Consume AI coding tools with confidence
docker run --name codegate -d -p 8989:8989 -p 9090:9090 -p 8990:8990 --mount type=volume,src=codegate_volume,dst=/app/codegate_volume --restart unless-stopped ghcr.io/stacklok/codegate:latest
CodeGate emboldens developers to use AI tools with confidence by providing blanket security and integrated controls across many LLMs and coding assistants.
CodeGate is a guardian angel that sits between your AI assistants and LLMs, always shielding your privacy and supercharging your productivity.
We built CodeGate to be:
Local
CodeGate is local, never passing your secrets to a service or to the cloud. Your secrets are safest when they stay on your desktop.
Open
CodeGate is open source because solving for privacy requires transparency. You can see how CodeGate works and contribute to its evolution.
Quiet
CodeGate is largely invisible—it works in the background to keep you safe and only alerts you to secret leakage and security risks.
Here’s how CodeGate works:
CodeGate is a gateway that sits between your AI coding assistant and LLM. From this position, CodeGate can protect your privacy and code; but more importantly, it can create value for the developer that wants to use many assistants and many models to reduce costs, increase speed and improve results.
The questions we hope you’re asking…
The team at Stacklok built CodeGate. It was important to us that CodeGate be open source, so we can build this with the community.
We’ve launched CodeGate as an MVP so we can start collaborating with other open source developers from the jump. This is a solution for developers, by developers. Come build with us in our GitHub repository.
Every other gateway we’ve found suffers from three major shortcomings: (1) they live in the cloud, so your secrets don’t stay on your desktop, (2) they are built by security professionals for security teams who want to measure risk, but not action it, and (3) they are not open source and therefore lack transparency.
No! CodeGate is a partner to, not an enemy of, coding assistants. Stacklok engineers created CodeGate to ensure we could embrace coding assistants without losing privacy and control of our own development environment.
It’s actually the underlying LLMs that have stale training data. They’re so expensive and cumbersome to update that their training cutoffs are months or even years in the past; so, they don’t always know when a dependency has been deprecated or compromised.
Stacklok supports this use case; just get in touch with us at hello@stacklok.com for more details.
Loose lips sink software ships. Take control right now with a single container:
docker run --name codegate -d -p 8989:8989 -p 9090:9090 -p 8990:8990 --mount type=volume,src=codegate_volume,dst=/app/codegate_volume --restart unless-stopped ghcr.io/stacklok/codegate:latest
LINKS
SUPPORT
