Eyes and hands on real Windows PCs — observe, click, type via Glasswarp API.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Bearer Glasswarp API key (gw_…). Create one at https://www.glasswarp.com/console → API Keys.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Launch an executable on the remote Windows rig (name on PATH or absolute path), optional args. Side effects: starts a process; Glasswarp tracks it and closes it on session.end. Use for notepad.exe, mspaint.exe, chrome with URL args, etc. Wait/re-observe after launch before clicking — do not assume the window is focused immediately.
Active session id
Executable name (e.g. notepad.exe) or absolute path
Process args, e.g. ["--new-window", "https://example.com"]
The server marks this tool as destructive. Running it may delete or overwrite data on the server or in any account you connected with your headers.
Launch an executable on the remote Windows rig (name on PATH or absolute path), optional args. Side effects: starts a process; Glasswarp tracks it and closes it on session.end. Use for notepad.exe, mspaint.exe, chrome with URL args, etc. Wait/re-observe after launch before clicking — do not assume the window is focused immediately.
| Argument | Type | Description |
|---|---|---|
| args | array | Process args, e.g. ["--new-window", "https://example.com"] |
| session_id* | string | Active session id |
| path_or_name* | string | Executable name (e.g. notepad.exe) or absolute path |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"path_or_name"
],
"properties": {
"args": {
"type": "array",
"items": {
"type": "string"
},
"description": "Process args, e.g. [\"--new-window\", \"https://example.com\"]"
},
"session_id": {
"type": "string",
"description": "Active session id"
},
"path_or_name": {
"type": "string",
"description": "Executable name (e.g. notepad.exe) or absolute path"
}
}
}Return one showcase run contract (install, command, needs, framing). Does NOT execute the demo or control a PC. Call after demos.list when you know the demo_id. If the client can run shell, offer the command; if chat-only, show the card. Do not replace this with a slow MCP click loop for solver demos.
| Argument | Type | Description |
|---|---|---|
| format | string | Response format: markdown (default) or json |
| demo_id* | string | Showcase id from demos.list |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"demo_id"
],
"properties": {
"format": {
"enum": [
"markdown",
"json"
],
"type": "string",
"description": "Response format: markdown (default) or json"
},
"demo_id": {
"enum": [
"minesweeper",
"mona-lisa",
"paint",
"notepad"
],
"type": "string",
"description": "Showcase id from demos.list"
}
}
}List showcase run contracts (id, title, install, command). Read-only catalog — does NOT start a session, touch a rig, or run solvers. Use when the user asks for Minesweeper/Mona Lisa/Paint demos or you need the glasswarp-demo command. Prefer demos.get for one full card. For ad-hoc UI work use rigs.list → session.start → screen.observe instead.
| Argument | Type | Description |
|---|---|---|
| format | string | Response format: markdown (default) or json |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {
"format": {
"enum": [
"markdown",
"json"
],
"type": "string",
"description": "Response format: markdown (default) or json"
}
}
}Left/right/middle-click a UIA target by id from the latest screen.observe (uses native center coords). Prefer over input.click_xy. Side effect: real mouse click on the remote Windows desktop. Do not reuse target_id after the screen may have changed — re-observe first. For click→type→keys sequences, use input.send_actions (one turn) instead of chaining this tool.
| Argument | Type | Description |
|---|---|---|
| button | string | Mouse button (default left) |
| double | boolean | If true, double-click (two clicks) |
| target_id* | string | Target id from the most recent screen.observe (e.g. uia-… ) |
| session_id* | string | Active session id |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"target_id"
],
"properties": {
"button": {
"enum": [
"left",
"right",
"middle"
],
"type": "string",
"description": "Mouse button (default left)"
},
"double": {
"type": "boolean",
"description": "If true, double-click (two clicks)"
},
"target_id": {
"type": "string",
"description": "Target id from the most recent screen.observe (e.g. uia-… )"
},
"session_id": {
"type": "string",
"description": "Active session id"
}
}
}Click at native screen coordinates (0…native_width-1, 0…native_height-1 from screen.observe). Last resort when no suitable UIA target exists — prefer input.click_target. Never use JPEG/downscaled pixel coords. Side effect: real mouse click on the remote desktop.
| Argument | Type | Description |
|---|---|---|
| x* | integer | Native X (0…native_width-1) |
| y* | integer | Native Y (0…native_height-1) |
| button | string | Mouse button (default left) |
| double | boolean | If true, double-click |
| session_id* | string | Active session id |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"x",
"y"
],
"properties": {
"x": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native X (0…native_width-1)"
},
"y": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native Y (0…native_height-1)"
},
"button": {
"enum": [
"left",
"right",
"middle"
],
"type": "string",
"description": "Mouse button (default left)"
},
"double": {
"type": "boolean",
"description": "If true, double-click"
},
"session_id": {
"type": "string",
"description": "Active session id"
}
}
}Press-move-release mouse drag in native capture coordinates. Use for drawing, sliders, selection boxes, and drag-and-drop. Side effect: mouse_down → moves → mouse_up on the remote desktop. Prefer input.send_actions if the drag is one step in a longer predictable sequence.
