Local file search
Searches files on a local system instead of uploading them to a cloud service.
Reference is a local file search tool for AI agents that adds citations to results and avoids cloud uploads. It is aimed at workflows where agents need to search files on a local system and return source-backed answers.
Reference is a local file search product for AI agents. It is designed to search files on a local system and return results with citations, while keeping the workflow off the cloud.
The product page presents Reference as a way to support agent workflows that need grounded answers from local files. Its emphasis is on local search and citations, so the output is meant to be traceable back to source material rather than treated as uncited text.
Searches files on a local system instead of uploading them to a cloud service.
Returns search results with citations so users can trace outputs back to source material.
Positions file search as part of an agent workflow rather than as a general consumer search tool.
Keeps file content local, which is the main workflow signal provided on the site.
Use Reference when an AI agent needs to answer questions using documents stored on a local machine.
Use the citation-backed output when a user needs to verify which file or source supported the result.
Choose the tool when keeping files off a cloud service is part of the required workflow.
Reference is a local file search product for AI agents that returns cited results.
No cloud uploads are mentioned on the product site, which suggests the workflow is kept local.
The site describes it as a tool for AI agents that need to search local files.
The provided page text does not include pricing details.