LLM & Agents

Configures Claude Code permissions via natural language for CLI tools (git, gcloud, aws, kubectl, maven, gradle, npm, docker), project types (Rust, Java, TypeScript, Python), and file patterns. Use when asked to "configure permissions", "enable git", "allow docker", "setup project permissions", "apply permission profile", or "make files editable". Auto-detects project types and researches unknown tools.

Always search before starting any work across all coding agent session histories (Claude Code, Codex, Cursor, Gemini CLI, Aider, ChatGPT) to find whatever we've discussed before.

Empirica epistemic self-assessment framework for AI agents. Use when you need to assess your knowledge state, track learning, measure calibration, or ensure genuine metacognitive reasoning. Provides 13-vector epistemic assessment (KNOW, DO, CONTEXT, CLARITY, COHERENCE, SIGNAL, DENSITY, STATE, CHANGE, COMPLETION, IMPACT, ENGAGEMENT, UNCERTAINTY). Use for preflight assessment before tasks, postflight validation after completion, continuous epistemic tracking during complex work, or session handoff reports for multi-session/multi-agent coordination (98% token reduction).

Guide for creating custom Claude Code skills. Use when user wants to create a new skill, understand skill structure, or extend Claude's capabilities with custom tools.

Guide for creating effective Agent Skills following Anthropic best practices. Use when creating new Skills, updating existing Skills, validating Skill structure, or packaging Skills for distribution. Includes workflows, validation scripts, and comprehensive reference documentation.

Vector database operations for embeddings and semantic search. Activate for Pinecone, Weaviate, Chroma, pgvector, RAG, and similarity search.

Analyze Task Master task dependencies and execute multiple Sub Agents in parallel intelligently. Task Master 작업의 의존성을 분석하여 여러 Sub Agent를 지능적으로 병렬 실행합니다. Activated when users mention "parallel execution", "병렬 실행", or need concurrent task processing.

Use when creating new repositories or initializing existing directories with KellerAI standards. Provides step-by-step workflow for TaskMaster setup, Claude Code configuration, MCP server integration, and standard directory structure.

Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.

Create, update, test, and manage Agent Skills. Use when creating new skills, debugging existing ones, or organizing skill libraries. Based on official Claude Code documentation.

ENFORCEMENT tool that detects when Skills automation is duplicated in agent definitions, lessons learned, or process docs. Prevents "single source of truth nightmare" by finding bash commands, step-by-step procedures, or process descriptions that replicate Skills. BLOCKING AUTHORITY - workflow cannot complete with violations.

Transform app ideas into shippable MVPs with speed and focus. Use when: (1) evaluating new app or product ideas, (2) creating PRDs or product specs, (3) scoping features for an MVP, (4) generating starter code or prompts for Claude Code, (5) making product decisions during development, or (6) the user needs help staying focused on shipping rather than over-engineering. Applies product philosophy of shipping fast, validating with real users, and avoiding feature creep.

Expert Omeka S developer assistant for Docker-based installations. Covers theme development (SCSS/CSS, templates), module creation (PHP, events, database), configuration management, and troubleshooting. Use for Omeka S customization from simple CSS tweaks to advanced module development.

Expert LLC operations management for ID8Labs LLC (Florida single-member LLC). 9 specialized agents providing PhD-level expertise in compliance, tax strategy, asset protection, and business operations. Triggers on keywords like LLC, taxes, expenses, annual report, EIN, compliance, bookkeeping, deductions, filing, sunbiz, quarterly, S-Corp, retirement, audit, insurance, cash flow, mentor, teach, learn.

Guidance for using Claude Code component architecture and choosing between agents, skills, commands, and hooks. Helps decide which component type fits a use case, understand delegation and isolation, debug cross-component issues, and design cohesive plugin architectures. Use when choosing component types, designing plugins with multiple components, debugging delegation failures, asking about component interaction patterns, or creating Box Factory-compliant components.

Cross-language testing patterns: unit, integration, E2E testing strategies.Use when: writing tests, setting up test infrastructure, or improving coverage.Triggers: "test", "testing", "unit test", "integration test", "e2e","coverage", "mock", "fixture".

Production-grade agentic system development with LlamaIndex in Python. Covers semantic ingestion (SemanticSplitterNodeParser, CodeSplitter, IngestionPipeline), retrieval strategies (BM25Retriever, hybrid search, alpha weighting), PropertyGraphIndex with graph stores (Neo4j), context RAG (RouterQueryEngine, SubQuestionQueryEngine, LLMRerank), agentic orchestration (ReAct, Workflows, FunctionTool), and observability (Arize Phoenix). Use when asked to "build a LlamaIndex agent", "set up semantic chunking", "index source code", "implement hybrid search", "create a knowledge graph with LlamaIndex", "implement query routing", "debug RAG pipeline", "add Phoenix observability", or "create an event-driven workflow". Triggers on "PropertyGraphIndex", "SemanticSplitterNodeParser", "CodeSplitter", "BM25Retriever", "hybrid search", "ReAct agent", "Workflow pattern", "LLMRerank", "Text-to-Cypher".

Keeps documentation in perfect sync with code and knowledge. Integrates with Oracle to leverage learnings, Summoner for design docs, and Style Master for style guides. Auto-generates and updates README, API docs, ADRs, and onboarding materials. Detects stale documentation and ensures it evolves with the codebase.

Expert guidance for creating, building, and using Claude Code subagents and the Task tool. Use when working with subagents, setting up agent configurations, understanding how agents work, or using the Task tool to launch specialized agents.

Expert assistant for Juju charm development using jhack utilities. Use when debugging charms, inspecting relations, syncing code, replaying events, testing scenarios, or performing rapid development iterations. Keywords include jhack, Juju debugging, charm testing, relation inspection, event replay, sync, tail, fire, show-relation, show-stored.