🎯

docs-write

🎯Skill

from lukasstrickler/ai-dev-atelier

VibeIndex|
What it does

Writes and updates documentation following predefined standards and best practices, ensuring consistent, high-quality technical documentation across a project.

πŸ“¦

Part of

lukasstrickler/ai-dev-atelier(10 items)

docs-write

Installation

Install ScriptRun install script
curl -sSfL https://raw.githubusercontent.com/LukasStrickler/ai-dev-atelier/main/install.sh | bash
git cloneClone repository
git clone https://github.com/LukasStrickler/ai-dev-atelier.git ~/ai-dev-atelier
MakeRun with Make
make -C ~/ai-dev-atelier install
MakeRun with Make
make -C ~/ai-dev-atelier validate
Install ScriptRun install script
bash ~/ai-dev-atelier/install.sh --no
πŸ“– Extracted from docs: lukasstrickler/ai-dev-atelier
1Installs
-
AddedFeb 4, 2026

Skill Details

SKILL.md

Overview

AI Dev Atelier - Mastering the New Stack

# AI Dev Atelier

> Mastering the New Stack. A production-ready pack of 10 skills for code quality, documentation, code review, and research, built on the open [Agent Skills standard](https://agentskills.io/specification) and designed for OpenCode.

![Version](https://img.shields.io/badge/version-1.0.0-blue) ![License](https://img.shields.io/badge/license-MIT-blue) ![Agent Skills](https://img.shields.io/badge/Agent%20Skills-open-2ea043) ![OpenCode](https://img.shields.io/badge/OpenCode-supported-1f6feb)

Table of Contents

  • [Overview](#overview)
  • [Quick Start](#quick-start)
  • [What You Get](#what-you-get)
  • [Docs Index](#docs-index)
  • [Prerequisites](#prerequisites)
  • [Contributing](#contributing)
  • [License](#license)
  • [Community](#community)

Overview

AI Dev Atelier is a curated skill pack for OpenCode and Cursor. Agent Skills are a simple, open format for giving agents portable capabilities and expertise through reusable instructions, scripts, and resources. This repo provides those skills, installs them to ~/.opencode/skills/ (OpenCode) and ~/.cursor/skills/ (Cursor, [per Cursor docs](https://cursor.com/docs/context/skills)), and configures MCP servers for both: OpenCode at ~/.opencode/opencode.json and Cursor at ~/.cursor/mcp.json ([Cursor MCP docs](https://cursor.com/docs/context/mcp)).

What You Get

  • 10 production-ready [Agent Skills](https://agentskills.io/specification) from config/skills.json
  • 9 MCP servers configured from config/mcps.json
  • PreToolUse guardrails defined in config/hooks.json
  • code-quality β€” Typecheck, lint, format, Markdown validation
  • docs-check β€” Detect docs needing updates from git diff
  • docs-write β€” Write/update docs with standards
  • git-commit β€” Write clear commits with Conventional Commits format
  • code-review β€” CodeRabbit reviews (task/pr modes)
  • resolve-pr-comments β€” Multi-agent PR comment resolution
  • search β€” Web + library docs + GitHub code search
  • research β€” Academic research with evidence cards
  • tdd β€” Test-Driven Development with Testing Trophy workflow
  • ui-animation β€” Tasteful UI animation & accessibility
  • use-graphite β€” Manage stacked PRs with Graphite CLI

See content/skills/README.md for full usage details and scripts.

Quick Start

```bash

# 1. Install AI Dev Atelier

curl -sSfL https://raw.githubusercontent.com/LukasStrickler/ai-dev-atelier/main/install.sh | bash

# 2. Verify installation

# Ask OpenCode: "What skills are available?"

# Prompt before overwriting skills

# bash ~/ai-dev-atelier/install.sh --no

```

Curl install notes:

  • Cache location: ~/.config/opencode/atelier/cache/repo (or ~/.opencode/atelier/cache/repo if OpenCode uses ~/.opencode).
  • Add API keys in .env at that path (copy from .env.example once).
  • Update by re-running the same curl command; it refreshes tracked files while keeping your .env and updating .env.example.

Install a specific ref:

```bash

AI_DEV_ATELIER_REF= curl -sSfL \

https://raw.githubusercontent.com/LukasStrickler/ai-dev-atelier//install.sh | bash

```

If you prefer a local checkout (recommended for dev work):

```bash

git clone https://github.com/LukasStrickler/ai-dev-atelier.git ~/ai-dev-atelier

# Install local changes into OpenCode

make -C ~/ai-dev-atelier install

# Validate skill structure

make -C ~/ai-dev-atelier validate

# Prompt before overwriting skills

bash ~/ai-dev-atelier/install.sh --no

```

Docs Index

| Document | Purpose |

|----------|---------|

| [INSTALL.md](./INSTALL.md) | Dependencies and MCP setup |

| [content/skills/README.md](./content/skills/README.md) | Skill catalog and usage reference |

| [AGENTS.md](./AGENTS.md) | Developer workflow guide and MCP reference |

| [MY_AGENTIC_DEV_SETUP.md](./MY_AGENTIC_DEV_SETUP.md) | OpenCode + Fulcrum + oh-my-opencode workflow |

Prerequisites

  • Git
  • Bash 4.0+
  • Node.js 18+ or Bun (for code-quality tooling)
  • OpenCode (skills runtime)
  • jq (optional, required for MCP config automation)

install.sh checks dependencies and prompts to install missing packages when possible.

See [INSTALL.md](./INSTALL.md) for platform-specific installation commands and optional tools.

Contributing

  1. Run make test before committing.
  2. Keep instructions and scripts inside content/skills/.
  3. Update content/skills/README.md when the catalog changes.
  4. Keep outputs under .ada/.

License

MIT License. See [LICENSE](./LICENSE).

Community

  • Issues: Report bugs or request features via [GitHub Issues](https://github.com/LukasStrickler/ai-dev-atelier/issues)

More from this repository9

🎯
ui-animation🎯Skill

Generates tasteful, accessible UI animations that enhance user experience while maintaining design consistency and interaction standards.

🎯
use-graphite🎯Skill

use-graphite skill from lukasstrickler/ai-dev-atelier

🎯
code-quality🎯Skill

Automatically checks code quality by performing typechecking, linting, formatting, and Markdown validation across project files.

🎯
code-review🎯Skill

Performs comprehensive code reviews using CodeRabbit, supporting both task-level and pull request review modes to provide detailed feedback and analysis.

🎯
tdd🎯Skill

Guides developers through Test-Driven Development (TDD) by generating comprehensive test suites, scaffolding tests before implementation, and ensuring code coverage using the Testing Trophy workflow.

🎯
research🎯Skill

Conducts academic research by generating evidence-based research cards with scholarly sources and citations.

🎯
search🎯Skill

Searches across web resources, library documentation, and GitHub repositories to help developers quickly find relevant code, references, and technical information.

🎯
docs-check🎯Skill

Detects documentation files that need updates by analyzing git diffs and identifying outdated or changed documentation relative to recent code changes.

🎯
resolve-pr-comments🎯Skill

Resolves pull request comments by coordinating multiple AI agents to systematically address and implement feedback across code changes.