Writ Cloud is an MCP server published by app.usewrit, hosted at api.usewrit.app. When last checked on 2026-11-22 it was working: it answers and lists its tools. It declares 41 tools, including writ_browser_use, writ_record_website, writ_website_to_api, writ_discovery_status, writ_browser_act.
Read, crawl and act on websites, signed in as the user, and turn any site into an API.
2 connections through 5 shared values. Tap any node.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
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.
| Parameter | Type | Description |
|---|---|---|
| url * | string | |
| goal | string | |
| auth_mode | string | one of ["reuse","fresh_login"] |
| fresh_exit | boolean | |
| persona_id | ["integer","string"] | |
| idempotency_key | string | |
| use_residential | boolean | Residential exit (bot walls, another country). Omitted: Writ's residential list decides. Pair with residential_country. |
| execution_target | string | cloud | auto | local (the user's own desktop app). |
| residential_country | string |
{
"name": "writ_browser_use",
"title": "Open a browser",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string"
},
"goal": {
"type": "string"
},
"auth_mode": {
"enum": [
"reuse",
"fresh_login"
],
"type": "string"
},
"fresh_exit": {
"type": "boolean"
},
"persona_id": {
"type": [
"integer",
"string"
]
},
"idempotency_key": {
"type": "string"
},
"use_residential": {
"type": "boolean",
"description": "Residential exit (bot walls, another country). Omitted: Writ's residential list decides. Pair with residential_country."
},
"execution_target": {
"type": "string",
"description": "cloud | auto | local (the user's own desktop app)."
},
"residential_country": {
"type": "string"
}
}
},
"annotations": {
"title": "Open a browser",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}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.
| Parameter | Type | Description |
|---|---|---|
| url * | string | |
| goal * | string | |
| device | string | |
| auth_mode | string | one of ["reuse","fresh_login"] |
| fresh_exit | boolean | |
| persona_id | ["integer","string"] | |
| skip_existing | boolean | |
| use_residential | boolean | Residential exit (bot walls, another country). Omitted: Writ's residential list decides. Pair with residential_country. |
| skip_marketplace | boolean | |
| residential_country | string |
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.
| Parameter | Type | Description |
|---|---|---|
| url | string | |
| goal | string | |
| device | string | |
| save_as | string | |
| build_id | integer | |
| anonymous | boolean | |
| persona_id | integer | |
| skip_existing | boolean | Skip the proposal of the user's own matching workflows. |
| idempotency_key | string |
Status of a website-to-API build (a guided one, or one started in the Writ app). 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 build_id=<id>; its map comes with it).
| Parameter | Type | Description |
|---|---|---|
| wait | boolean | One held call until done or parked. |
| build_id * | integer | |
| timeout_seconds | integer | Ceiling for wait=true (up to 75). |
{
"name": "writ_discovery_status",
"title": "Check an API build",
"description": "Status of a website-to-API build (a guided one, or one started in the Writ app). 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 build_id=<id>; its map comes with it).",
"inputSchema": {
"type": "object",
"required": [
"build_id"
],
"properties": {
"wait": {
"type": "boolean",
"description": "One held call until done or parked."
},
"build_id": {
"type": "integer"
},
"timeout_seconds": {
"type": "integer",
"description": "Ceiling for wait=true (up to 75)."
}
}
},
"annotations": {
"title": "Check an API build",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}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 (a flow is built step by step with writ_browser_compose add_flow_step), 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').
| Parameter | Type | Description |
|---|---|---|
| inputs | object | |
| actions * | array | Ordered action objects, e.g. [{"action":"click","selector":"#login"}]. |
| max_chars | integer | Clip of the returned page (default 40000). |
| session_id * | string |
{
"name": "writ_browser_act",
"title": "Act in a browser session",
"description": "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 (a flow is built step by step with writ_browser_compose add_flow_step), 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').",
"inputSchema": {
"type": "object",
"required": [
"session_id",
"actions"
],
"properties": {
"inputs": {
"type": "object"
},
"actions": {
"type": "array",
"items": {
"type": "object"
},
"description": "Ordered action objects, e.g. [{\"action\":\"click\",\"selector\":\"#login\"}]."
},
"max_chars": {
"type": "integer",
"description": "Clip of the returned page (default 40000)."
},
"session_id": {
"type": "string"
}
}
},
"annotations": {
"title": "Act in a browser session",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}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).
| Parameter | Type | Description |
|---|---|---|
| offset | integer | |
| section | string | page (default) | lists | map | explorer | concierge_api. |
| max_chars | integer | |
| session_id * | string |
{
"name": "writ_browser_context",
"title": "Read a browser session",
"description": "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).",
"inputSchema": {
"type": "object",
"required": [
"session_id"
],
"properties": {
"offset": {
"type": "integer",
"minimum": 0
},
"section": {
"enum": [
"page",
"lists",
"map",
"explorer",
"concierge_api"
],
"type": "string",
"description": "page (default) | lists | map | explorer | concierge_api."
},
"max_chars": {
"type": "integer"
},
"session_id": {
"type": "string"
}
}
},
"annotations": {
"title": "Read a browser session",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}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.
| Parameter | Type | Description |
|---|---|---|
| index | integer | detail: which call. |
| query | string | Substring filter. |
| method | string | HTTP method filter. |
| offset | integer | |
| max_chars | integer | |
| operation | string | search (default) | detail. |
| session_id * | string |
{
"name": "writ_browser_network",
"title": "Read a session's network calls",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"session_id"
],
"properties": {
"index": {
"type": "integer",
"minimum": 0,
"description": "detail: which call."
},
"query": {
"type": "string",
"description": "Substring filter."
},
"method": {
"type": "string",
"description": "HTTP method filter."
},
"offset": {
"type": "integer",
"minimum": 0
},
"max_chars": {
"type": "integer"
},
"operation": {
"enum": [
"search",
"detail"
],
"type": "string",
"description": "search (default) | detail."
},
"session_id": {
"type": "string"
}
}
},
"annotations": {
"title": "Read a session's network calls",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Authors the workflow in an open session beyond a recording: named API functions (define_function with a `flow`, preferred in an API build, or 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. A flow is built in small calls: define_function with its first request, then add_flow_step {function, step, at?} per step, each tested and answered with the step list; a set step's value is read as {ref:'vars.<name>'}. Shapes: writ_help('writ_browser_compose').
| Parameter | Type | Description |
|---|---|---|
| payload | object | The operation's arguments; add_flow_step: {function, step, at?}, remove_flow_step: {function, at}. |
| operation * | string | The action; add_flow_step, replace_flow_step and remove_flow_step change one step of a function's flow. |
| session_id * | string |
{
"name": "writ_browser_compose",
"title": "Compose a workflow in a session",
"description": "Authors the workflow in an open session beyond a recording: named API functions (define_function with a `flow`, preferred in an API build, or 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. A flow is built in small calls: define_function with its first request, then add_flow_step {function, step, at?} per step, each tested and answered with the step list; a set step's value is read as {ref:'vars.<name>'}. Shapes: writ_help('writ_browser_compose').",
"inputSchema": {
"type": "object",
"required": [
"session_id",
"operation"
],
"properties": {
"payload": {
"type": "object",
"description": "The operation's arguments; add_flow_step: {function, step, at?}, remove_flow_step: {function, at}."
