Skip to content
| Marketplace
Sign in
Visual Studio Code>Programming Languages>ManulMcpServerNew to Visual Studio Code? Get it now.
ManulMcpServer

ManulMcpServer

alexbeatnik

| (0) | Free
VS Code extension that publishes ManulEngine as an MCP server and supports .hunt authoring.
Installation
Launch VS Code Quick Open (Ctrl+P), paste the following command, and press enter.
Copied to clipboard
More Info

ManulMcpServer

Status: Alpha

VS Code extension that turns ManulEngine into a native MCP server available directly in GitHub Copilot chat. Write .hunt automation scripts, run them from the editor, or invoke them through Copilot using natural language.

Alpha. Developed alongside ManulEngine — both projects are in active development. API and file formats are subject to change.


ManulEngine

This extension requires ManulEngine — the deterministic web and desktop automation runtime that powers the .hunt DSL.

PyPI PyPI
VS Code Marketplace VS Code Marketplace
GitHub alexbeatnik/ManulEngine
Status Alpha — battle-tested on real-world DOMs, APIs may evolve

What It Does

  • MCP server — registers ManulMcpServer in VS Code's MCP Servers view so Copilot can drive a real browser via chat
  • Hunt language support — syntax highlighting, IntelliSense, hover docs, and inline diagnostics for .hunt files
  • Run from editor — run button in the editor title bar, command palette, and status bar for .hunt files
  • Natural language input — type click login or fill email with test@example.com and the extension normalizes it to proper DSL before sending to the engine

Requirements

  • VS Code 1.110 or newer
  • Python 3.10+ with ManulEngine installed:
pip install manul-engine
playwright install

A workspace-local .venv is automatically detected and used if present.


Installation

Install the .vsix file:

code --install-extension manul-mcp-server-0.0.1.vsix

Then Reload Window (Ctrl+Shift+P → Developer: Reload Window).

After reload, ManulMcpServer appears in the MCP Servers panel and Copilot chat gains the Manul tools automatically.


MCP Tools Available in Copilot Chat

Tool What it does
manul_run_step Run a single DSL step or natural-language action in the browser
manul_run_goal Convert a natural-language goal into steps and execute them
manul_run_hunt Run a full .hunt document passed as text
manul_run_hunt_file Run a .hunt file from disk
manul_validate_hunt Validate a .hunt document without running it
manul_normalize_step Preview how a step will be normalized to DSL
manul_get_state Get current browser and session state
manul_preview_goal Preview goal-to-DSL conversion without execution
manul_scan_page List all interactive elements on the current page
manul_save_hunt Save a .hunt file to disk

Hunt File Quick Start

Create a file with the .hunt extension and write your automation:

@context: Login flow smoke test
@title: SauceDemo Login

STEP 1: Login
    NAVIGATE to 'https://www.saucedemo.com/'
    VERIFY that 'Username' is present
    Fill 'Username' field with 'standard_user'
    Fill 'Password' field with 'secret_sauce'
    Click the 'login-button' button
    VERIFY that 'Sauce Labs Backpack' is present

DONE.

Click the ▷ button in the editor title bar to run it.


Configuration

Open Settings (Ctrl+,) and search for manul:

Setting Default Description
manul.pythonPath python3 Python executable. Leave as python3 to auto-detect a local .venv.
manul.headless false Run browser in headless mode.
manul.requestTimeoutMs 60000 Timeout in ms for engine calls.
manul.logNormalizedDsl true Log auto-corrected DSL in the output panel.
manul.mcpServerLabel ManulMcpServer Label shown in the MCP Servers view.
manul.apiBaseUrl http://127.0.0.1:8000 Base URL for the ManulEngine HTTP API used by the editor Run Step and Run Hunt File commands.

Note: The editor title bar Run commands (Run Step, Run Hunt File) connect to the ManulEngine HTTP API at manul.apiBaseUrl. You need a running ManulEngine server for those commands to work:

manul serve

MCP tools (manul_run_step, manul_run_goal, etc.) use the bundled Python runner directly and do not require a separate HTTP server.


Hunt File Format

@context: <context for the AI planner>
@title: <suite name>

STEP 1: <description>
    NAVIGATE to 'https://...'
    VERIFY that 'element' is present
    Fill 'Field' field with 'value'
    Click the 'Button' button
    VERIFY that 'result' is present

DONE.

Rules:

  • @context:, @title:, STEP N:, and DONE. are flush-left
  • Action lines inside a STEP are indented with 4 spaces
  • Lines starting with # are comments

What's New

  • Initial release: MCP server integration, .hunt language support, Python runner bridge, manul_scan_page and manul_save_hunt tools.

License

Apache-2.0

  • Contact us
  • Jobs
  • Privacy
  • Manage cookies
  • Terms of use
  • Trademarks
© 2026 Microsoft