GreyFox is a self-hosted, telemetry-free AI traffic tooling solution designed for local operator workflows. It acts as a powerful local proxy that intercepts and routes AI application traffic, enabling users to inspect, debug, and enforce policies on LLM data streams without sending operational data to the cloud. This platform is ideal for developers, researchers, and organizations prioritizing privacy and granular control over their AI integrations.
Key Features
- Telemetry-Free Control Plane: Ensures operational data remains self-hosted and private, never sent to the cloud.
- Local Proxy Integration: Sits between your application and OpenAI-compatible providers, routing only chosen traffic.
- LLM Data Stream Inspection: Allows detailed inspection and debugging of local AI application traffic.
- Policy Enforcement: Apply custom policies to manage proxy behavior and operator intent.
- Managed User Controls: Administer usage boundaries and operational states for known users (Essential edition).
- Advanced Analytics Workspace: Deep insights into tokens, cache, traffic sources, models, and latency (Essential edition).
Use Cases
GreyFox is invaluable for developers and teams conducting local AI experiments. By routing traffic through its local proxy, operators can inspect requests and responses, understand model interactions, and debug integrations in a controlled, private environment. This eliminates the need to rely on external services for monitoring, keeping sensitive development data secure.
For organizations with stringent data privacy requirements, GreyFox provides a critical self-hosted solution. It ensures that all LLM application traffic and operational telemetry remain within the user's Docker environment, preventing any data leakage to third-party cloud providers. This is crucial for compliance and maintaining data sovereignty.
Furthermore, GreyFox empowers operators to establish repeatable and controlled AI routing setups. With features like managed user controls, deterministic request/response presets, and advanced analytics, it bridges the gap between experimental utilities and productized operator workflows, allowing for consistent policy application and resource management across various AI providers.
Pricing Information
GreyFox offers a free, open-source Community Edition (0.1.1) for basic local AI traffic inspection and simple proxy control. The Essential edition (0.2.1) is a commercial subscription available monthly (EUR 19) or annually (EUR 179), providing advanced features like managed users, saved profiles, request presets, and deep analytics. A Pro edition with enterprise audit logs and programmable routing is currently in development. Subscriptions for paid editions are processed externally by Stripe.
User Experience and Support
The GreyFox interface is designed as a local operator panel, not a hosted surveillance layer, providing a clear and intuitive control surface for managing AI traffic workflows. It helps operators reason about local routing, users, providers, and usage signals directly within their self-hosted environment. The Community Edition 0.1.1 includes UI template and styling refinements. Support inquiries can be directed to
support@skilful-fox.com, and community discussions are encouraged.
Technical Details
GreyFox operates as a local proxy server, integrating with applications by simply changing the base URL of any OpenAI-compatible LLM client to point at the GreyFox endpoint. It runs within a Docker environment, ensuring self-containment and ease of deployment. The package is published via GitHub Container Registry, and examples often feature TypeScript for client-side integration.
Pros and Cons
- Pros: Telemetry-free by design, strong privacy focus, self-hosted control, detailed local traffic inspection, supports OpenAI-compatible providers, open-source Community Edition, clear commercial boundary.
- Cons: Requires Docker setup, manual base URL configuration, advanced features are behind a paywall, Pro edition is still under development, limited public documentation beyond feature descriptions.
Conclusion
GreyFox stands out as an essential tool for anyone seeking robust, privacy-preserving control over their local AI application traffic. Its self-hosted nature and comprehensive feature set empower operators to inspect, debug, and manage LLM data streams with confidence. Explore the Community Edition today to take command of your AI workflows.