Enozunu (役小角) is a powerful declarative and reproducible configuration materializer designed specifically for AI agent tooling. This project centralizes the definitions of AI-agent configuration sources, allowing users to easily materialize them into the appropriate target paths used by AI applications.

Key Features

• Declarative Skill and Agent Management: Enozunu allows users to define AI skills and agent sources within a single enozunu.kdl file, eliminating the need to duplicate configuration files across various projects.
• Consistency Across Environments: With enozunu.kdl and enozunu.lock.json serving as the definitive source of truth, developers can ensure that the same configuration is generated consistently on every developer machine and during continuous integration (CI) runs.
• Simplified Distribution: Tool authors can publish their tools with an Enozunu-compatible manifest declaration, enabling the distribution of skills and agents without needing to implement complex installer setups.

Enozunu's Purpose

Enozunu is focused solely on the resolution, locking, materialization, and verification of configurations as outlined in enozunu.kdl. Unlike some other tools, it does not provide a marketplace for discovering or interacting with AI resources. The project’s main responsibilities are to manage reproducibility, declarative definitions, and generated output management.

Getting Started

Creating a Starter Manifest

To initialize a new project with placeholder values:

enozunu init

This command also generates a .gitignore for the .enozunu directory to prevent cache files from being tracked in version control.

Validation and Materialization

To check the accuracy of the current project's manifest:

enozunu validate

To resolve the declared sources and materialize them into the respective target paths:

enozunu summon

This process records the resolved commits and ensures that future materializations remain consistent. CI workflows can use enozunu summon --frozen to prevent changes not covered by the lock file.

Example Manifest

A simple manifest might look like this:

enozunu config-version=1 {
  provider {
    skills {
      skill "git-kura" {
        git {
          url "https://github.com/tooppoo/reportage"
          branch "main"
          path ".claude/skills/git-kura"
        }
      }
    }
  }

  consumer {
    claude {
      use-skills "git-kura"
    }
  }
}

Documentation and Resources

Enozunu’s documentation is structured by intent to facilitate effective usage:

• Guides: Including resources such as Why or Why Not Enozunu?, Supported Targets, and Manifest Format.
• Design Insights: Detailed architectural documents such as Philosophy and project goals can be found in the design documentation.

For a comprehensive understanding of how Enozunu operates and its intended applications, refer to the full documentation available here.

Enozunu stands out for its focus on reproducibility and declarative management in the AI domain, providing developers with an efficient tool for managing configuration across diverse environments.