AegisDB

AegisDB is a powerful, self-hosted database solution designed to provide AI agents with durable, long-term memory capabilities. Built in C and packaged as a single, lightweight binary, AegisDB operates privately within your infrastructure, ensuring that sensitive data remains secure and never leaks to third-party services. This multi-tenant, encrypted database supports essential features such as backups, read replicas, and a seamless integration with Claude Code, all while utilizing straightforward JSON-over-TCP communication.

Key Features

• Durable Memory Storage: AegisDB offers robust episodic and semantic memory storage using an append-only log structure that ensures data integrity through corruption-resilient recovery.
• Multi-Tenant Architecture: Designed for teams, it includes support for per-namespace scoped tokens, user authentication, and customizable rate limits to manage resource usage effectively.
• Easy Setup & Use: The application runs as a single binary without external dependencies; simply deploy it using Docker and interact with the built-in client commands.

Getting Started

Deploy AegisDB using a simple Docker command:

docker run -d --name aegisdb -p 9470:9470 -v aegis-data:/data ghcr.io/d4n-larsson/aegisdb:latest

Once running, perform operations like health checks, memory inserts, and searches effortlessly:

docker exec aegisdb aegisdb client ping
docker exec aegisdb aegisdb client put --type semantic --tags user "prefers dark mode"
docker exec aegisdb aegisdb client search --tags user --top-k 5

Observability & Monitoring

AegisDB comes equipped with built-in observability features, including a Grafana dashboard for real-time analytics of memory usage and operation metrics. To set up the observability stack, simply use:

docker compose --profile monitoring up

Integration with Claude Code

With AegisDB, giving Claude Code a persistent memory is as simple as executing a single command for seamless memory management. This powerful integration allows for storing and recalling information across interactions, significantly enhancing the AI's capabilities.

Security & Compliance

AegisDB is designed with security in mind. User data is encrypted at rest, and a configurable authentication model allows for stringent access controls, ensuring that users can securely manage their data without external exposure.

The system also features comprehensive documentation which includes guides on setting up a team server, wire protocol specifics, and an architecture overview to facilitate smooth deployment and operation.

Conclusion

In summary, AegisDB provides a compelling solution for teams looking to enhance the memory capabilities of their AI agents. By combining self-hosting, robust security features, and simple deployment, AegisDB empowers developers to build intelligent applications that rely on accurate, long-term memory storage.