},
"operation": {
"enum": [
"list",
"define_function",
"compile_function",
"test_function",
"remove_function",
"add_flow_step",
"replace_flow_step",
"remove_flow_step",
"add_steps",
"remove_step",
"set_inputs"
],
"type": "string",
"description": "The action; add_flow_step, replace_flow_step and remove_flow_step change one step of a function's flow."
},
"session_id": {
"type": "string"
}
}
},
"annotations": {
"title": "Compose a workflow in a session",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}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.
| Parameter | Type | Description |
|---|---|---|
| name | string | |
| listing | object | Its Marketplace title, summary, tags, category and intent: people-facing, English, no build status. |
| save_to | string | one of ["desktop","cloud","both"] |
| keep_open | boolean | |
| session_id * | string | |
| description | string | |
| allow_no_data | boolean | Save a workflow that returns no data, on purpose. |
| contribute_to_catalog | boolean |
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.
| Parameter | Type | Description |
|---|---|---|
| kind | string | captcha | twofa | sign_in | question. |
| question | string | kind='question': one short sentence. |
| session_id * | string | |
| wait_seconds | integer | Hold up to this long (1-60). |
{
"name": "writ_browser_ask_user",
"title": "Ask the user to step in",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"session_id"
],
"properties": {
"kind": {
"enum": [
"captcha",
"question",
"twofa",
"sign_in"
],
"type": "string",
"description": "captcha | twofa | sign_in | question."
},
"question": {
"type": "string",
"description": "kind='question': one short sentence."
},
"session_id": {
"type": "string"
},
"wait_seconds": {
"type": "integer",
"description": "Hold up to this long (1-60)."
}
}
},
"annotations": {
"title": "Ask the user to step in",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}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.
| Parameter | Type | Description |
|---|---|---|
| discard | boolean | Throw away unsaved steps instead of auto-saving. |
| session_id * | string |
{
"name": "writ_browser_cancel",
"title": "Close a browser session",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"session_id"
],
"properties": {
"discard": {
"type": "boolean",
"description": "Throw away unsaved steps instead of auto-saving."
},
"session_id": {
"type": "string"
}
}
},
"annotations": {
"title": "Close a browser session",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": true
}
}Lists the account's open cloud browser sessions (each resumable by its session_id).
| Parameter | Type | Description |
|---|---|---|
| limit | integer | |
| include_closed | boolean |
{
"name": "writ_browser_sessions",
"title": "List open browser sessions",
"description": "Lists the account's open cloud browser sessions (each resumable by its session_id).",
"inputSchema": {
"type": "object",
"properties": {
"limit": {
"type": "integer"
},
"include_closed": {
"type": "boolean"
}
}
},
"annotations": {
"title": "List open browser sessions",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}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.
| Parameter | Type | Description |
|---|---|---|
| why | string | request: one line the user sees in the window. |
| wait | boolean | |
| force | boolean | |
| action * | string | list (default) | get | request | wait | sign_in | record_login. |
| device | string | |
| domain | string | list: personas usable on this site. request: the site. |
| link_id | string | wait: the link_id of a persona_needed / request answer. |
| login_url | string |
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').
| Parameter | Type | Description |
|---|---|---|
| site | string | request: the store's domain. |
| steps | array | |
| action * | string | request | wait | checkout | fill | list. |
| fields | object | |
| purpose | string | One line the user reads. |
| session | string | The browser session on the checkout page. |
| summary | string | |
| currency | string | |
| grant_id | string | |
| max_amount | number | The most it may charge. |
The user's linked Writ desktops: list, use (scope this connection to one: runs, browsers, personas and monitors then target it), clear, monitors, datasets.
| Parameter | Type | Description |
|---|---|---|
| action | string | list (default) | use | clear | monitors | datasets. |
| device | string | use: the desktop's agent_id or name. |
{
"name": "writ_devices",
"title": "Choose a linked desktop",
"description": "The user's linked Writ desktops: list, use (scope this connection to one: runs, browsers, personas and monitors then target it), clear, monitors, datasets.",
"inputSchema": {
"type": "object",
"properties": {
"action": {
"enum": [
"list",
"use",
"clear",
"monitors",
"datasets"
],
"type": "string",
"description": "list (default) | use | clear | monitors | datasets."
},
"device": {
"type": "string",
"description": "use: the desktop's agent_id or name."
}
}
},
"annotations": {
"title": "Choose a linked desktop",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Lists the workflows saved in the Writ account; each runs on demand without live browsing. Returns id, name, declared inputs (with `input_rules`: the values and format an input accepts), schedule, whether it is pinned as its own run_<name> tool, and, for an API with several functions, their names (`functions`, `write_functions`): call one with function_name.
| Parameter | Type | Description |
|---|---|---|
| search | string | Name/description filter, e.g. a site. |
{
"name": "writ_list_workflows",
"title": "List saved workflows",
"description": "Lists the workflows saved in the Writ account; each runs on demand without live browsing. Returns id, name, declared inputs (with `input_rules`: the values and format an input accepts), schedule, whether it is pinned as its own run_<name> tool, and, for an API with several functions, their names (`functions`, `write_functions`): call one with function_name.",
"inputSchema": {
"type": "object",
"properties": {
"search": {
"type": "string",
"description": "Name/description filter, e.g. a site."
}
}
},
"annotations": {
"title": "List saved workflows",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Runs a saved workflow by workflow_id or name and returns its extracted data (it waits by default). The answer's `data` is what the run returned and `records_path` names its records; pass `output` with only the fields the question needs (fields: ['price', 'data.date as date']) for a far smaller answer. 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').
| Parameter | Type | Description |
|---|---|---|
| wait | boolean | |
| files | object | |
| device | string | |
| inputs | object | Run inputs (top-level fields work too). |
| output | object | |
| max_age | integer | Reuse a result younger than this many seconds (free). 0 = fresh. |
| workflow | string | Workflow name (or workflow_id). |
| persona_id | ["integer","string"] | Run as this saved identity. |
| workflow_id | ["integer","string"] |
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').
| Parameter | Type | Description |
|---|---|---|
| path | string | |
| limit | integer | |
| patch | object | Settings and metadata to change; {"name": "New name"} renames it. |
| device | string | |
| offset | integer | |
| copy_to | string | one of ["cloud","desktop"] |
| section | string | one of ["source","flow","contract"] |
| step_id | string | |
| verbose | boolean | |
| operator | string | |
| workflow | string | Name or id; local:<n> for a desktop workflow. |
| max_chars | integer |
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').
| Parameter | Type | Description |
|---|---|---|
| goal * | string | |
| apply | boolean | false (default) drafts; true writes. |
| workflow | string | |
| step_index | integer | |
| workflow_id | integer | |
| requirements | string | |
| desired_inputs | array | |
| request_samples | array | |
| response_sample | string |
{
"name": "writ_create_http_extraction",
"title": "Build an HTTP extraction",
"description": "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').",
"inputSchema": {
"type": "object",
"required": [
"goal"
],
"properties": {
"goal": {
"type": "string"
},
"apply": {
"type": "boolean",
"description": "false (default) drafts; true writes."
