AiMarkdown

Markdown

Sanitized markdown to HTML renderer.

Retrieval-Augmented Generation

RAG combines retrieval systems with language models to produce grounded responses.

The pipeline works in three stages:

  1. Indexing — documents are chunked and embedded into a vector store
  2. Retrieval — a query embedding finds the top-k most similar chunks
  3. Generation — retrieved context is passed to the LLM via system prompt

Use similarity_threshold: 0.8 to filter low-relevance matches.

Usage 

<AiMarkdown content="# Hello\n\nGFM **bold**, tables, and task lists." />

Props

Prop Type Default Description
contentstring''—
parser'marked' | 'simple'—Use simpleParse for lightweight regex parsing

Slots

  • default