Atelier

Atelier - A collaborative workshop for software development

An atelier is the private workshop or studio where a principal master and a number of assistants, students, and apprentices can work together producing fine art or visual art released under the master's name or supervision.

>

Wikipedia

A personal development toolkit for AI agents - spec-driven development, code quality, deep thinking, and ecosystem patterns.

Quick Start

Install Atelier with a single command. It auto-detects your AI harness (Claude Code or OpenCode), configures agents, and sets up session hooks:

# Initialize atelier in your project
bunx @martinffx/atelier@latest init

# Or install everything including skills
bunx @martinffx/atelier@latest init --all

# Non-interactive mode (CI/CD)
bunx @martinffx/atelier@latest init --yes

That's it. Your project is now configured for spec-driven development.

What Gets Installed

Atelier sets up three things in your project:

1. Skills (34 available)

Specialized knowledge modules that auto-invoke based on context. Install them via the CLI (--all flag) or separately:

npx skills add martinffx/atelier

2. Agent Personas (3 subagents)

Harness-agnostic agent definitions configured with appropriate models:

Agents are generated into .claude/agents/ or .opencode/agents/ with harness-specific model identifiers.

3. Session Hooks

Context injection that runs on every session start, clear, or compact:

• Claude Code: hooks/atelier-session-start outputs project context as JSON
• OpenCode: .opencode/plugins/atelier.js provides config and transform hooks

4. Task Tracking (optional)

The spec workflow skills support beads for dependency-aware task tracking:

# Install beads (optional but recommended)
npm install -g beads

Why beads? It provides bd ready (finds next unblocked task), bd dep add (dependency management), and bd list (progress tracking) that harness-native todos can't match.

Fallback: If beads isn't installed, skills fall back to the harness's native todo system (TodoWrite for Claude Code, built-in todos for OpenCode).

5. Configuration

Single source of truth in .atelier/config.json:

{
  "version": "1.0.0",
  "harness": "claude",
  "skills_source": "martinffx/atelier",
  "agents": [
    { "template": "scout", "model": "haiku" },
    { "template": "oracle", "model": "opus" },
    { "template": "architect", "model": "opus" }
  ]
}

CLI Commands

init (default)

Initialize atelier in the current project.

bunx @martinffx/atelier@latest init [options]

Options:

• --harness <type> - Force harness type (claude or opencode)
• --all - Also install skills via npx skills add
• --yes - Non-interactive mode with default models
• --project - Install skills in project directory (default: global)

Idempotent: Re-running init is safe. It regenerates files without deleting anything unless you switch harnesses.

update

Refresh hooks and agents from the latest templates without touching skills:

bunx @martinffx/atelier@latest update

remove

Remove all atelier-generated files from the project:

bunx @martinffx/atelier@latest remove

Skills remain installed. Run npx skills remove martinffx/atelier to remove them separately.

Skills

This repository includes 34 skills that enhance AI agents with specialized knowledge and workflows.

Installing Skills

If you didn't use --all during init, install skills manually:

# Install all skills
npx skills add martinffx/atelier

# Install specific skills
npx skills add martinffx/atelier --skill typescript-drizzle-orm
npx skills add martinffx/atelier --skill python-fastapi
npx skills add martinffx/atelier --skill spec-brainstorm

Skill Types

Skills fall into four categories based on what they do:

#### Workflow Skills (spec:*)

Process-oriented skills that guide you through structured development workflows. These produce artifacts and should be followed step-by-step.

• spec-brainstorm → spec.md — Discovery, requirements, architecture
• spec-plan → plan.json — Break spec into implementable tasks
• spec-implement — Execute tasks with TDD
• spec-finish — Validate, review, prepare for PR
• spec-orchestrator — Route to the right skill based on context

#### Thinking Skills (oracle:*)

Analytical skills that provide patterns, principles, and deep reasoning. These adapt to your specific situation.

• oracle-architect — DDD, hexagonal architecture, component design
• oracle-challenge — Challenge assumptions, validate decisions
• oracle-doubt — Adversarial review of non-trivial decisions
• oracle-grillme — Socratic interrogation of plans against domain model
• oracle-security — Security architecture and threat modeling
• oracle-testing — Stub-driven TDD, layer boundary testing
• oracle-thinkdeep — Extended sequential reasoning for complex problems

#### Domain Knowledge (python:*, typescript:*)

Technology-specific patterns and best practices. These are like having a senior engineer for that stack.

TypeScript (8 skills)

• typescript-api-design — REST conventions, error responses, pagination
• typescript-fastify — Fastify + TypeBox route handlers
• typescript-drizzle-orm — Type-safe SQL schemas and queries
• typescript-dynamodb-toolbox — Single-table design, GSIs
• typescript-functional-patterns — ADTs, branded types, Option/Result
• typescript-effect-ts — Functional effects, error handling, resources
• typescript-build-tools — Bun, Vitest, Biome, Turborepo
• typescript-testing — Mocking, MSW, snapshot testing

Python (8 skills)

• python-architecture — Functional core/shell, DDD, layered architecture
• python-fastapi — Pydantic validation, dependency injection, OpenAPI
• python-sqlalchemy — ORM patterns, queries, async, upserts
• python-temporal — Workflow orchestration, activities, error handling
• python-modern-python — Type hints, generics, pattern matching
• python-monorepo — uv workspaces, mise task orchestration
• python-testing — Stub-driven TDD, pytest patterns
• python-build-tools — uv, ruff, basedpyright, pytest config

#### Utility Skills (code:*)

Task-specific tools you invoke when you need them.

• code-commit — Generate and validate conventional commits
• code-debug — Systematic debugging workflow
• code-docs — README, API docs, changelog generation
• code-handoff — Compact conversation into handoff document
• code-review — Multi-agent code review with specialized reviewers
• code-subagents — Dispatch patterns for parallel implementation

Skills are auto-invoked based on their description when you work with relevant technologies. No commands needed—just install and AI agents will use them when appropriate.

How Skills Work

Skills are auto-invoked based on context. When you say "create a spec for user auth", the AI matches this to spec-brainstorm and loads it automatically.

Namespace Philosophy

Skills are organized into four categories based on their role:

• Workflow (spec:) — Sequential steps that produce artifacts. Follow them in order.
• Thinking (oracle:) — Analytical capabilities that reason about your specific problem.
• Domain Knowledge (python-*, typescript-*) — Stack-specific patterns and best practices. Like having a senior engineer for that technology.
• Utility (code:) — Task-specific tools you invoke directly when needed.

The Spec Workflow

graph LR
    A[spec-brainstorm] -->|spec.md| B[spec-plan]
    B -->|plan.json| C[spec-implement]
    C --> D[spec-finish]
    
    B -.->|design flaw| A
    C -.->|missing tasks| B
    C -.->|fundamental issue| A
    D -.->|bugs found| C

Standard flow:

1. Research - Discovery + research + architecture → spec.md
2. Plan - Break into tasks → plan.json
3. Implement - Execute with TDD
4. Finish - Validate and review

Iteration is normal - Backflows (dotted lines) are expected when:

• Planning reveals design flaws → back to research
• Implementation finds missing tasks → update plan
• Validation finds bugs → back to implement

When to Use Which Skill

Development

For local development with Claude Code, use the --plugin-dir flag to load skills directly:

claude --plugin-dir ./atelier

Restart Claude Code after making changes to reload skills.

To work on the CLI itself:

# Build the CLI
bun run build

# Test locally
bun ./dist/atelier.js init --yes

License

MIT Copyright (c) 2026 Martin Richards