Install TSUNAGI

What TSUNAGI Is

TSUNAGI is an independent Cardano node implementation written in Zig. It provides a deterministic ledger pipeline with shadow verification, designed to follow and decode Cardano blocks with full transparency into every stage of processing.

The runtime includes self-healing infrastructure subsystems — KAGAMI for state mirroring, YAMORI for anomaly detection and recovery, TATE for vertical integrity checks, and KURA for persistent storage management. Each subsystem exposes its own state, giving operators complete observability into node behavior.

TSUNAGI is operator-focused. It is built for people who want to understand exactly what their node is doing, why it is doing it, and how to verify every output.

Current Status

TSUNAGI is under active, continuous engineering. This section is an honest assessment of where the project stands.

• Active development — the codebase is under continuous engineering with regular architectural changes
• Preview network operational — TSUNAGI follows and decodes Cardano blocks on the preview network
• Infrastructure and observability already live — the status page, explorer, and backup systems are operational
• Interface and runtime still evolving — expect changes between versions in CLI flags, configuration, and internal APIs

This is pre-release software. It is suitable for research, experimentation, and contributing to development. It is not yet suitable for production stake pool operation.

Installation Philosophy

TSUNAGI prioritizes four principles in its build and deployment model:

• Transparency — documented architecture at every layer, public release planned after stability milestone
• Reproducibility — deterministic builds using the Zig toolchain, producing identical binaries from identical source
• Operator visibility — every subsystem exposes its internal state through structured endpoints and logs
• Deterministic behavior — identical inputs always produce identical outputs, with no hidden state or undefined transitions

Installation Steps

The following commands illustrate a typical workflow for cloning, building, and running TSUNAGI. These are example commands and may change as the project evolves.

Exact flags, paths, and runtime options may change between versions. The state endpoint returns JSON with chain tip, convergence, producer readiness, and all subsystem metrics.

Operational Verification

Once TSUNAGI is running, use the following methods to confirm node health and correct operation:

• Status endpoint — the status page provides real-time node telemetry including sync progress, slot height, and subsystem health
• Explorer — the explorer page exposes block decode telemetry, showing decoded transaction counts and block metadata
• Runtime logs — structured logs are written to the TSUNAGI_HOME directory, capturing every state transition and subsystem event
• Backup workflow — the scripts/backup-tsunagi-site.sh script handles incremental backup of node state and site data

Support and Contribution

TSUNAGI is an open project. Contributions, questions, and review are welcome.

• Source — currently private during active development; public release planned after stability milestone
• Treasury — project sustainability and funding information at the treasury page
• Whitepaper — complete architecture documentation at the whitepaper page