Standard RAG Tool

Runs the full RAG pipeline — retrieval from the knowledge base followed by LLM answer generation with citations.

Endpoint

POST /components/{chatbot_id}:standard-rag/run

Example Request

curl -X POST "https://glchat.glair.ai/components/{chatbot_id}:standard-rag/run" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: {your_user_api_key}" \
  -d '{
    "inputs": {
      "query": "What is our refund policy?"
    },
    "config": {}
  }'

Response

{
  "response": "Our refund policy allows returns within 30 days of purchase...",
  "references": [
    {
      "id": "54652883-dc0e-4e08-928e",
      "content": "Returns are accepted within 30 days of purchase...",
      "metadata": { "source": "policy-doc.pdf", ...},
      "score": 0.71
    }
  ]
}

Field	Type	Description
response	string	LLM-generated answer grounded in the retrieved chunks
references	list	Source chunks used to generate the response

Inputs Schema

Field	Type	Required	Default	Description
query	string	Yes	—	Shorthand alias — if standalone_query or generation_query are absent, both fall back to this value
generation_query	string	No	""	Query used for LLM generation. If absent, falls back to query. If query is also absent, defaults to ""
standalone_query	string	No	—	Query used for knowledge base retrieval
history	list[Message]	No	[]	Conversation history for context-aware generation
context	string	No	""	Additional context string to inject into the prompt
additional_instructions	string	No	""	Extra instructions appended to the system prompt
use_case_id	string	No	"answer_question"	Use case identifier that selects the prompt template
generation_strategy	string	No	"stuff"	"stuff" or "refine" — see [Runtime Config]([LINK: runtime-config subpage])
attachment_chunks	list[Chunk]	No	[]	Pre-loaded chunks from file attachments
search_chunk_results	list[Chunk]	No	[]	Pre-fetched search chunks to inject alongside retrieval results
memory_results	list[Chunk]	No	[]	Memory chunks from the user's memory store
retrieval_params	dict	No	{}	Low-level parameters forwarded to the vector store query

---