| Argument | Type | Description |
|---|---|---|
| to_x* | integer | Native end X |
| to_y* | integer | Native end Y |
| from_x* | integer | Native start X |
| from_y* | integer | Native start Y |
| session_id* | string | Active session id |
| duration_ms | integer | Approx drag duration in ms; more steps when larger (default ~200) |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"from_x",
"from_y",
"to_x",
"to_y"
],
"properties": {
"to_x": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native end X"
},
"to_y": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native end Y"
},
"from_x": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native start X"
},
"from_y": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native start Y"
},
"session_id": {
"type": "string",
"description": "Active session id"
},
"duration_ms": {
"type": "integer",
"maximum": 5000,
"minimum": 0,
"description": "Approx drag duration in ms; more steps when larger (default ~200)"
}
}
}Move the cursor to native (x,y) then apply a vertical mouse-wheel delta. Side effect: scroll on whatever is under that point. Negative delta scrolls toward the bottom of the page. Prefer input.send_actions when scroll is part of a multi-step sequence. Re-observe after scrolling lists/pages before clicking targets.
| Argument | Type | Description |
|---|---|---|
| x* | integer | Native X to hover before scrolling |
| y* | integer | Native Y to hover before scrolling |
| delta* | integer | Vertical wheel delta (negative = toward bottom of page) |
| session_id* | string | Active session id |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"x",
"y",
"delta"
],
"properties": {
"x": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native X to hover before scrolling"
},
"y": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Native Y to hover before scrolling"
},
"delta": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": -9007199254740991,
"description": "Vertical wheel delta (negative = toward bottom of page)"
},
"session_id": {
"type": "string",
"description": "Active session id"
}
}
}PREFERRED multi-step tool: run 1–10 predictable UI actions in one call (input.click_target, input.click_xy, input.type_text, input.send_keys, input.drag, input.scroll). Side effects: all actions execute on the remote desktop; fails fast before sending if any action is invalid. observe_after defaults true (verification observe: text+targets; set observe_image=true for JPEG). Do not batch across unpredictable waits (page loads, installers, modals) — single-step those. Prefer this over chaining solo click/type/keys tools.
| Argument | Type | Description |
|---|---|---|
| actions* | array | Ordered actions (max 10); input.click_target needs target_id from latest screen.observe |
| session_id* | string | Active session id |
| observe_after | boolean | Default true. When true, return a verification observe in the same result |
| observe_image | boolean | Default false. When true with observe_after, include verification JPEG unless changed=false |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"actions"
],
"properties": {
"actions": {
"type": "array",
"items": {
"type": "object",
"required": [
"type"
],
"properties": {
"x": {
"type": "number"
},
"y": {
"type": "number"
},
"keys": {
"type": "string"
},
"text": {
"type": "string"
},
"to_x": {
"type": "number"
},
"to_y": {
"type": "number"
},
"type": {
"enum": [
"click_target",
"click_xy",
"type_text",
"send_keys",
"drag",
"scroll"
],
"type": "string"
},
"delta": {
"type": "number"
},
"button": {
"enum": [
"left",
"right",
"middle"
],
"type": "string"
},
"double": {
"type": "boolean"
},
"delay_ms": {
"type": "integer",
"maximum": 2000,
"minimum": 0,
"description": "Optional settle delay after this action (0–2000 ms)"
},
"target_id": {
"type": "string"
}
}
},
"maxItems": 10,
"minItems": 1,
"description": "Ordered actions (max 10); input.click_target needs target_id from latest screen.observe"
},
"session_id": {
"type": "string",
"description": "Active session id"
},
"observe_after": {
"type": "boolean",
"description": "Default true. When true, return a verification observe in the same result"
},
"observe_image": {
"type": "boolean",
"description": "Default false. When true with observe_after, include verification JPEG unless changed=false"
}
}
}Send a key or chord to the focused window (e.g. enter, tab, ctrl+s, alt+f4, win). Side effect: real key events on the remote desktop. Prefer bundling into input.send_actions when the shortcut follows a click/type in the same planned sequence. Use input.type_text for literal strings, not this tool.
