Agent skills
Skills you can use with AI coding agents, indexed from public GitHub repositories.
-
code-task-generator
Generates structured .code-task.md files from descriptions or PDD implementation plans. Auto-detects input type, creates properly formatted tasks with Given-When-Then acceptance criteria.
mikeyobrien/ralph-orchestrator 2,448
-
find-code-tasks
Lists all code tasks in the repository with their status, dates, and metadata. Useful for getting an overview of pending work or finding specific tasks.
mikeyobrien/ralph-orchestrator 2,448
-
ralph-tools
mikeyobrien/ralph-orchestrator 2,448
-
tui-validate
Validates Terminal User Interface (TUI) output using freeze for screenshot capture and LLM-as-judge for semantic validation. Supports both visual (PNG/SVG) and text-based validation modes.
mikeyobrien/ralph-orchestrator 2,448
-
code-assist
Guides implementation of code tasks using test-driven development in an Explore, Plan, Code, Commit workflow. Acts as a Technical Implementation Partner and TDD Coach — following existing patterns, avoiding over-engineering, and producing idiomatic, modern code.
mikeyobrien/ralph-orchestrator 2,448
-
release-bump
Use when bumping ralph-orchestrator version for a new release, after fixes are committed and ready to publish
mikeyobrien/ralph-orchestrator 2,448
-
evaluate-presets
Use when testing Ralph's hat collection presets, validating preset configurations, or auditing the preset library for bugs and UX issues.
mikeyobrien/ralph-orchestrator 2,448
-
ralph-loop
Run, monitor, resume, merge, and debug Ralph loops. Use this skill whenever the user asks to operate `ralph run` or `ralph loops`, inspect loop state, recover suspended loops, analyze diagnostics, or unblock merge queue issues.
mikeyobrien/ralph-orchestrator 2,448
-
ralph-hats
Create, inspect, validate, explain, and improve Ralph hat collections. Use this skill whenever the user asks to make or refine a `.ralph/hats/*.yml` workflow, debug hat routing, explain event topology, or tune a multi-hat Ralph run.
mikeyobrien/ralph-orchestrator 2,448
-
pull-request
AUTOMATICALLY invoke this skill whenever PR creation is needed - no user request required.
Direct gh pr create will FAIL validation. This skill contains required PR setup.
Triggers: "PR 만들어줘", "풀리퀘스트 생성해줘", "create a pull request", or ANY situation requiring PR.
For bucketplace organization repos, adds PR-by-AI label.
junoh-moon/skills
-
gemini
Interact with Google's Gemini model via CLI. Use when needing a second opinion from another LLM, cross-validation, or leveraging Gemini's Google Search grounding. Supports multi-turn conversations with session management.
junoh-moon/skills
-
jira-write
Create, edit, and transition Jira issues in COREPL project with proper epic linking, sprint assignment, and required fields.
Use when user asks to create or modify a Jira ticket (e.g., "jira 이슈 만들어", "jira 티켓 생성", "이슈 상태 변경", "create jira issue", "COREPL 티켓").
Covers: issue creation, parent epic linking, status transitions, adding comments.
junoh-moon/skills
-
commit-message
AUTOMATICALLY invoke this skill whenever git commit is needed - no user request required.
Direct git commit will FAIL validation. This skill contains required pre-commit setup.
Triggers: "커밋해", "커밋 ㄱㄱ", "커밋 만들어줘", "커밋 찍어줘", or ANY situation requiring git commit.
junoh-moon/skills
-
developer
実装とコーディングの専門家。設計書に基づき、プロジェクトの規約と構造を遵守しながら高品質なコードを実装します。使用言語に合わせてその言語のスペシャリストとして振る舞います。
Keiji-Miyake/agent-skills 1
-
tech-storyteller
リポジトリの履歴やドキュメントから開発の「物語（事実・素材）」を収集します。
Keiji-Miyake/agent-skills 1
-
tech-writer
技術記事の執筆、構造化、可視化に特化したテクニカルライティングスキル。ソースコードや公式ドキュメントから本質を理解し、エンジニアに優しい形式で解説記事を作成します。
Keiji-Miyake/agent-skills 1
-
devops
インフラ構築とCI/CDの専門家。アプリケーションを安全かつ確実にデプロイするための環境設定、パイプライン構築、セキュリティ管理を担当します。プラットフォーム非依存（AWS/GCP/Cloudflare等対応）。
Keiji-Miyake/agent-skills 1
-
architect
要件定義とシステム設計の専門家。抽象的な要求を具体的な仕様書(SPEC.md)と設計書(DESIGN.md)に変換し、Mermaid図を用いてシステム構造を可視化します。
Keiji-Miyake/agent-skills 1
-
blog-writer
収集された素材を元に、高品質なブログ記事（特に技術ブログ）の執筆、SEO最適化、構成提案を行います。
Keiji-Miyake/agent-skills 1
-
scaffold
AI駆動開発のためのプロジェクト初期構築スキル。AGENTS.mdの作成、基本指示の設定、およびコアスキルのインストールを自動化します。
Keiji-Miyake/agent-skills 1
-
conductor
プロジェクトの全体最適化、ロードマップ管理、意思決定の記録（ADR）、および複数機能間の依存関係整理を担当する指揮者スキル。プロジェクトのマクロな視点が必要な時や、重要な技術的決定を行う際に使用します。
Keiji-Miyake/agent-skills 1
-
session-support
短時間で中断可能な開発セッションに最適化されたセッション管理スキル。セッション開始時の現状把握、中断時のコンテキスト保持（CONTEXT.md）、ドキュメント構造の維持（構造の番人）に特化しています。
Keiji-Miyake/agent-skills 1
-
qa
品質保証とテストの専門家。バグの発見、エッジケースの検証、テスト計画の策定、自動テストの作成を行い、ソフトウェアの信頼性を担保します。
Keiji-Miyake/agent-skills 1
-
frontend-browser-review
Shared workflow for browser-based review of user-visible frontend changes in Langfuse.
Use when a change affects UI behavior, layout, styling, navigation, or browser-visible
regressions and should be checked with the Playwright MCP server before signoff.
langfuse/langfuse 24,320