},
"workflow": {
"type": "string"
},
"step_index": {
"type": "integer"
},
"workflow_id": {
"type": "integer"
},
"requirements": {
"type": "string"
},
"desired_inputs": {
"type": "array",
"items": {
"type": "string"
}
},
"request_samples": {
"type": "array",
"items": {
"type": "object"
}
},
"response_sample": {
"type": "string"
}
}
},
"annotations": {
"title": "Build an HTTP extraction",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}Explains why a saved workflow's run returned nothing or failed: HTTP-lane readiness and, with task_id, the run's requests and responses.
| Parameter | Type | Description |
|---|---|---|
| task_id | integer | |
| workflow | string | |
| workflow_id | integer |
{
"name": "writ_diagnose_http_workflow",
"title": "Diagnose a workflow's HTTP lane",
"description": "Explains why a saved workflow's run returned nothing or failed: HTTP-lane readiness and, with task_id, the run's requests and responses.",
"inputSchema": {
"type": "object",
"properties": {
"task_id": {
"type": "integer"
},
"workflow": {
"type": "string"
},
"workflow_id": {
"type": "integer"
}
}
},
"annotations": {
"title": "Diagnose a workflow's HTTP lane",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Pins (or unpins) a saved workflow as its own run_<name> tool on this server.
| Parameter | Type | Description |
|---|---|---|
| pinned | boolean | |
| workflow | string | |
| workflow_id | integer |
{
"name": "writ_pin_workflow_tool",
"title": "Pin a workflow as a tool",
"description": "Pins (or unpins) a saved workflow as its own run_<name> tool on this server.",
"inputSchema": {
"type": "object",
"properties": {
"pinned": {
"type": "boolean"
},
"workflow": {
"type": "string"
},
"workflow_id": {
"type": "integer"
}
}
},
"annotations": {
"title": "Pin a workflow as a tool",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Reads a saved workflow's collected data as a table: q filters, run_id picks one run, refs returns full records.
| Parameter | Type | Description |
|---|---|---|
| q | string | |
| refs | array | Full records by '<run_id>:<record_index>'. |
| view | string | |
| limit | integer | |
| device | string | |
| output | object | |
| run_id | integer | |
| workflow | string | |
| workflow_id | integer | |
| preview_chars | integer |
{
"name": "writ_workflow_data",
"title": "Read a workflow's collected data",
"description": "Reads a saved workflow's collected data as a table: q filters, run_id picks one run, refs returns full records.",
"inputSchema": {
"type": "object",
"properties": {
"q": {
"type": "string"
},
"refs": {
"type": "array",
"items": {
"type": "string"
},
"description": "Full records by '<run_id>:<record_index>'."
},
"view": {
"type": "string"
},
"limit": {
"type": "integer"
},
"device": {
"type": "string"
},
"output": {
"type": "object"
},
"run_id": {
"type": "integer"
},
"workflow": {
"type": "string"
},
"workflow_id": {
"type": "integer"
},
"preview_chars": {
"type": "integer"
}
}
},
"annotations": {
"title": "Read a workflow's collected data",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Searches everything the account's workflows already collected for a term, without running anything.
| Parameter | Type | Description |
|---|---|---|
| q * | string | |
| limit | integer | |
| device | string | |
| workflow | string | |
| workflow_id | integer | |
| preview_chars | integer |
{
"name": "writ_search_data",
"title": "Search collected data",
"description": "Searches everything the account's workflows already collected for a term, without running anything.",
"inputSchema": {
"type": "object",
"required": [
"q"
],
"properties": {
"q": {
"type": "string"
},
"limit": {
"type": "integer"
},
"device": {
"type": "string"
},
"workflow": {
"type": "string"
},
"workflow_id": {
"type": "integer"
},
"preview_chars": {
"type": "integer"
}
}
},
"annotations": {
"title": "Search collected data",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Exports a workflow's full data table as CSV or JSON.
| Parameter | Type | Description |
|---|---|---|
| q | string | |
| view | string | |
| format | string | csv (default) | json. |
| workflow | string | |
| workflow_id | integer |
{
"name": "writ_export_data",
"title": "Export collected data",
"description": "Exports a workflow's full data table as CSV or JSON.",
"inputSchema": {
"type": "object",
"properties": {
"q": {
"type": "string"
},
"view": {
"type": "string"
},
"format": {
"type": "string",
"description": "csv (default) | json."
},
"workflow": {
"type": "string"
},
"workflow_id": {
"type": "integer"
}
}
},
"annotations": {
"title": "Export collected data",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Run history (status, timing, errors, venue) of one workflow or all of them.
| Parameter | Type | Description |
|---|---|---|
| limit | integer | |
| device | string | |
| status | string | Filter by status. |
| workflow | string | |
| workflow_id | integer |
{
"name": "writ_workflow_runs",
"title": "List workflow runs",
"description": "Run history (status, timing, errors, venue) of one workflow or all of them.",
"inputSchema": {
"type": "object",
"properties": {
"limit": {
"type": "integer"
},
"device": {
"type": "string"
},
"status": {
"type": "string",
"description": "Filter by status."
},
"workflow": {
"type": "string"
},
"workflow_id": {
"type": "integer"
}
}
},
"annotations": {
"title": "List workflow runs",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}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.
| Parameter | Type | Description |
|---|---|---|
| tz | string | IANA timezone. |
| days | array | Weekdays 1=Mon..7=Sun (weekly). |
| kind | string | interval | daily | weekly. |
| time | string | HH:MM (daily/weekly). |
| inputs | object | |
| enabled | boolean | |
| function | string | |
| workflow | string | |
| functions | array | |
| workflow_id | integer | |
| every_minutes | integer | Interval in minutes. |
Publishes a saved workflow as a REST endpoint (a POST URL that waits and returns its JSON). Repeated calls reuse it.
| Parameter | Type | Description |
|---|---|---|
| label | string | |
| workflow | string | |
| workflow_id | integer | |
| wait_timeout | integer | |
| timeout_seconds | integer |
{
"name": "writ_expose_workflow_api",
"title": "Publish a workflow as a REST API",
"description": "Publishes a saved workflow as a REST endpoint (a POST URL that waits and returns its JSON). Repeated calls reuse it.",
"inputSchema": {
"type": "object",
"properties": {
"label": {
"type": "string"
},
"workflow": {
"type": "string"
},
"workflow_id": {
"type": "integer"
},
"wait_timeout": {
"type": "integer",
"maximum": 300,
"minimum": 5
},
"timeout_seconds": {
"type": "integer",
"maximum": 300,
"minimum": 5
}
}
},
"annotations": {
"title": "Publish a workflow as a REST API",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}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').
