# AI LLM Application

> Group skill: AI/LLM application — provider selection, app patterns, RAG, agents, prompts, evaluation, safety, and monitoring.

- Skill: `may215/ai-llm-application` (Agent Skill)
- Install (CLI): `npx skillmds@latest add may215/ai-llm-application`
- Raw SKILL.md: https://api.skillmd.com/api/skills/may215/ai-llm-application/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: may215 (https://skillmd.com/u/may215)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/may215/ai-llm-application

---


# 🤖 AI / LLM Application

> **Group Skill** — Orchestrates sub-skills for building an application powered by LLMs.
> Activate by saying: "Build an AI app" or "LLM application"

## When to Use

- Building an AI-powered application
- Need to integrate with LLM providers
- Implementing RAG, agents, or prompt engineering

## Step-by-Step Workflow

### Step 1: Provider
**Skill**: `gemini-api-dev` / `azure-ai-projects-py` / `azure-ai-projects-ts`
- Choose LLM provider
- Set up API access
- Configure authentication

### Step 2: Patterns
**Skill**: `llm-app-patterns` / `llm-structured-output`
- Define app architecture
- Design prompt flows
- Handle structured outputs

### Step 3: RAG
**Skill**: `rag-engineer` / `rag-implementation`
- Implement retrieval-augmented generation
- Set up vector databases
- Build embedding pipelines

### Step 4: Agents
**Skill**: `langchain-architecture` / `langgraph` / `crewai`
- Design agent orchestration
- Implement tool use
- Handle multi-agent coordination

### Step 5: Prompts
**Skill**: `prompt-engineering` / `prompt-engineering-patterns`
- Optimize prompts
- Use structured prompt patterns
- Manage prompt templates

### Step 6: Evaluation
**Skill**: `llm-evaluation` / `advanced-evaluation`
- Set up quality assurance
- Evaluate LLM outputs
- Track performance metrics

### Step 7: Safety
**Skill**: `yes-md` / `tool-use-guardian`
- Implement AI governance
- Add content safety filters
- Monitor for misuse

### Step 8: Monitoring
**Skill**: `langfuse` / `posthog-automation`
- Track LLM calls and costs
- Monitor latency and errors
- Analyze user interactions

## Completion

AI application is built with RAG, agents, safety, and monitoring. 🤖

## Sub-Skills Referenced

1. `gemini-api-dev`
2. `azure-ai-projects-py`
3. `azure-ai-projects-ts`
4. `llm-app-patterns`
5. `llm-structured-output`
6. `rag-engineer`
7. `rag-implementation`
8. `langchain-architecture`
9. `langgraph`
10. `crewai`
11. `prompt-engineering`
12. `prompt-engineering-patterns`
13. `llm-evaluation`
14. `advanced-evaluation`
15. `yes-md`
16. `tool-use-guardian`
17. `langfuse`
18. `posthog-automation`