LLM & Agents

AI content generation with OpenAI and Claude, callAIWithPrompt usage, prompt storage in app_settings, structured outputs, response format validation, multi-criteria scoring, rate limiting, JSON schema, and AI API best practices. Use when generating content, creating prompts, scoring articles, or working with OpenAI/Claude APIs.

This skill provides comprehensive knowledge for working with the Anthropic Claude Agent SDK. It should be used when building autonomous AI agents, creating multi-step reasoning workflows, orchestrating specialized subagents, integrating custom tools and MCP servers, or implementing production-ready agentic systems with Claude Code's capabilities.Use when building coding agents, SRE systems, security auditors, incident responders, code review bots, or any autonomous system that requires programmatic interaction with Claude Code CLI, persistent sessions, tool orchestration, and fine-grained permission control.Keywords: claude agent sdk, @anthropic-ai/claude-agent-sdk, query(), createSdkMcpServer, AgentDefinition, tool(), claude subagents, mcp servers, autonomous agents, agentic loops, session management, permissionMode, canUseTool, multi-agent orchestration, settingSources, CLI not found, context length exceeded

Complete Amplify Gen 2 handbook - instructs Claude to read resources/handbook/ before implementing any code

Coordinate multi-agent software development from requirements through verified implementation using spec-driven workflows. Use when planning features, creating technical specifications, implementing complex tasks through specialized agents, or managing spec-dev workflows (PLAN/BUILD/ITERATE).

Generate and maintain the FastAPI backend structure for the RAG chatbot with proper project organization.

Test implementations against requirements, collect feedback, and create improvement loops

Comprehensive JWT authentication expert for senior developers (10+ years experience). Intelligently detects project language/framework and implements production-ready JWT auth systems with refresh tokens, secure HTTP-only cookies, token rotation, blacklisting, RBAC, MFA, and complete security. Covers Express, FastAPI, Next.js, React, Django, Flask, NestJS, and more. Automatically audits JWT implementations, generates complete auth systems (registration, login, logout, refresh, password reset), implements middleware, prevents XSS/CSRF attacks, uses bcrypt/argon2 hashing, and follows OWASP best practices. Use for implementing JWT authentication, token refresh flows, secure cookie storage, protected routes, role-based access control, security audits, and complete auth system generation.

Semantic compression of verbose documentation into OCTAVE format achieving 60-80% reduction with 100% decision-logic fidelity. Use when converting natural language to OCTAVE, streamlining documentation, creating semantic density, optimizing agent prompts, or compressing knowledge artifacts. Triggers on: octave compression, compress to octave, semantic compression, documentation streamlining, OCTAVE optimization, compress documentation, semantic density.

Guide for creating Claude Code hooks. This skill should be used when users want to create, modify, or troubleshoot hooks that automate actions at specific points in Claude Code's workflow, such as validating code before commits, running linters after edits, or protecting sensitive files.

Codex CLI（read-only）を用いて、レビュー→Claude Code修正→再レビュー（ok: true まで）を反復し収束させるレビューゲート。仕様書/SPEC/PRD/要件定義/設計、実装計画（PLANS.md等）の作成・更新直後、major step（>=5 files / 新規モジュール / 公開API / infra・config変更）完了後、および git commit / PR / merge / release 前に使用する。キーワード: Codexレビュー, codex review, レビューゲート.

Interactive Notion workspace management using MCP tools. Use for searching pages/databases, querying database entries, creating pages with custom properties, managing workspace structure, and exploratory operations requiring user interaction.

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

Generates jest.config.cjs for unit testing configuration. Configures Jest with Vue, TypeScript support, coverage thresholds, and test environment settings.

Generate intelligent, non-redundant test plans based on implementation changes. This skill should be used after implementing features to create comprehensive yet efficient test plans with proper coverage across unit, integration, API, and E2E tests without duplication.

Queries cached external documentation and blog content for authoritative, up-to-date information. Sources include Claude Code docs, engineering blog, and configured knowledge bases with 24-hour TTL caching. Use when you need current information about Claude Code features, hooks, or best practices. Do NOT use for general coding questions - rely on training knowledge or web search instead.

Create OpenCode plugins using @opencode-ai/plugin SDK. Use when user wants to BUILD a plugin in .opencode/plugin/ or ~/.config/opencode/plugin/ - for custom tools, event hooks, LLM parameter control, auth providers, or tool execution interception. DO NOT use when modifying OpenCode source code (packages/plugin, packages/opencode, etc.) or working on OC internals - those are core development, not plugin creation.

Collect a single Schwab MCP data snapshot (accounts, quotes, indicators, option chains) with a five-minute cache so trading agents can reuse consistent inputs. Use whenever an options persona needs fresh risk, market, or chain data before recommending trades.

Automatically assigns GitHub issues to appropriate GitHub Copilot agents based on issue type, labels, and content. Triggers on "assign agent", "who should work on this", "route this issue", or after creating issues via github-mcp-orchestrator.

Use when optimizing CLAUDE.md, AGENTS.md, custom commands, or skill files for Claude 4.5 models - applies documented Anthropic best practices systematically instead of inventing improvements

Multi-agent coordination, tier approvals, decision logging