_residential` (boolean): Residential network for a bot wall (automatic retry when refused).

### `writ_crawl_status` (~65 tokens)

Check a crawl

Status of a crawl by crawl_id; wait=true holds up to 75 s and returns the collected rows when it converges.

Input parameters:

- `crawl_id` (integer, required)
- `limit` (integer)
- `wait` (boolean): Hold until done and inline the rows.

### `writ_crawl_files` (~50 tokens)

List a crawl's captured files

The original files (PDFs, office documents, images, CSVs) a crawl captured, with short-lived download URLs.

Input parameters:

- `crawl` (string)
- `crawl_id` (integer)

### `writ_create_automation` (~212 tokens)

Create an automation

Creates an automation: on an event (a workflow or crawl completed, a monitor change, a webhook, a schedule) it runs a workflow, sends a notification templated over the event's data, and/or wakes an AI agent. Event, template and webhook options: writ_help('writ_create_automation').

Input parameters:

- `ai_prompt` (string): Wake an AI agent with this task.
- `channels` (array): Notification channels.
- `name` (string, required)
- `notify` (string): Message template, e.g. {{result.extracted_data.0.title}}.
- `on_workflow` (string): Source workflow for workflow_* events.
- `run_workflow` (string): Workflow to run.
- `schedule` (object): when='scheduled': {kind, interval_minutes | time, days, tz}.
- `target_id` (integer): change_detected: the monitor_id.
- `when` (string): workflow_completed | crawl_completed | change_detected | webhook_received | scheduled | ...

### `writ_list_webhooks` (~35 tokens)

List inbound webhook URLs

The account's inbound webhook URLs and the automations they fire.

Input parameters:

- `webhook_trigger_id` (integer): Only this webhook.

### `writ_create_monitor` (~269 tokens)

Watch a page for changes

WATCHES a page for changes: Writ checks it on a schedule and fires change_detected when the page, a CSS selector's text, a value read by `extract`, or a visual zone changes. Returns the monitor_id that writ_wire_monitor connects to an alert, a workflow or an AI task (a price watch takes a threshold there). Without `interval` the answer is needs_input: this plan's options and a tell_user for the user, nothing created. The page is read first: a sign-in answers needs_persona, a bot check offers use_residential.

Input parameters:

- `extract` (object|string): Browserless read instead of a selector, e.g. {"from":"html_css","selector":".price"}.
- `interval` (string|integer): How often: an option id from needs_input ('5m', '1h', '24h') or seconds.
- `persona_id` (integer|string)
- `selector` (string): CSS selector whose text is watched (unset = whole page / uptime).
- `session_id` (string)
- `url` (string, required)
- `use_residential` (boolean): Check through a residential exit (bot walls).
- `watch` (string): content (default) | price (the value must hold a number).

### `writ_wire_monitor` (~249 tokens)

Connect a monitor to an action

Connects a monitor's change_detected event to an action: action='notify' (channels and recipients the account has configured), 'workflow' (runs a saved workflow), or 'ai_task' (an AI agent opens the page and works `prompt`). A price watch takes `threshold` (+ threshold_op): the action then runs once, when a check reads a price at or below it. Buying at a price (`buy`): writ_help('writ_wire_monitor').

Input parameters:

- `action` (string, required): notify | workflow | ai_task.
- `buy` (object): Run the workflow or task as a purchase (saved as a rehearsal until armed).
- `channels` (array): Notification channels, e.g. ["email","pushover"].
- `monitor_id` (integer, required)
- `prompt` (string): action='ai_task': what the agent does.
- `recipients` (array): e.g. ["email:3"]; omitted = every enabled recipient.
- `threshold` (number|string): Price watch: act once when the price reaches this number.
- `threshold_op` (string): lte (default) | lt | gte | gt.
- `workflow` (string)

### `writ_help` (~158 tokens)

Writ tool reference

The full reference of any Writ tool (every parameter and its rules) and the complete guide. topic = a tool name (e.g. 'writ_browser_compose', a 'run_<workflow>' tool), 'guide' for the manual (recurring digests, HTTP flow building, buying on a price, crawl and build modes), 'actions' / 'recording' / '2fa' for the browser rules, 'http_flows' / 'flow_operators' for writing an HTTP flow, or empty for the list of topics.

Input parameters:

- `topic` (string): A tool name, 'guide', 'actions', 'recording', '2fa', 'http_flows', 'flow_operators', or empty.

### `writ_browser_live_view` (~106 tokens)

Live browser view

The live browser card's own access: the card ChatGPT and Claude show under writ_browser_use and writ_record_website calls it to read the session's state and get one-time passes for the session's screencast and its take-over channel. The user watches the page in the card and can take it over there. It is not part of a conversation's work.

Input parameters:

- `session_id` (string, required): The session_id the card was opened with (cbs_...).

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/app-usewrit-writ-cloud/api#diagnostics

## Score history

- 2026-10-17: 76
- 2026-10-16: 76
- 2026-10-15: 75
- 2026-10-14: 75
- 2026-10-13: 74
- 2026-10-12: 73
- 2026-10-11: 73

## Common questions

### What is the Writ Cloud MCP server?

Writ Cloud is an MCP server listed in the public MCP registry as app.usewrit/writ-cloud. Read, crawl and act on websites, signed in as the user, and turn any site into an API. This page covers its hosted endpoint (https://api.usewrit.app/mcp).

### Is the Writ Cloud MCP server safe to use?

Writ Cloud scores 76 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Writ Cloud MCP server expose?

Writ Cloud exposes 41 tools: writ_browser_use, writ_record_website, writ_website_to_api, writ_discovery_status, writ_browser_act, and 36 more. Their descriptions and schemas cost roughly 5,359 tokens of context every time the server is loaded.

### Does the Writ Cloud MCP server require authentication?

Yes. Writ Cloud asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

### Is the Writ Cloud MCP server still maintained?

Writ Cloud is still listed as active in the MCP registry. We last reached this channel on 17 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://api.usewrit.app/mcp
- Repository: https://github.com/usewrit/writ-mcp
- Website: https://usewrit.app/
- Changelog RSS feed: https://verifymcp.io/servers/app-usewrit-writ-cloud/api.xml
- Changelog JSON feed: https://verifymcp.io/servers/app-usewrit-writ-cloud/api.json
- HTML version of this page: https://verifymcp.io/servers/app-usewrit-writ-cloud/api