kindling-server

Daemon runtime for kindling — local-first memory for AI-assisted development.

kindling-server provides the daemon used by the kindling CLI and Rust client. It exposes kindling memory over a local HTTP/1 API, using Unix domain sockets where available and a TCP fallback on Windows.

Most users should not depend on this crate directly. Run the daemon with:

kindling serve

Use this crate when you are extending, embedding or testing the daemon layer itself.

When to use this crate

Use kindling-server if you are:

• Building a custom runtime around the kindling daemon.
• Testing daemon behaviour directly.
• Extending local API endpoints.
• Working on per-project database routing or process lifecycle behaviour.

For normal Rust integrations, use kindling-client.

Runtime role

The daemon exists to make kindling safe and predictable when multiple tools are interacting with project memory at the same time. It handles:

• Local API access.
• Project-aware routing.
• Store/service orchestration.
• Daemon lifecycle.
• Cross-tool coordination.

Documentation

Full docs: docs.eddacraft.ai/kindling

Relevant docs:

• Configuration
• Core concepts

Licence

Apache-2.0