{
  "name": "rag_sqlite_vec_demo",
  "description": "Zero-infrastructure RAG using the bundled SQLite + sqlite-vec vector store. Requires conductor.db.type=sqlite and conductor.integrations.ai.enabled=true, which auto-registers the 'default' vector DB instance. Embeddings are requested at 256 dimensions to match the default instance.",
  "version": 1,
  "schemaVersion": 2,
  "tasks": [
    {
      "name": "index_doc_1",
      "taskReferenceName": "index_doc_1_ref",
      "type": "LLM_INDEX_TEXT",
      "inputParameters": {
        "vectorDB": "default",
        "index": "demo_index",
        "namespace": "demo_docs",
        "docId": "intro-001",
        "text": "Conductor is a distributed workflow orchestration engine. It lets developers build complex stateful applications by orchestrating microservices and AI agents.",
        "embeddingModelProvider": "openai",
        "embeddingModel": "text-embedding-3-small",
        "dimensions": 256,
        "metadata": { "category": "introduction" }
      }
    },
    {
      "name": "index_doc_2",
      "taskReferenceName": "index_doc_2_ref",
      "type": "LLM_INDEX_TEXT",
      "inputParameters": {
        "vectorDB": "default",
        "index": "demo_index",
        "namespace": "demo_docs",
        "docId": "vectordb-002",
        "text": "Conductor supports several vector databases: PostgreSQL (pgvector), MongoDB Atlas, Pinecone, and an embedded SQLite backend powered by the sqlite-vec extension that needs no external server.",
        "embeddingModelProvider": "openai",
        "embeddingModel": "text-embedding-3-small",
        "dimensions": 256,
        "metadata": { "category": "features" }
      }
    },
    {
      "name": "index_doc_3",
      "taskReferenceName": "index_doc_3_ref",
      "type": "LLM_INDEX_TEXT",
      "inputParameters": {
        "vectorDB": "default",
        "index": "demo_index",
        "namespace": "demo_docs",
        "docId": "sqlite-003",
        "text": "When SQLite persistence and the AI integration are both enabled, Conductor bundles the sqlite-vec native extension and registers a default vector store automatically, so semantic search works out of the box.",
        "embeddingModelProvider": "openai",
        "embeddingModel": "text-embedding-3-small",
        "dimensions": 256,
        "metadata": { "category": "configuration" }
      }
    },
    {
      "name": "search_index",
      "taskReferenceName": "search_ref",
      "type": "LLM_SEARCH_INDEX",
      "inputParameters": {
        "vectorDB": "default",
        "index": "demo_index",
        "namespace": "demo_docs",
        "query": "${workflow.input.question}",
        "embeddingModelProvider": "openai",
        "embeddingModel": "text-embedding-3-small",
        "dimensions": 256,
        "maxResults": 3
      }
    },
    {
      "name": "generate_rag_answer",
      "taskReferenceName": "answer_ref",
      "type": "LLM_CHAT_COMPLETE",
      "inputParameters": {
        "llmProvider": "openai",
        "model": "gpt-4o-mini",
        "messages": [
          {
            "role": "system",
            "message": "You are a technical expert. Answer the question using only the provided context."
          },
          {
            "role": "user",
            "message": "Context:\n${search_ref.output.result}\n\nQuestion: ${workflow.input.question}"
          }
        ],
        "temperature": 0.2
      }
    }
  ],
  "inputParameters": ["question"],
  "outputParameters": {
    "search_results": "${search_ref.output.result}",
    "answer": "${answer_ref.output.result}"
  }
}