Skip to content

Repository files navigation

BigFix Ops Studio

BigFix Ops Studio is a localized, AI-powered IDE designed specifically for IT and Security professionals managing BigFix environments. It provides a seamless interface for writing, evaluating, and debugging endpoint management scripts with built-in, context-aware AI assistance.

The studio is organized into three core workspaces:

  • Client Relevance Studio — Query endpoint properties locally
  • Session Relevance Studio — Session-based queries for the BigFix Console
  • Action Script Studio — Write and reference BigFix Action Script snippets

✨ Key Features

  • Local Evaluation Engine: Safely evaluate complex relevance expressions locally without sending infrastructure queries to the cloud.
  • Semantic RAG Engine: Uses a local ChromaDB Vector Database to mathematically match your question to the most relevant snippets in the knowledge base, ensuring accurate answers while minimising token consumption.
  • Multi-Provider AI Assistant: Connect to your preferred AI provider to generate scripts, debug errors, and answer BigFix-specific questions. Supports Groq, Google Gemini, and Anthropic Claude.
  • Rate-Limit Resilience: Built-in cascading fallback logic automatically switches to secondary AI models if the primary model hits an API rate limit, ensuring uninterrupted workflow.
  • SmartSense Code Explanation: Highlight any complex relevance statement and instantly generate a plain-English explanation of exactly how it works.
  • Standalone Desktop App: Runs completely natively on Windows via an Electron wrapper — no Python or Node.js installation required on end-user machines.

🚀 Getting Started

Prerequisites

  • Windows OS (required for local BigFix evaluation and Registry key storage)
  • Python 3.8+ (for developers building from source)
  • An API key from one of the supported AI providers (Groq, Google Gemini, or Anthropic Claude)

Installation

Download the latest release. The release contains:

  • BigFix Ops Studio Setup.exe — Windows installer
  • How_to_Use_BigFix_Ops_Studio.docx — User guide

Run the installer and follow the on-screen prompts. No Python or Node.js installation is required on end-user machines.

Note: On first launch, the AI knowledge base syncs in the background. The application window appears immediately — full AI capability is available within a few seconds.

Setting Up Your AI Provider

BigFix Ops Studio requires an API key to power the AI Assistant. Open a Command Prompt in the application folder and inject your key via the command line. It is saved securely to your Windows Registry for all future sessions.

Groq (Recommended — fastest, generous free tier)

"BigFix Ops Studio.exe" --provider=groq --groq-api-key="YOUR_KEY_HERE"

Google Gemini (Recommended — most generous free quota at 1M TPM)

"BigFix Ops Studio.exe" --provider=gemini --gemini-api-key="YOUR_KEY_HERE"

Anthropic Claude (Best reasoning quality)

"BigFix Ops Studio.exe" --provider=claude --claude-api-key="YOUR_KEY_HERE"

Standard Usage

Once your API key is saved, simply double-click BigFix Ops Studio.exe to launch — no extra flags needed.

Tip: To switch providers, re-run with --provider= and the new provider name. To replace an expired key, re-run with the appropriate --xxx-api-key= flag.


🛠️ Building from Source

Prerequisites

  • Python 3.8+
  • Node.js
  • PyInstaller (pip install pyinstaller)

Step 1 — Install Dependencies

pip install -r requirements.txt

Step 2 — Build

Run the provided PowerShell build script from the root directory:

.\build.ps1

This freezes the Python backend into dist\server.exe, installs Electron dependencies, and packages everything into a Windows installer at:

electron\release\BigFix Ops Studio Setup.exe

Dev tip: Set the environment variable BIGFIX_DEBUG=1 before launching to enable verbose logging. Production builds log at WARNING level only.


📁 Architecture & Directory Structure

bigfix-ops-studio/
├── frontend/                  # Web Interface (HTML, CSS, JS)
│   ├── index.html             # Main IDE UI layout
│   ├── styles.css             # Application styling
│   └── app.js                 # Frontend logic & AI Chat UI
│
├── backend/                   # Python Core
│   ├── server.py              # HTTP routing (/chat, /evaluate, /ping)
│   ├── config.py              # Env variables, key management, paths
│   ├── ai_handler.py          # LLM routing, fallback logic, ChromaDB RAG
│   └── qna_engine.py          # Local qna.exe bridge
│
├── electron/                  # Desktop wrapper
│   ├── main.js                # Electron entry point
│   └── package.json           # Electron build config
│
├── .chroma_db/                # Vector DB storage (auto-generated)
├── library.enc                # Encrypted AI snippet library
├── inspectors.enc             # Encrypted syntax dictionary
├── backend_entry.py           # PyInstaller entry point
├── build.ps1                  # Full build script
└── requirements.txt           # Python dependencies

🔒 Security & Data Privacy

Area Detail
Evaluations All relevance and Action Script evaluations run entirely on the local machine via qna_engine.py. No infrastructure data is ever sent to the AI.
AI Chat The semantic embedding model runs 100% locally. Only text actively typed into the AI Chat or highlighted via SmartSense is sent to the configured LLM provider.
Key Storage API keys are stored in HKEY_CURRENT_USER Windows Registry. They are never written to plain-text files and never transmitted outside the provider's own API.
Logging Production builds log at WARNING level only. Set BIGFIX_DEBUG=1 for verbose output during development.

❓ Troubleshooting

Symptom Resolution
Port 5055 in use on relaunch The application auto-detects and clears stale processes on port 5055 at startup. If the error persists, manually kill any remaining BigFix Ops Studio processes.
App slow to load on first launch The ChromaDB vector database syncs in the background. The window appears immediately; AI will be ready within a few seconds.
AI error: cannot import 'genai' Ensure google-genai is installed: pip install google-genai
Engine not found Ensure qna.exe is in the same folder as the .exe, or that BESClient is installed and registered in the Windows Registry.
Rate limit errors All models for the selected provider are rate-limited. Wait 60 seconds, or switch to Gemini (1M TPM free quota).
AI response garbled Rephrase your prompt or restart the assistant panel.

About

BigFix Ops Studio

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors

Languages