Where is the source code for Oswright?
The source code is linked from the repository URL on this page.
@Ask-812
Basic information
MCP server for Windows desktop automation. Incremental perception instead of a screenshot per step: measured 16.9x fewer tokens per action than the alternatives, with the benchmarks to check it.
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"oswright": {
"command": "uvx",
"args": [
"oswright"
]
}
}
}- Take a screenshot of the screen or a region. Returns the image as native MCP image content. Optionally saves to a file path.
- Get screen dimensions and monitor count.
- Find all occurrences of text on screen using OCR. Returns matches with coordinates and confidence.
- Read ALL visible text on the screen using OCR. Returns every detected text element with position.
- Find all occurrences of a template image on screen using OpenCV template matching.
- Click the mouse at coordinates or current position. Returns screenshot.
- Double-click at coordinates or current position. Returns screenshot.
- Move the mouse cursor to screen coordinates.
- Scroll the mouse wheel. Returns screenshot.
- Drag from one point to another. Returns screenshot.
- Get the current mouse cursor position.
- Type text character by character. Returns screenshot.
- Press a key or combo like `Enter`, `Ctrl+C`, `Alt+Tab`. Returns screenshot.
- Find text via OCR and click on it. Auto-retries until found or timeout. Returns screenshot.
- Find text via OCR and double-click on it. Returns screenshot.
- Find text via OCR and right-click on it. Returns screenshot.
- Find text via OCR and hover over it. Returns screenshot.
- Find a label, click it, clear, and type a value. Returns screenshot.
- Fill multiple fields in one call. Reduces round-trips.
- Wait for text to appear on screen. Polls via OCR.
- Wait for text to disappear from screen.
- Wait for a specified duration (capped at 30s), then screenshot.
- List all visible windows. Optionally filter by title substring.
- Bring a window to the foreground by title. Returns screenshot.
- Close a window by title (sends WM_CLOSE). Returns screenshot.
- Minimize a window by title. Returns screenshot.
- Capture a screenshot of just one window.
- Get the current text content of the system clipboard.
- Copy text to the system clipboard.
- Launch an application and optionally wait for it to load. Runs the program directly, never through a shell.
- Get info about the active OCR backend and available backends.
- Report what changed on screen since the last observation. Rescans only the regions that moved. Prefer this over `screenshot` for tracking state.
- Find on-screen text using the cheapest method that can answer. Reports which cascade rung responded.
- Find text via the cascade and click it. The cheap alternative to `click_text`.
- Read on-screen text from the incremental model without re-OCRing the display.
- Report how much perception work the model has avoided.
- Remember the current screen so future visits skip reading it. Persists across sessions.
- Report what the screen atlas has remembered and how often it helped.
- Get the accessibility tree of the focused window. Returns all interactive elements with names, types, positions. Deterministic and instant.
- Click a UI element using the accessibility tree. More reliable than OCR.
- Set the value of a UI element (e.g., text box). More reliable than OCR-based fill.
- Get info about the currently focused window.
- Wait for the screen to visually change. Takes a baseline screenshot, polls until different.
MCP server for Windows desktop automation. Incremental perception instead of a screenshot per step: measured 16.9x fewer tokens per action than the alternatives, with the benchmarks to check it.
The README includes setup instructions such as "command": "uvx",.
The source code is linked from the repository URL on this page.
If the README contains a parseable MCP configuration block, it is shown in the Config tab.

Connect AI assistants to Mailercloud, the email marketing platform — manage campaigns, contacts, lists, templates, and sending, plus AI-powered campaign analysis and deliverability audits. Endpoint: https://mcp.mailerclo

Day Off is an AI-powered PTO and time-tracking platform trusted by more than 50,000 companies worldwide. Set up your company and onboard your employees in just a few clicks. Track attendance and time spent on tasks and

Governed, read-only AI access to your enterprise email archive. Search years of correspondence across teams and mail systems in plain language.

Qorenext Trade screening MCP is a Model Context Protocol server that enables Claude and other AI clients to run sanctions screening and trade-screening due-diligence directly in chat.

EM+x turns your conversation into board-ready deliverables: steering-committee readouts, executive briefs, market scans, QBR decks, and program updates - each built by an agentic consulting team using your own templates
Comments