LLM & Agents

Centralizes persona-driven code reviews (Carmack, React core, Anders, React Router, etc.) so Claude can pick or combine expert viewpoints when the user asks for a code review or perspective-specific critique.

Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors.

Guide for building stateful, conversational agents using LangGraph. This skill should be used when creating new LangGraph agents, implementing StateGraph workflows, adding checkpointing, or integrating agents with the Agent Chat UI.

Use when diagnosing openai_harmony.HarmonyError or gpt-oss tool calling issues with vLLM. Identifies error sources (vLLM server vs client), maps specific error messages to known GitHub issues, and provides configuration fixes for tool calling problems with gpt-oss models.

Comprehensive Python programming guidelines based on Google's Python Style Guide. Use when Claude needs to write Python code, review Python code for style issues, refactor Python code, or provide Python programming guidance. Covers language rules (imports, exceptions, type annotations), style rules (naming conventions, formatting, docstrings), and best practices for clean, maintainable Python code.

Orient to current project by reading CLAUDE.md and architecture docs. Auto-invokes when Claude detects a new or unfamiliar repository. Can also be triggered by "orient me", "what is this project", or "prime session".

Safe filesystem operations for agents, including path normalization vs resolution, temp file handling, atomic replacement, and spooled buffers. Use when reading/writing user-supplied paths, staging outputs, or managing temporary files; triggers: filesystem, os.path, tempfile, path normalization, realpath, atomic write.

Best practices for activating and utilizing the Serena MCP toolset. Use this skill when working in a Serena-enabled project to ensure correct project activation (including WSL handling) and to leverage semantic tools for code navigation and editing.

Integrates MongoDB document database with Node.js applications using the official driver or Mongoose ODM. Use when building applications with flexible schemas, document storage, or when user mentions MongoDB, NoSQL, or document databases.

Turn Innovation PRDs into concrete experiment plans with explicit hypotheses, metrics, and evaluation methods for RAG quality, agents, and automation outcomes.

An easy way to force agent to think again if it's statement that the "Job is done and production ready" is actually done - usually it's not. This skill ensures that that the AI agent verifies their task and user does not have to check after the agent if they did their job.

Generate adversarial inputs, edge cases, and boundary test payloads. Use when testing input robustness, finding edge case failures, or stress-testing LLM boundaries.

Claude Rulesファイルを.claude/rules/ディレクトリに作成・管理する。プロジェクト固有のルール、コーディング規約、ワークフロー設定を追加したい場合に使用。

プロジェクトの .claude/skills/ に新しいスキルを作成する。「スキル作成」「新しいスキル」「スキルを作って」「スキル追加」「skill 作成」「スキルを追加したい」「新規スキル」などで起動。プロジェクト固有のスキルファイルを生成。

Use when working with trait extraction, place enrichment, or the learning loop. Load for understanding how the system learns from gameplay, how traits are extracted from Nominatim + player sessions, async queue processing, and LLM prompt configuration. Covers update_place_traits, pgmq, and edge function flow.

技術ブログの記事執筆を支援。トピックや題材を受け取ったら、Zenn形式の技術記事の骨子を作成します。実体験ベースで思考過程や試行錯誤を重視した記事構成を提案します。技術検証が必要な場合は、technical-verification-plannerスキルを使用してください。詳細な情報や最新の情報が必要な場合は、deep-research-agentスキルを使用してください。記事の骨子が完成したら、draft-preparation-assistantスキルを使用して下書きファイルを準備することを推奨します。記事の下書きが完成したら、article-review-assistantスキルを使用してレビューすることを推奨します。ブログ記事を書く時、記事の構成を考える時に使用してください。

Used to create a new agent. Used when a user wants to create a new agent

Agent spécialisé dans la génération de documents Markdown professionnels pour dossiers bancaires.

Control interactive terminal processes via tmux - launch CLI applications, send input, capture output, wait for completion. This skill should be used when tasks require interactive debugging (pdb, node inspect), REPL exploration, long-running observable processes, or spawning another Claude instance for independent analysis or second opinions.

Guide for creating effective skills that extend agent capabilities with specialized knowledge, workflows, or tool integrations. Use this skill when the user asks to; (1) create a new skill, (2) make a skill, (3) build a skill, (4) set up a skill, (5) initialize a skill, (6) scaffold a skill, (7) update or modify an existing skill, (8) validate a skill, (9) learn about skill structure, (10) understand how skills work, or (11) get guidance on skill design patterns. Trigger on phrases like "create a skill", "new skill", "make a skill", "skill for X", "how do I create a skill", or "help me build a skill".