← Latest products

Noisegate

a differential-privacy gateway for untrusted AI agents. Noisegate: a differential privacy gateway that lets an untrusted LLM agent query sensitive data over MCP (Model Context Protocol), with a formal guarantee no individual's record can leak even if the agent is adversarial - enforcement lives in trusted code below the model, validated by a runnable attack gallery.

Open source

28 stars · GitHubexternal site

Visit github.comexternal site

Reactions (1)

Similar products

by description similarity

  1. Halo

    github.com

    open-source, tamper-evident runtime evidence for AI agents

    Hacker News37 pts

  2. Context Gateway

    github.com

    Compress agent context before it hits the LLM

    Hacker News97 pts

  3. HoundDog.ai

    github.com

    Ultra-Fast Code Scanner for Data Privacy

    Hacker News16 pts

  4. An MCP Gateway to block the lethal trifecta

    github.com

    🔐 Firewall Your Data, Control Agents. Prevent agent data exfiltration. Gain visibility into AI's interactions with your data / systems of record / existing software. https://discord.gg/tXjATaKgTV

    Hacker News51 pts

  5. Open Bias

    github.com

    proxy that enforces agent behavior at runtime

    Hacker News21 pts