📺 Watch the full tutorial on YouTube

📚 WikiStream

Visual Interface for OpenWiki with Local Ollama Models

📖 Overview

WikiStream is a lightweight, high-performance visual interface built with Streamlit that wraps the powerful OpenWiki CLI tool. By directly integrating with your local Ollama server, WikiStream allows you to generate and maintain comprehensive codebase documentation completely offline and securely, without relying on external APIs.

With a fixed focus on performance, this tool utilizes the qwen3:4b local model (with reasoning/thinking disabled for maximum speed) to deliver rapid documentation generation directly to your outputs.md.

---

🚀 Tech Stack

• Frontend/UI: Streamlit
• Documentation Engine: OpenWiki CLI (npx openwiki)
• Local AI Provider: Ollama (OpenAI-compatible endpoints)
• Primary Model: qwen3:4b

---

⚡ Setup & Installation

Follow these steps in your Windows PowerShell terminal to get started immediately.

1. Prerequisites

Ensure you have the following installed:

• Python 3.9+
• Node.js & npm (for OpenWiki)
• Ollama with the qwen3:4b model pulled (ollama run qwen3:4b)

2. Clone & Install Dependencies

# Create a virtual environment
python -m venv venv
.\venv\Scripts\activate

# Install Python requirements
pip install -r requirements.txt

# Install OpenWiki globally (required for fast execution)
npm install -g openwiki

---

🎯 Run Instructions

Execute the following command to launch the visual interface:

streamlit run app.py

This will open your default web browser to http://localhost:8501. From the interface, you can enter the path to any local directory and a prompt, then click Generate Documentation to run OpenWiki.

---

📁 File Structure

• app.py: The core Streamlit application providing the visual interface for OpenWiki.
• requirements.txt: Contains minimal Python dependencies (streamlit).
• README.md: This comprehensive documentation file.
• outputs.md: The generated markdown documentation created by the OpenWiki execution.

---

💡 5 Use Cases

1. Legacy Code Onboarding: Point WikiStream at a massive, undocumented legacy repository and instantly generate a high-level architectural overview.
2. Automated Sprint Documentation: Run WikiStream at the end of every sprint to document the new features added to the codebase.
3. API Endpoint Summaries: Request OpenWiki to specifically document all REST API routes and their request/response schemas.
4. Security Audits: Generate a breakdown of authentication and authorization flows across the project to assist security researchers.
5. Offline Development: Generate documentation in air-gapped or restricted network environments since all processing is handled locally via Ollama.

---

🔮 5 Future Features

1. Batch Processing: Ability to select multiple repositories and generate documentation sequentially.
2. Git Auto-Commit: Automatically stage and commit the generated documentation directly back to the target repository.
3. Custom Model Selector: Dropdown in the UI to seamlessly switch between different local Ollama models (e.g., qwen2.5-coder:1.5b, llama3.2:3b).
4. LangSmith Integration UI: Visual toggle and input field to enable LangSmith tracing for OpenWiki runs directly from the dashboard.
5. Interactive Chat Mode: Incorporate OpenWiki's interactive chat mode into the Streamlit UI, allowing back-and-forth documentation refinement.

---

Keywords: OpenWiki, Ollama, qwen3:4b, Local LLM, Offline Documentation, Streamlit UI, AI Agent, Codebase Documentation, Python, Machine Learning, OpenRouter, Self-Hosted AI, Visual Interface, RAG, GitHub Actions.