| Parameter | Type | Description |
|---|---|---|
| url * | string | |
| name | string | |
| wait | boolean | Hold until done and return the rows (default true when bounded). |
| limit | integer | |
| speed | string | |
| device | string | |
| intent | string | The section wanted, in plain words; the server derives paths and depth. |
| output | object | |
| max_age | integer | |
| save_as |
Lists the crawls saved for re-running, with their scope.
| Parameter | Type | Description |
|---|---|---|
| limit | integer |
{
"name": "writ_saved_crawls",
"title": "List saved crawls",
"description": "Lists the crawls saved for re-running, with their scope.",
"inputSchema": {
"type": "object",
"properties": {
"limit": {
"type": "integer"
}
}
},
"annotations": {
"title": "List saved crawls",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}Inspects or changes a saved crawl's settings in place.
| Parameter | Type | Description |
|---|---|---|
| name | string | |
| crawl * | string | |
| settings | object | Sparse settings patch. |
| description | string | |
| default_max_age_seconds | integer |
{
"name": "writ_update_saved_crawl",
"title": "Edit a saved crawl",
"description": "Inspects or changes a saved crawl's settings in place.",
"inputSchema": {
"type": "object",
"required": [
"crawl"
],
"properties": {
"name": {
"type": "string"
},
"crawl": {
"type": "string"
},
"settings": {
"type": "object",
"description": "Sparse settings patch."
},
"description": {
"type": "string"
},
"default_max_age_seconds": {
"type": "integer",
"minimum": 0
}
}
},
"annotations": {
"title": "Edit a saved crawl",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": true
}
}Runs a saved crawl; max_age returns its recent data at no cost instead of re-crawling.
| Parameter | Type | Description |
|---|---|---|
| wait | boolean | Hold until done. |
| crawl * | string | |
| limit | integer | |
| output | object | |
| max_age | integer | Reuse a run younger than this (seconds). |
| preview_chars | integer | |
| timeout_seconds | integer |
{
"name": "writ_run_saved_crawl",
"title": "Run a saved crawl",
"description": "Runs a saved crawl; max_age returns its recent data at no cost instead of re-crawling.",
"inputSchema": {
"type": "object",
"required": [
"crawl"
],
"properties": {
"wait": {
"type": "boolean",
"description": "Hold until done."
},
"crawl": {
"type": "string"
},
"limit": {
"type": "integer"
},
"output": {
"type": "object"
},
"max_age": {
"type": "integer",
"minimum": 0,
"description": "Reuse a run younger than this (seconds)."
},
"preview_chars": {
"type": "integer"
},
"timeout_seconds": {
"type": "integer"
}
}
},
"annotations": {
"title": "Run a saved crawl",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}Reads the data a saved crawl collected on its last completed run, without crawling.
| Parameter | Type | Description |
|---|---|---|
| crawl * | string | |
| limit | integer | |
| output | object | |
| preview_chars | integer |
{
"name": "writ_saved_crawl_data",
"title": "Read a saved crawl's data",
"description": "Reads the data a saved crawl collected on its last completed run, without crawling.",
"inputSchema": {
"type": "object",
"required": [
"crawl"
],
"properties": {
"crawl": {
"type": "string"
},
"limit": {
"type": "integer"
},
"output": {
"type": "object"
},
"preview_chars": {
"type": "integer"
}
}
},
"annotations": {
"title": "Read a saved crawl's data",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}READ PAGES NOW, no browser session: one page (url), up to 20 pages in parallel (urls, on one site or several), 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; render_mode='http' opens no browser, so it never passes a check), or behind the user's sign-in (persona_id). Discussion threads keep their comments. format='html' returns the raw HTML for selectors, for one page or a batch. A page that cannot be read says why (error.code, e.g. scrape_not_found: the URL does not exist on the site) and gives retry_with when another read can help. NOT FOR: a site-wide dataset (writ_crawl_site) or clicking and typing (writ_browser_use).
| Parameter | Type | Description |
|---|---|---|
| url | string | |
| urls | array | Up to 20 known pages, read in parallel; they may be on different sites. |
| top_n | integer | With url = a listing: also read its N top-ranked item pages (max 20). |
| device | string | |
| format | string | markdown (default) | html (raw HTML) | both; with url, urls or top_n. |
| 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 or meets a check) | http (no browser) | browser. |
Status of a crawl by crawl_id; wait=true holds up to 75 s and returns the collected rows when it converges.
| Parameter | Type | Description |
|---|---|---|
| wait | boolean | Hold until done and inline the rows. |
| limit | integer | |
| output | object | |
| crawl_id * | integer | |
| preview_chars | integer | |
| timeout_seconds | integer |
{
"name": "writ_crawl_status",
"title": "Check a crawl",
"description": "Status of a crawl by crawl_id; wait=true holds up to 75 s and returns the collected rows when it converges.",
"inputSchema": {
"type": "object",
"required": [
"crawl_id"
],
"properties": {
"wait": {
"type": "boolean",
"description": "Hold until done and inline the rows."
},
"limit": {
"type": "integer"
},
"output": {
"type": "object"
},
"crawl_id": {
"type": "integer"
},
"preview_chars": {
"type": "integer"
},
"timeout_seconds": {
"type": "integer"
}
}
},
"annotations": {
"title": "Check a crawl",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}The original files (PDFs, office documents, images, CSVs) a crawl captured, with short-lived download URLs.
| Parameter | Type | Description |
|---|---|---|
| runs | integer | |
| crawl | string | |
| limit | integer | |
| crawl_id | integer |
{
"name": "writ_crawl_files",
"title": "List a crawl's captured files",
"description": "The original files (PDFs, office documents, images, CSVs) a crawl captured, with short-lived download URLs.",
"inputSchema": {
"type": "object",
"properties": {
"runs": {
"type": "integer"
},
"crawl": {
"type": "string"
},
"limit": {
"type": "integer"
},
"crawl_id": {
"type": "integer"
}
}
},
"annotations": {
"title": "List a crawl's captured files",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}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').
| Parameter | Type | Description |
|---|---|---|
| name * | string | |
| when | string | workflow_completed | crawl_completed | change_detected | webhook_received | scheduled | ... |
| title | string | |
| blocks | array | |
| inputs | object | |
| notify | string | Message template, e.g. {{result.extracted_data.0.title}}. |
| enabled | boolean | |
| channels | array | Notification channels. |
| priority | integer | |
| schedule | object | when='scheduled': {kind, interval_minutes | time, days, tz}. |
| ai_prompt | string |
The account's inbound webhook URLs and the automations they fire.
| Parameter | Type | Description |
|---|---|---|
| webhook_trigger_id | integer | Only this webhook. |
{
"name": "writ_list_webhooks",
"title": "List inbound webhook URLs",
"description": "The account's inbound webhook URLs and the automations they fire.",
"inputSchema": {
"type": "object",
"properties": {
"webhook_trigger_id": {
"type": "integer",
"description": "Only this webhook."
