> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tensorpro.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Provenance and grounding

> Read grounded answers and citations in the Workspace.

## Citations in answers

Answers in the Workspace carry inline citation markers. Selecting a citation
shows the source it points to, along with a provenance tag that tells you where
that information came from, for example whether a user stated it, a user
confirmed it, or a model inferred it.

## The Memory panel

The Memory panel on the right shows how an answer was grounded:

* The Context tab lists the records used for the latest turn, each with its
  provenance tag.
* The Records tab shows your recent memory records.

Records that came from a model inference or generation can be confirmed from
this panel. Confirming a record promotes it to a user-confirmed fact.

## Why this matters

Provenance is what makes EDN answers auditable: you can trace each part of an
answer back to its origin, and inferences are always shown as inferences rather
than as established fact. For the model behind this behavior, see
[Core concepts](/developer/core-concepts).