Knowledge Base & RAG
Upload documents and connect your voice agent to native semantic search.
Orbitali provides a built-in Retrieval-Augmented Generation (RAG) system. You can upload company policy sheets, FAQs, or product catalogs directly, and the agent will use this context to answer caller inquiries natively.
This eliminates the need to build a custom vector database, handle document chunking, or create external search APIs.
Ingesting Documents
You can manage knowledge documents through the Knowledge Base tab in the developer dashboard, or programmatically through the public API.
Supported formats
- Text (
.txt): Used as-is. - Markdown (
.md): Parsed directly to plain text. - PDF (
.pdf): Text is extracted programmatically.
Uploading via the API
Use the public API to add documents from scripts or coding agents. Send raw text as JSON, or upload a file (including PDF) as multipart form data.
# JSON with text — ideal for coding agents
curl https://api.orbitali.ai/public/v1/agents/$AGENT_ID/knowledge \
-X POST \
-H "Authorization: Bearer $ORBITALI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"name": "Cancellation policy",
"content": "Customers may cancel up to 24 hours before their appointment..."
}'
name and description are optional; when omitted, Orbitali suggests them from the document content. See the API reference for the multipart upload, listing, and deletion endpoints.
Document processing
When a document is uploaded:
- Orbitali extracts its text and divides it into searchable passages.
- The passages are indexed for semantic retrieval within the agent's knowledge base.
- Once processing completes, the document status becomes
ready, and the agent can use it during conversations.
Orbitali manages the indexing and retrieval infrastructure. You do not need to configure an embedding model, vector dimensions, or database indexes.
The search_knowledge System Tool
As soon as an agent has at least one ready document, Orbitali automatically makes the search_knowledge tool available during conversations.
You do not need to register this tool manually; it is handled natively by Orbitali and does not call your backend Server URL.
Tool Contract
The tool is exposed to the model with the following function declaration:
{
"name": "search_knowledge",
"description": "Search the agent's knowledge base for information relevant to the caller's question. Use this before saying you don't know something.",
"parameters": {
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "The question or topic to search for."
}
},
"required": ["query"]
}
}
Runtime Query Flow
When a caller asks a question that requires knowledge lookup:
- Knowledge lookup: The agent determines it needs to call
search_knowledgewith a query string:{ "name": "search_knowledge", "arguments": { "query": "What is your cancellation policy?" } } - Semantic search: Orbitali searches the current agent's knowledge base for passages relevant to the query.
- Context retrieval: Relevant passages are returned to the agent to help it answer the question.
- Response: The agent uses the retrieved facts to respond to the caller.