}
}
},
"annotations": {
"title": "List inbound webhook URLs",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}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.
| Parameter | Type | Description |
|---|---|---|
| url * | string | |
| mode | string | one of ["selector","visual"] |
| watch | string | content (default) | price (the value must hold a number). |
| device | string | |
| enabled | boolean | |
| 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. |
| selector | string | CSS selector whose text is watched (unset = whole page / uptime). |
| zone_text | string |
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').
| Parameter | Type | Description |
|---|---|---|
| buy | object | Run the workflow or task as a purchase (saved as a rehearsal until armed). |
| name | string | |
| title | string | |
| action * | string | notify | workflow | ai_task. |
| device | string | |
| prompt | string | action='ai_task': what the agent does. |
| enabled | boolean | |
| message | string | |
| channels | array | Notification channels, e.g. ["email","pushover"]. |
| workflow | string | |
| entry_url |
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 modes), 'actions' / 'recording' / '2fa' for the browser rules, 'http_flows' / 'flow_operators' for writing an HTTP flow, or empty for the list of topics.
| Parameter | Type | Description |
|---|---|---|
| topic | string | A tool name, 'guide', 'actions', 'recording', '2fa', 'http_flows', 'flow_operators', or empty. |
{
"name": "writ_help",
"title": "Writ tool reference",
"description": "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 modes), 'actions' / 'recording' / '2fa' for the browser rules, 'http_flows' / 'flow_operators' for writing an HTTP flow, or empty for the list of topics.",
"inputSchema": {
"type": "object",
"properties": {
"topic": {
"type": "string",
"description": "A tool name, 'guide', 'actions', 'recording', '2fa', 'http_flows', 'flow_operators', or empty."
}
}
},
"annotations": {
"title": "Writ tool reference",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}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.
| Parameter | Type | Description |
|---|---|---|
| session_id * | string | The session_id the card was opened with (cbs_...). |
{
"name": "writ_browser_live_view",
"title": "Live browser view",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"session_id"
],
"properties": {
"session_id": {
"type": "string",
"description": "The session_id the card was opened with (cbs_...)."
}
}
},
"annotations": {
"title": "Live browser view",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}WRIT IS FOR: the user's signed-in account on a site, a click/form on a site, a page a plain fetch cannot open (JS, 403), no usable API, an endpoint that outlives the chat, a job that repeats. TOOLS BY INTENT: - read a page, several pages, or a listing's top N -> writ_scrape (one call; bot walls and 403s are retried there, JavaScript pages read with render_mode='browser'). Reading never needs a browser session. - watch a page or a price -> writ_create_monitor, then writ_wire_monitor (alert, workflow or AI task); relay its tell_user. - a site as a callable API -> YOU build it: writ_website_to_api opens a browser bound to the build; capture_network, act like a user, read the calls (writ_browser_network), probe them (api_call); a failed probe is a clue, not a stop. Prefer an HTTP flow (define_function `flow`; writ_help('http_flows')) mixed with single requests (from_index); DOM extraction last. writ_browser_save, then its run_example on two inputs. - something already saved -> writ_run_workflow (or its run_<name> tool); writ_list_workflows(search=<site>) finds it, before any browser opens. - a whole site or a section as a dataset -> writ_crawl_site. - click, type, sign in, submit -> writ_browser_use, then writ_browser_act; writ_browser_save keeps it as a workflow, writ_browser_cancel closes it. RULES: - A browser turn costs ~20k tokens, a saved workflow ~1k: a task that may come again is saved ({{name}} values + `inputs`, then writ_browser_save) and proven on two inputs; a one-off is not saved. - Behind a login: writ_personas action='list' domain=<site> first, and pass its persona_id. Never ask for a password or a code in chat: relay persona_needed's tell_user and create_url, then writ_personas action='wait' link_id=<id>. - A CAPTCHA or a code Writ cannot mint -> writ_browser_ask_user. - A site that refuses Writ's servers -> the answer carries site_block: what gets past it on this account (the plan first, then Writ Desktop, then credits) and a tell_user for the user. - Never poll: tools that wait take wait=true (one held call, up to 75s). - A run that returns nothing -> writ_diagnose_http_workflow, then writ_update_workflow. - A call whose arguments arrive as one block of text is decoded by Writ when the text is JSON; when it is not, the answer names the broken character and nothing ran: resend it in smaller parts (a flow: define_function with its first request, then one add_flow_step per step). - A 'local' run whose desktop is offline moves to the cloud (off the user's machine and IP) only with the user's agreement. - Paying -> writ_payment; never ask for a card number. - Each answer names its next step and the rules for it. writ_help(topic) returns any tool's full reference; writ_help('guide') the complete manual (recurring digests, HTTP flow building, buying on a price, crawl modes). - This server is "Writ Cloud", the user's cloud account (sessions run on Writ's fleet). A server named "Writ" is their desktop app. On any surface a saved workflow comes before a new build.
Names, descriptions and schemas were written by the server and are shown as data.
Pattern matches on what the server declares. They show what it says it can do, not what it does.
None yet. Changes appear after a server's second daily check.
{
"name": "writ_record_website",
"title": "Record a website task",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"goal",
"url"
],
"properties": {
"url": {
"type": "string"
},
"goal": {
"type": "string"
},
"device": {
"type": "string"
},
"auth_mode": {
"enum": [
"reuse",
"fresh_login"
],
"type": "string"
},
"fresh_exit": {
"type": "boolean"
},
"persona_id": {
"type": [
"integer",
"string"
]
},
"skip_existing": {
"type": "boolean"
},
"use_residential": {
"type": "boolean",
"description": "Residential exit (bot walls, another country). Omitted: Writ's residential list decides. Pair with residential_country."
},
"skip_marketplace": {
"type": "boolean"
},
"residential_country": {
"type": "string"
}
}
},
"annotations": {
"title": "Record a website task",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}| use_residential | boolean |
| execution_target | string |
| skip_marketplace | boolean | Skip the marketplace proposals. |
| residential_country | string |
{
"name": "writ_website_to_api",
"title": "Turn a website into an API",
"description": "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.",
"inputSchema": {
"type": "object",
"properties": {
"url": {
"type": "string"
},
"goal": {
"type": "string"
},
"device": {
"type": "string"
},
"save_as": {
"type": "string"
},
"build_id": {
"type": "integer"
},
"anonymous": {
"type": "boolean"
},
"persona_id": {
"type": "integer"
},
"skip_existing": {
"type": "boolean",
"description": "Skip the proposal of the user's own matching workflows."
},
"idempotency_key": {
"type": "string"
},
"use_residential": {
"type": "boolean"
},
"execution_target": {
"type": "string"
},
"skip_marketplace": {
"type": "boolean",
"description": "Skip the marketplace proposals."