| Argument | Type | Description |
|---|---|---|
| keys* | string | Key or combo, e.g. enter, tab, ctrl+s, alt+f4, win |
| session_id* | string | Active session id |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"keys"
],
"properties": {
"keys": {
"type": "string",
"description": "Key or combo, e.g. enter, tab, ctrl+s, alt+f4, win"
},
"session_id": {
"type": "string",
"description": "Active session id"
}
}
}Type a Unicode string into the currently focused control via native input. Does not click first — focus the field (input.click_target / input.send_actions) before calling. Side effect: keystrokes on the remote desktop. For form fills (click → type → tab/enter), prefer input.send_actions in one call.
| Argument | Type | Description |
|---|---|---|
| text* | string | Literal text to type (not a key combo) |
| session_id* | string | Active session id |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id",
"text"
],
"properties": {
"text": {
"type": "string",
"description": "Literal text to type (not a key combo)"
},
"session_id": {
"type": "string",
"description": "Active session id"
}
}
}List Windows machines (rigs) paired to this API key: id, name, online, api_access_enabled, and USABLE flag. Read-only — does not start a session. Call first before start_session. A rig is USABLE only when online AND the owner enabled API access. If none are USABLE, tell the user to install the host agent, pair in Console → Rigs, and enable API access — never ask for OS passwords.
No arguments.
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {}
}Read the current screen: UIA targets (numbered ids + native coords) and a text summary. Does not move mouse/keyboard. Default image=false (no JPEG) for speed; set image=true only when you must judge pixels visually (then max_width≈960, quality≈60). If changed=false, JPEG is omitted even when requested — do not re-analyze; wait or act differently. If dirty is null, assume changed. Prefer input.send_actions for multi-step UI; observe after meaningful steps, not after every click. Target ids are valid only until the next UI change.
| Argument | Type | Description |
|---|---|---|
| mark | boolean | Overlay numbered targets on JPEG when image=true (default true; ignored if image=false) |
| image | boolean | Include JPEG (default false). True only to visually read/judge the screen. |
| quality | integer | JPEG quality 40–100 when image=true (default 60) |
| max_width | integer | JPEG max width when image=true (default 960). Clicks always use native coords, not JPEG pixels. |
| session_id* | string | Active session id from session.start |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id"
],
"properties": {
"mark": {
"type": "boolean",
"description": "Overlay numbered targets on JPEG when image=true (default true; ignored if image=false)"
},
"image": {
"type": "boolean",
"description": "Include JPEG (default false). True only to visually read/judge the screen."
},
"quality": {
"type": "integer",
"maximum": 100,
"minimum": 40,
"description": "JPEG quality 40–100 when image=true (default 60)"
},
"max_width": {
"type": "integer",
"maximum": 3840,
"minimum": 320,
"description": "JPEG max width when image=true (default 960). Clicks always use native coords, not JPEG pixels."
},
"session_id": {
"type": "string",
"description": "Active session id from session.start"
}
}
}End an active session. Side effects: stops billing, runs host safety_restore, closes apps launched via app.launch. Always call when finished or abandoning — do not leave sessions open. Safe to call once; further observe/input on that session_id will fail.
| Argument | Type | Description |
|---|---|---|
| session_id* | string | Session id returned by session.start |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id"
],
"properties": {
"session_id": {
"type": "string",
"description": "Session id returned by session.start"
}
}
}Return the console Live View URL (≈60fps) for the rig owner to watch and intervene. Read-only for the agent — does not grant the API key console access. Offer on long or sensitive tasks. Owner must be signed into Glasswarp; API keys alone cannot open the player.
| Argument | Type | Description |
|---|---|---|
| session_id* | string | Active session id |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id"
],
"properties": {
"session_id": {
"type": "string",
"description": "Active session id"
}
}
}Start a metered desktop session on a USABLE rig from rigs.list. Side effects: begins wall-clock billing, shows an on-screen “API session active” indicator, enables observe/input until session.end. Idle sessions auto-end after ~15 minutes. Always call session.end when done or abandoning. Do not call if no USABLE rig exists. Returns session_id and Live View URL (owner console login required).
| Argument | Type | Description |
|---|---|---|
| rig_id* | string | Rig id from rigs.list (must be online with api_access_enabled) |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"rig_id"
],
"properties": {
"rig_id": {
"type": "string",
"description": "Rig id from rigs.list (must be online with api_access_enabled)"
}
}
}Fetch session metadata: status, host, mode, created_at, action_count, billed_minutes. Read-only — no input side effects. Use to tell the user about metered time or confirm the session is still active before more actions.
| Argument | Type | Description |
|---|---|---|
| session_id* | string | Session id from session.start |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"session_id"
],
"properties": {
"session_id": {
"type": "string",
"description": "Session id from session.start"
}
}
}