Stacks Documentation
SkillsChangelog
TeamSponsorsPartnersPostcardwareAwesome StacksContributing

AI

Copy page
Copy page as Markdown for LLMs
View as Markdown
View this page as plain text
Open in ChatGPT
Copy content and open ChatGPT
Open in Claude
Copy content and open Claude
Open in Perplexity
Search with Perplexity AI
On this page 5

stacks-ai · Backend and API · model-invoked

The AI layer: Anthropic, OpenAI, Ollama and AWS Bedrock drivers, image generation, vision, RAG and embeddings, MCP clients, and the higher-level helpers for summarization, sentiment and classification.

When to reach for it

  • Using Anthropic/OpenAI/Ollama/AWS Bedrock drivers
  • Image generation (DALL-E)
  • Vision analysis
  • RAG/vector search
  • Embeddings
  • MCP (Model Context Protocol) clients
  • Text summarization
  • Sentiment analysis
  • Content classification
  • Personalization
  • The buddy AI assistant

Covers

@stacksjs/ai, config/ai.ts.

Inside the skill

The sections an agent reads once the skill loads.

  • Key Paths
  • Source Files
  • Anthropic Driver
  • OpenAI Driver
  • Provider-Neutral Client
  • Ollama Driver (Local LLMs)
  • Image Generation
  • RAG & Vector Search
  • MCP (Model Context Protocol)
  • Personalization
  • Buddy AI Assistant
  • Compact Project Context
  • Claude Agent
  • Claude Agent SDK
  • config/ai.ts
  • Gotchas

Where the code lives

  • Core package: storage/framework/core/ai/src/
  • Configuration: config/ai.ts

Using it

This one is model-invoked. Your agent reaches for it on its own when the task matches, and you can also call it by name:

/stacks-ai

Source: stacks-ai/SKILL.md. Shadow it for one project with app/Skills/stacks-ai/SKILL.md, then re-run buddy setup:ai. See Writing your own.