},
"residential_country": {
"type": "string"
}
}
},
"annotations": {
"title": "Turn a website into an API",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}{
"name": "writ_browser_save",
"title": "Save a session as a workflow",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"session_id"
],
"properties": {
"name": {
"type": "string"
},
"listing": {
"type": "object",
"description": "Its Marketplace title, summary, tags, category and intent: people-facing, English, no build status."
},
"save_to": {
"enum": [
"desktop",
"cloud",
"both"
],
"type": "string"
},
"keep_open": {
"type": "boolean"
},
"session_id": {
"type": "string"
},
"description": {
"type": "string"
},
"allow_no_data": {
"type": "boolean",
"description": "Save a workflow that returns no data, on purpose."
},
"contribute_to_catalog": {
"type": "boolean"
}
}
},
"annotations": {
"title": "Save a session as a workflow",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}| persona_id | integer |
| include_runs | boolean |
| execution_target | string | one of ["cloud"] |
{
"name": "writ_personas",
"title": "Use the user's saved sign-ins",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"action"
],
"properties": {
"why": {
"type": "string",
"description": "request: one line the user sees in the window."
},
"wait": {
"type": "boolean"
},
"force": {
"type": "boolean"
},
"action": {
"enum": [
"list",
"get",
"request",
"wait",
"sign_in",
"record_login"
],
"type": "string",
"description": "list (default) | get | request | wait | sign_in | record_login."
},
"device": {
"type": "string"
},
"domain": {
"type": "string",
"description": "list: personas usable on this site. request: the site."
},
"link_id": {
"type": "string",
"description": "wait: the link_id of a persona_needed / request answer."
},
"login_url": {
"type": "string"
},
"persona_id": {
"type": "integer"
},
"include_runs": {
"type": "boolean"
},
"execution_target": {
"enum": [
"cloud"
],
"type": "string"
}
}
},
"annotations": {
"title": "Use the user's saved sign-ins",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}| automation_id |
| integer |
| total_selector | string |
| commit_selector | string | checkout: the place-order button. |
| confirmation_id | string |
| payment_method_id | string |
{
"name": "writ_payment",
"title": "Pay with a card the user approves",
"description": "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').",
"inputSchema": {
"type": "object",
"required": [
"action"
],
"properties": {
"site": {
"type": "string",
"description": "request: the store's domain."
},
"steps": {
"type": "array",
"items": {
"type": "object"
}
},
"action": {
"enum": [
"request",
"wait",
"checkout",
"fill",
"list"
],
"type": "string",
"description": "request | wait | checkout | fill | list."
},
"fields": {
"type": "object"
},
"purpose": {
"type": "string",
"description": "One line the user reads."
},
"session": {
"type": "string",
"description": "The browser session on the checkout page."
},
"summary": {
"type": "string"
},
"currency": {
"type": "string"
},
"grant_id": {
"type": "string"
},
"max_amount": {
"type": "number",
"description": "The most it may charge."
},
"automation_id": {
"type": "integer"
},
"total_selector": {
"type": "string"
},
"commit_selector": {
"type": "string",
"description": "checkout: the place-order button."
},
"confirmation_id": {
"type": "string"
},
"payment_method_id": {
"type": "string"
}
}
},
"annotations": {
"title": "Pay with a card the user approves",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}| A number, or 'local:<id>' (a workflow on the user's desktop). |
| function_name | string | One function of a multi-function workflow; writ_list_workflows names them (`functions`). |
| mutation_mode | string | live (default) | dry_run (preview, nothing sent) | private_test. |
| function_names | array |
| idempotency_key | string |
| timeout_seconds | integer |
| use_residential | boolean |
| execution_target | string | cloud | auto | local (the user's own desktop app). |
| residential_country | string |
{
"name": "writ_run_workflow",
"title": "Run a saved workflow",
"description": "Runs a saved workflow by workflow_id or name and returns its extracted data (it waits by default). The answer's `data` is what the run returned and `records_path` names its records; pass `output` with only the fields the question needs (fields: ['price', 'data.date as date']) for a far smaller answer. 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').",
"inputSchema": {
"type": "object",
"properties": {
"wait": {
"type": "boolean"
},
"files": {
"type": "object"
},
"device": {
"type": "string"
},
"inputs": {
"type": "object",
"description": "Run inputs (top-level fields work too)."
},
"output": {
"type": "object"
},
"max_age": {
"type": "integer",
"minimum": 0,
"description": "Reuse a result younger than this many seconds (free). 0 = fresh."
},
"workflow": {
"type": "string",
"description": "Workflow name (or workflow_id)."
},
"persona_id": {
"type": [
"integer",
"string"
],
"description": "Run as this saved identity."
},
"workflow_id": {
"type": [
"integer",
"string"
],
"description": "A number, or 'local:<id>' (a workflow on the user's desktop)."
},
"function_name": {
"type": "string",
"description": "One function of a multi-function workflow; writ_list_workflows names them (`functions`)."
},
"mutation_mode": {
"enum": [
"live",
"dry_run",
"private_test"
],
"type": "string",
"description": "live (default) | dry_run (preview, nothing sent) | private_test."
},
"function_names": {
"type": "array",
"items": {
"type": "string"
}
},
"idempotency_key": {
"type": "string"
},
"timeout_seconds": {
"type": "integer"
},
"use_residential": {
"type": "boolean"
},
"execution_target": {
"type": "string",
"description": "cloud | auto | local (the user's own desktop app)."
},
"residential_country": {
"type": "string"
}
}
},
"annotations": {
"title": "Run a saved workflow",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}| workflow_id | integer |
| step_updates | array | [{index, patch}] merged into those recorded steps. |
| function_name | string |
| input_updates | array |
| replace_steps | array |
| validate_only | boolean | true previews and validates; nothing is saved. |
| function_updates | array | [{function_name, patch|json_edits|script_edits}] (add /config/flow/steps/- appends a step); add:{step} adds a function. |
| regenerate_skill | boolean |
| expected_updated_at | string |
| confirm_listing_change | boolean |
{
"name": "writ_update_workflow",
"title": "Edit a saved workflow",
"description": "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').",
"inputSchema": {
"type": "object",
"properties": {
"path": {
"type": "string"
},
"limit": {
"type": "integer",
"maximum": 100,
"minimum": 1
},
"patch": {
"type": "object",
"description": "Settings and metadata to change; {\"name\": \"New name\"} renames it."
},
"device": {
"type": "string"
},
"offset": {
"type": "integer",
"minimum": 0
},
"copy_to": {
"enum": [
"cloud",
"desktop"
],
"type": "string"
},
"section": {
"enum": [
"source",
"flow",
"contract"
],
"type": "string"
},
"step_id": {
"type": "string"
},
"verbose": {
"type": "boolean"
},
"operator": {
"type": "string"
},
"workflow": {
"type": "string",
"description": "Name or id; local:<n> for a desktop workflow."
