SearXNG — Windows Native Port

A lightweight, 100% native Windows port of SearXNG running via Waitress WSGI for zero Docker overhead and no Redis/Valkey daemon requirement. Reason for making this was because I am working on a local AI harness and am thinking of directly integrating this engine into the harness itself.

---

🌟 Key Features

• Native Windows Execution: Runs purely in standard Python (3.10 – 3.14+) using the built-in Windows asyncio event loop.
• No Docker or WSL Required: Bypasses POSIX-only dependencies (uWSGI, uvloop).
• Zero Background Daemon Dependencies: Configured with server.limiter: false and valkey.url: false for immediate, single-command operation.
• Fast Multithreaded Serving: Powered by waitress WSGI server (configurable thread pool).
• JSON & HTML Support: Supports both browser searching and programmatic JSON API queries (/search?q=<query>&format=json).
• Windows Path Normalization: Fully patched template resolver and static file mapper for Windows filesystem separators (\ vs /).

---

🚀 Quickstart

1. Requirements

• Windows 10 / 11 (x64)
• Python 3.10+ (or uv)

2. Installation

# Using uv (recommended - ultra fast):
uv venv .venv
uv pip install --python .venv/Scripts/python.exe -r requirements-windows.txt

# Or using standard python:
python -m venv .venv
.\.venv\Scripts\pip install -r requirements-windows.txt

3. Running the Server

Option A: PowerShell Launcher

powershell -ExecutionPolicy Bypass -File .\start_searxng.ps1 -Port 8888

Option B: Direct Python Invocation

.\.venv\Scripts\python.exe run_server.py --host 127.0.0.1 --port 8888 --threads 6

---

🔍 API Usage

Query JSON API:

Invoke-RestMethod -Uri "http://127.0.0.1:8888/search?q=Rust+programming&format=json" -Method Get

Access Web UI:

Open http://127.0.0.1:8888 in your browser.

---

📄 License

This port is licensed under the GNU Affero General Public License v3.0 or later (AGPL-3.0-or-later), matching upstream SearXNG.