Add knowledge - Veryfront

Add knowledge when agents or apps need project-specific retrieval context.

Steps

Add or import the source content.
Create or update RAG documents.
Connect knowledge to the agents or apps that need it.
Check available context before relying on retrieval.

Try it with REST

Search project knowledge: POST /projects/{project_reference}/knowledge/lookup

Terminal

curl -X POST https://api.veryfront.com/projects/support-assistant/knowledge/lookup \
  -H "Authorization: Bearer $VERYFRONT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "refund policy"
  }'

Import a file from connected storage: POST /providers/{provider}/files/{file_id}/import

Terminal

curl -X POST https://api.veryfront.com/providers/google_drive/files/<FILE_ID>/import \
  -H "Authorization: Bearer $VERYFRONT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "project_id": "<PROJECT_ID>",
    "base_path": "knowledge"
  }'

API surfaces

Surface	Start here
REST	Knowledge Lookup, RAG Documents
GraphQL	Semantic Search Files
MCP	Search Knowledge, Import External File

Verify

Search for a phrase from the imported file. Connect the knowledge source to the agent only after search returns the expected context.

Knowledge surfaces

Surface	Use it for
Knowledge manifest	Project-level retrieval configuration.
RAG document	Searchable document context.
Agent connection	Runtime access to retrieval context.

API reference

API	Use
REST	Knowledge and RAG documents.
GraphQL	Knowledge views.
MCP	Knowledge tools.

​Steps

​Try it with REST

​API surfaces

​Verify

​Knowledge surfaces

​API reference