},
"max_chars": {
"type": "integer",
"maximum": 20000,
"minimum": 1000
},
"workflow_id": {
"type": "integer"
},
"step_updates": {
"type": "array",
"items": {
"type": "object"
},
"description": "[{index, patch}] merged into those recorded steps."
},
"function_name": {
"type": "string"
},
"input_updates": {
"type": "array",
"items": {
"type": "object"
},
"maxItems": 50,
"minItems": 1
},
"replace_steps": {
"type": "array",
"items": {
"type": "object"
}
},
"validate_only": {
"type": "boolean",
"description": "true previews and validates; nothing is saved."
},
"function_updates": {
"type": "array",
"items": {
"type": "object"
},
"maxItems": 50,
"minItems": 1,
"description": "[{function_name, patch|json_edits|script_edits}] (add /config/flow/steps/- appends a step); add:{step} adds a function."
},
"regenerate_skill": {
"type": "boolean"
},
"expected_updated_at": {
"type": "string"
},
"confirm_listing_change": {
"type": "boolean"
}
}
},
"annotations": {
"title": "Edit a saved workflow",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": true
}
}{
"name": "writ_set_schedule",
"title": "Schedule a workflow",
"description": "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.",
"inputSchema": {
"type": "object",
"properties": {
"tz": {
"type": "string",
"description": "IANA timezone."
},
"days": {
"type": "array",
"items": {
"type": "integer"
},
"description": "Weekdays 1=Mon..7=Sun (weekly)."
},
"kind": {
"type": "string",
"description": "interval | daily | weekly."
},
"time": {
"type": "string",
"description": "HH:MM (daily/weekly)."
},
"inputs": {
"type": "object"
},
"enabled": {
"type": "boolean"
},
"function": {
"type": "string"
},
"workflow": {
"type": "string"
},
"functions": {
"type": "array",
"items": {
"type": "string"
}
},
"workflow_id": {
"type": "integer"
},
"every_minutes": {
"type": "integer",
"description": "Interval in minutes."
}
}
},
"annotations": {
"title": "Schedule a workflow",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": true
}
}| string |
| delay_ms | integer |
| executor | string | regular (default, no AI) | ai (5x the page rate). |
| ocr_mode | string |
| rank_cap | integer | Top N items of a listing (one page per agent, in parallel). |
| max_depth | integer |
| seed_urls | array | Exact pages to collect (no discovery). |
| persona_id | ["integer","string"] |
| shard_size | integer |
| page_budget | integer | Cap on pages fetched. |
| render_mode | string |
| same_domain | boolean |
| content_spec | object |
| extract_mode | string | markdown (default) | schema | html. |
| exclude_paths | array |
| include_paths | array |
| preview_chars | integer |
| extract_prompt | string | With executor=ai: what to extract from each page. |
| extract_schema | object | With extract_mode=schema: CSS {row_selector, fields: [{name, selector, attr?}]}, not a JSON Schema. |
| respect_robots | boolean |
| timeout_seconds | integer |
| use_residential | boolean |
| allow_subdomains | boolean |
| relevance_threshold | number |
| residential_country | string |
| max_concurrent_shards | integer |
{
"name": "writ_crawl_site",
"title": "Crawl a site into a dataset",
"description": "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').",
"inputSchema": {
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string"
},
"name": {
"type": "string"
},
"wait": {
"type": "boolean",
"description": "Hold until done and return the rows (default true when bounded)."
},
"limit": {
"type": "integer"
},
"speed": {
"type": "string"
},
"device": {
"type": "string"
},
"intent": {
"type": "string",
"description": "The section wanted, in plain words; the server derives paths and depth."
},
"output": {
"type": "object"
},
"max_age": {
"type": "integer",
"minimum": 0
},
"save_as": {
"type": "string"
},
"delay_ms": {
"type": "integer"
},
"executor": {
"type": "string",
"description": "regular (default, no AI) | ai (5x the page rate)."
},
"ocr_mode": {
"type": "string"
},
"rank_cap": {
"type": "integer",
"maximum": 50,
"minimum": 1,
"description": "Top N items of a listing (one page per agent, in parallel)."
},
"max_depth": {
"type": "integer"
},
"seed_urls": {
"type": "array",
"items": {
"type": "string"
},
"description": "Exact pages to collect (no discovery)."
},
"persona_id": {
"type": [
"integer",
"string"
]
},
"shard_size": {
"type": "integer"
},
"page_budget": {
"type": "integer",
"description": "Cap on pages fetched."
},
"render_mode": {
"type": "string"
},
"same_domain": {
"type": "boolean"
},
"content_spec": {
"type": "object"
},
"extract_mode": {
"type": "string",
"description": "markdown (default) | schema | html."
},
"exclude_paths": {
"type": "array",
"items": {
"type": "string"
}
},
"include_paths": {
"type": "array",
"items": {
"type": "string"
}
},
"preview_chars": {
"type": "integer"
},
"extract_prompt": {
"type": "string",
"description": "With executor=ai: what to extract from each page."
},
"extract_schema": {
"type": "object",
"description": "With extract_mode=schema: CSS {row_selector, fields: [{name, selector, attr?}]}, not a JSON Schema."
},
"respect_robots": {
"type": "boolean"
},
"timeout_seconds": {
"type": "integer"
},
"use_residential": {
"type": "boolean"
},
"allow_subdomains": {
"type": "boolean"
},
"relevance_threshold": {
"type": "number",
"maximum": 1,
"minimum": 0
},
"residential_country": {
"type": "string"
},
"max_concurrent_shards": {
"type": "integer"
}
}
},
"annotations": {
"title": "Crawl a site into a dataset",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}| include_paths | array |
| preview_chars | integer |
| respect_robots | boolean |
| use_residential | boolean | Residential network for a bot wall (automatic retry when refused). |
| residential_country | string |
{
"name": "writ_scrape",
"title": "Read web pages",
"description": "READ PAGES NOW, no browser session: one page (url), up to 20 pages in parallel (urls, on one site or several), 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; render_mode='http' opens no browser, so it never passes a check), or behind the user's sign-in (persona_id). Discussion threads keep their comments. format='html' returns the raw HTML for selectors, for one page or a batch. A page that cannot be read says why (error.code, e.g. scrape_not_found: the URL does not exist on the site) and gives retry_with when another read can help. NOT FOR: a site-wide dataset (writ_crawl_site) or clicking and typing (writ_browser_use).",
"inputSchema": {
"type": "object",
"properties": {
"url": {
"type": "string"
},
"urls": {
"type": "array",
"items": {
"type": "string"
},
"description": "Up to 20 known pages, read in parallel; they may be on different sites."
},
"top_n": {
"type": "integer",
"maximum": 20,
"minimum": 1,
"description": "With url = a listing: also read its N top-ranked item pages (max 20)."
},
"device": {
"type": "string"
},
"format": {
"enum": [
"markdown",
"html",
"both"
],
"type": "string",
"description": "markdown (default) | html (raw HTML) | both; with url, urls or top_n."
