igmarin/hanakai-yaku

Curated library of atomic AI agent skills for Hanami, dry-rb, and ROM Ruby development. Covers actions, slices, repositories, relations, changesets, providers, DI, operations, TDD, CLI, views, routing, and validation. Shared Ruby process skills have moved to ruby-core-skills. Uses Markdown + Front-matter architecture.

1.33x

Quality

94%

Does it follow best practices?

Impact

92%

1.33x

Average score across 35 eval scenarios

Securityby

Passed

No known issues

hanakai-yaku

Name: igmarin/hanakai-yaku
Rating: 92.2 (1 reviews)
Author: igmarin

hanakai-yaku logo

hanakai-yaku turns AI coding assistants into disciplined Hanami collaborators.

A curated library of atomic skills for Hanami, dry-rb, and ROM Ruby development. 3 initial skills and 1 agent (expanding to 18 skills + 3 agents) that teach AI tools how to configure providers, implement dependency injection, write TDD tests, create repositories, design slices, and build operations — using Hanami conventions.

The project is built around one non-negotiable rule:

Write test -> Run test -> Verify it FAILS for the right reason -> Implement -> Verify it PASSES

That TDD gate is encoded directly into the skills and agents.

Part of the AI Skill Ecosystem

This repo is one of 6 in a composable AI skill ecosystem:

Repo	Role
`ruby-core-skills`	15 shared Ruby skills + process discipline
`rails-agent-skills`	28 Rails-specific skills + 9 agents
`hanakai-yaku`	35 Hanami/dry-rb skills + 10 agents
`agnostic-planning-skills`	10 planning skills + 4 agents
`agent-mcp-runtime`	Rust CLI runtime (pack resolution, MCP)
`ruby-skill-bench`	Benchmark/eval engine

See the Ecosystem Overview for the full architecture.

This repo depends on ruby-core-skills. See Migration Guide.

Supported agent environments

Who This Is For

What Is In The Repository

Area	Purpose
`skills/`	Atomic Hanami skills organized by category: context, providers, actions, persistence, dry-rb, testing, slices, views.
`agents/`	Orchestrated agents that chain skills into guided workflows.
`docs/`	Architecture, agent guide, reference catalog, and calling conventions.
`CONTEXT.md`	Domain glossary — the canonical vocabulary for Hanami concepts.

Start Here

Skills are invoked via chat commands:

Method	Syntax	Example
Chat Command	`@skill-name`	`@load-context`

@load-context            # Discover the app's slices, providers, and routes
@configure-providers     # Set up providers and settings
@implement-di            # Configure dependency injection patterns
@hanami-setup            # Full project onboarding workflow

Skill Catalog

Skill	Category	Description
`load-context`	Context	Load the Hanami app structure before coding
`configure-providers`	Providers	Set up Hanami providers, settings, and .env
`implement-di`	Providers	Dependency injection patterns with dry-system

Agent

Agent	Description
`hanami-setup`	Project onboarding: Context → Providers → DI → Verify

See docs/reference/skill-catalog.md for the complete catalog.

How Skills Work

Each skill is a single SKILL.md file with YAML frontmatter and a 6-section body:

1. Frontmatter (YAML)       — name, description, metadata
2. Quick Reference          — scannable table for fast lookup
3. HARD-GATE               — non-negotiable blocking rules
4. Core Process             — step-by-step procedure
5. Output Style             — exact shape of artifacts
6. Integration              — predecessor/successor skills

Documentation Map

Need	Document
Understand the docs system	docs/index.md
Browse all skills	docs/reference/skill-catalog.md
Understand skill chaining	docs/reference/integration-matrix.md
Follow agent guides	docs/agent-guide.md
Understand repository structure	docs/architecture.md

Contributing

When contributing skills, agents, or docs:

Keep generated artifacts in English unless the user explicitly asks for another language.
Preserve the tests-gate-implementation rule for every code-producing skill.
Keep public docs consistent with tile.json, agents.json, and CONTEXT.md.

docs

evals

skills

README.md

tile.json