Overview
Built-in RAG Knowledge Base to give your agents access to your data.
Plug your agents straight into your company's knowledge across Jira, Confluence, GitHub, Notion, SharePoint, Google Drive, Salesforce, and more, so they can answer from your own data.
The full RAG stack (chunking, embedding, hybrid search, reranking) runs inside Archestra. No external vector database or separate retrieval service required.
Configuration
Open Settings > Knowledge. Both an embedding and a reranking model must be set before Knowledge Bases and can be used.
Embedding Configuration
Pick the API key and embedding model. The embedding model vectorizes ingested documents so they can be queried semantically. The same model is used for both indexing and querying, which is why it is locked once saved.
- Key — only keys whose synced models have configured embedding dimensions appear in this list. If yours is missing, go to LLM Providers > Models, sync the provider, and set the dimensions for the embedding model. Supported dimensions: 768, 1536, 3072.
- Model — any embedding-capable model exposed by the selected key.
To change the embedding model, click Drop to clear the existing index — every document will need to be re-embedded on the next connector sync.
Reranking Configuration
Pick the LLM that scores and reorders search results by relevance.
- Key — any LLM provider key.
- Model — any chat model from that provider.
Creating a Knowledge Base
A Knowledge Base is a set of connectors. Create one from the Knowledge page and assign connectors to get data from. The same Knowledge Base can be reused across multiple agents and MCP Gateways.
Files
Files are static documents uploaded from Knowledge > Files and assigned directly to agents or MCP Gateways. Use files when you want reusable retrieval from .txt, .md, .csv, .json, .xml, or .pdf documents without setting up an external connector.
Files use the same visibility model as other knowledge resources.
- Owner — only the uploader can view and query the file.
- Teams — only members of selected teams can view and query the file.
- Organization — anyone in the organization can view and query the file.
Chat attachments stay with one conversation unless they are saved to Knowledge > Files.
For production deployments, file bytes are stored in PostgreSQL by default. To store bytes in S3, configure external blob storage in Knowledge Base Configuration.
Creating a Connector
Connectors pull data from external tools (Jira, Confluence, GitHub, etc.) and feed it into one or more Knowledge Bases. Each connector has a visibility setting that controls who can query its data — see Connector Visibility. For supported types and configuration, see Connectors.
Assigning to an Agent
- Go to Agents in the left sidebar and click the agent you want to attach knowledge to (or create a new one).
- In the Edit Agent dialog, scroll to Knowledge Sources.
- Click Select connectors or knowledge bases and pick one or more entries from the Knowledge Bases and Connectors lists. An agent can be assigned multiple Knowledge Bases or individual connectors.
- Click Update to save.
Once assigned, the agent gains a query_knowledge_sources tool that searches across everything attached to it and pulls back the most relevant documents to answer the user's question.