},
"persona_id": {
"type": [
"integer",
"string"
],
"description": "Read signed in as this saved identity (writ_personas)."
},
"render_mode": {
"type": "string",
"description": "auto (default: HTTP, a browser when the page needs JS or meets a check) | http (no browser) | browser."
},
"include_paths": {
"type": "array",
"items": {
"type": "string"
}
},
"preview_chars": {
"type": "integer"
},
"respect_robots": {
"type": "boolean"
},
"use_residential": {
"type": "boolean",
"description": "Residential network for a bot wall (automatic retry when refused)."
},
"residential_country": {
"type": "string"
}
}
},
"annotations": {
"title": "Read web pages",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}| Wake an AI agent with this task. |
| target_id | integer | change_detected: the monitor_id. |
| recipients | array |
| description | string |
| on_workflow | string | Source workflow for workflow_* events. |
| ai_entry_url | string |
| run_function | string |
| run_workflow | string | Workflow to run. |
| ai_session_id | integer |
| run_functions | array |
| on_workflow_id | integer |
| run_workflow_id | integer |
| cooldown_minutes | integer |
| target_selector_id | integer |
| webhook_trigger_id | integer |
{
"name": "writ_create_automation",
"title": "Create an automation",
"description": "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').",
"inputSchema": {
"type": "object",
"required": [
"name"
],
"properties": {
"name": {
"type": "string"
},
"when": {
"enum": [
"workflow_completed",
"workflow_started",
"crawl_completed",
"crawl_failed",
"ai_session_completed",
"ai_session_started",
"change_detected",
"webhook_received",
"scheduled"
],
"type": "string",
"description": "workflow_completed | crawl_completed | change_detected | webhook_received | scheduled | ..."
},
"title": {
"type": "string"
},
"blocks": {
"type": "array",
"items": {
"type": "object"
}
},
"inputs": {
"type": "object"
},
"notify": {
"type": "string",
"description": "Message template, e.g. {{result.extracted_data.0.title}}."
},
"enabled": {
"type": "boolean"
},
"channels": {
"type": "array",
"items": {
"type": "string"
},
"description": "Notification channels."
},
"priority": {
"type": "integer"
},
"schedule": {
"type": "object",
"description": "when='scheduled': {kind, interval_minutes | time, days, tz}."
},
"ai_prompt": {
"type": "string",
"description": "Wake an AI agent with this task."
},
"target_id": {
"type": "integer",
"description": "change_detected: the monitor_id."
},
"recipients": {
"type": "array",
"items": {
"type": "string"
}
},
"description": {
"type": "string"
},
"on_workflow": {
"type": "string",
"description": "Source workflow for workflow_* events."
},
"ai_entry_url": {
"type": "string"
},
"run_function": {
"type": "string"
},
"run_workflow": {
"type": "string",
"description": "Workflow to run."
},
"ai_session_id": {
"type": "integer"
},
"run_functions": {
"type": "array",
"items": {
"type": "string"
}
},
"on_workflow_id": {
"type": "integer"
},
"run_workflow_id": {
"type": "integer"
},
"cooldown_minutes": {
"type": "integer"
},
"target_selector_id": {
"type": "integer"
},
"webhook_trigger_id": {
"type": "integer"
}
}
},
"annotations": {
"title": "Create an automation",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}| persona_id | ["integer","string"] |
| session_id | string |
| try_anyway | boolean |
| request_url | string |
| use_residential | boolean | Check through a residential exit (bot walls). |
| interval_minutes | integer |
| requires_browser | boolean |
| residential_country | string |
{
"name": "writ_create_monitor",
"title": "Watch a page for changes",
"description": "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.",
"inputSchema": {
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string"
},
"mode": {
"enum": [
"selector",
"visual"
],
"type": "string"
},
"watch": {
"enum": [
"content",
"price"
],
"type": "string",
"description": "content (default) | price (the value must hold a number)."
},
"device": {
"type": "string"
},
"enabled": {
"type": "boolean"
},
"extract": {
"type": [
"object",
"string"
],
"description": "Browserless read instead of a selector, e.g. {\"from\":\"html_css\",\"selector\":\".price\"}."
},
"interval": {
"type": [
"string",
"integer"
],
"description": "How often: an option id from needs_input ('5m', '1h', '24h') or seconds."
},
"selector": {
"type": "string",
"description": "CSS selector whose text is watched (unset = whole page / uptime)."
},
"zone_text": {
"type": "string"
},
"persona_id": {
"type": [
"integer",
"string"
]
},
"session_id": {
"type": "string"
},
"try_anyway": {
"type": "boolean"
},
"request_url": {
"type": "string"
},
"use_residential": {
"type": "boolean",
"description": "Check through a residential exit (bot walls)."
},
"interval_minutes": {
"type": "integer"
},
"requires_browser": {
"type": "boolean"
},
"residential_country": {
"type": "string"
}
}
},
"annotations": {
"title": "Watch a page for changes",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}| string |
| max_steps | integer |
| threshold | ["number","string"] | Price watch: act once when the price reaches this number. |
| monitor_id * | integer |
| recipients | array | e.g. ["email:3"]; omitted = every enabled recipient. |
| workflow_id | integer |
| threshold_op | string | lte (default) | lt | gte | gt. |
| ai_session_id | integer |
| cooldown_minutes | integer |
{
"name": "writ_wire_monitor",
"title": "Connect a monitor to an action",
"description": "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').",
"inputSchema": {
"type": "object",
"required": [
"monitor_id",
"action"
],
"properties": {
"buy": {
"type": "object",
"description": "Run the workflow or task as a purchase (saved as a rehearsal until armed)."
},
"name": {
"type": "string"
},
"title": {
"type": "string"
},
"action": {
"enum": [
"workflow",
"notify",
"ai_task"
],
"type": "string",
"description": "notify | workflow | ai_task."
},
"device": {
"type": "string"
},
"prompt": {
"type": "string",
"description": "action='ai_task': what the agent does."
},
"enabled": {
"type": "boolean"
},
"message": {
"type": "string"
},
"channels": {
"type": "array",
"items": {
"type": "string"
},
"description": "Notification channels, e.g. [\"email\",\"pushover\"]."
},
"workflow": {
"type": "string"
},
"entry_url": {
"type": "string"
},
"max_steps": {
"type": "integer"
},
"threshold": {
"type": [
"number",
"string"
],
"description": "Price watch: act once when the price reaches this number."
},
"monitor_id": {
"type": "integer"
},
"recipients": {
"type": "array",
"items": {
"type": "string"
},
"description": "e.g. [\"email:3\"]; omitted = every enabled recipient."
},
"workflow_id": {
"type": "integer"
},
"threshold_op": {
"enum": [
"lt",
"lte",
"gt",
"gte"
],
"type": "string",
"description": "lte (default) | lt | gte | gt."
},
"ai_session_id": {
"type": "integer"
},
"cooldown_minutes": {
"type": "integer"
}
}
},
"annotations": {
"title": "Connect a monitor to an action",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}