skills.rest
limaon
limaonCOMMUNITY
Explore all skills by this author →
GitHub

llm-application-dev

Build AI applications with prompt templates, RAG patterns, and LLM API calls.

Updated Jul 8, 2022
Search Tags:#TypeScript#OpenAI#llm#rag#prompt-engineering#vector-store
One-click install
npx skills add https://github.com/limaon/dotfiles --skill llm-application-dev-limaon
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: llm-application-dev
Source: https://github.com/limaon/dotfiles/tree/main/.config/opencode/skills/llm-application-dev
Command: npx skills add https://github.com/limaon/dotfiles --skill llm-application-dev-limaon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Building AI-powered applications often requires combining prompt design, retrieval-augmented workflows, and robust LLM integration. This Skill guides developers to structure prompts, orchestrate multi-step LLM calls, and connect to vector stores or APIs for contextual knowledge.

Core Features & Use Cases

  • Prompt engineering templates for task-specific LLM prompts.
  • RAG patterns to fetch and incorporate relevant documents.
  • End-to-end guidance for designing AI assistants, chatbots, and automation tools.
  • Real-world example: create a product FAQ assistant that queries a vector store for user questions and responds with concise, sourced answers.

Quick Start

Use this skill to draft a basic LLM-powered assistant by defining a prompt template, wiring a retrieval step, and calling an LLM API to generate answers.

Frequently Asked Questions about llm-application-dev

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a chatbot with retrieval-augmented generation?▼

Building a chatbot with retrieval-augmented generation involves defining prompt templates, wiring a retrieval step to fetch relevant documents from a vector store, and calling an LLM API to generate contextual answers.

What's the best way to structure prompts for multi-step LLM calls?▼

Structuring prompts for multi-step LLM calls requires designing task-specific prompt templates that guide the LLM through sequential reasoning, ensuring each step passes context effectively to the next.

How does RAG connect to a vector store for AI assistant knowledge?▼

RAG connects to a vector store by querying stored document embeddings to fetch relevant information, which is then injected into prompt templates to provide the LLM with contextual external knowledge.

Can I use TypeScript for LLM application development with OpenAI?▼

TypeScript supports LLM application development with OpenAI by structuring multi-step API calls, defining typed prompt templates, and integrating vector store connections for RAG workflows.

Do I need a vector store to build an AI assistant?▼

A vector store is optional for building an AI assistant. You can build basic LLM-powered applications using prompt engineering alone, but integrating a vector store enables RAG patterns for external knowledge retrieval.

What are the limitations of prompt engineering for automation tasks?▼

Prompt engineering for automation tasks may lack real-time contextual knowledge without a vector store, requiring multi-step LLM calls to fetch external data and maintain accuracy across complex workflows.

Related Skills

Explore More Software Engineering Skills→
dotlab-hqdotlab-hq

langchain-rag

Build RAG pipelines that ingest, chunk, embed, and retrieve documents for LLMs.

Official
Intermediate
VrooliVrooli
54

Funnel Builder

Build multi-step conversion funnels with PostgreSQL storage and CLI/API endpoints.

Official
Advanced
abdullahhqaiserabdullahhqaiser

langchain-rag

Build RAG pipelines that load, split, embed, store, and retrieve documents.

Community
Intermediate
ejolietejoliet

senior-prompt-engineer

Design prompts, evaluate LLM outputs, and build agentic workflows.

Community
Advanced