OneProLabs · Apache 2.0

Your backups know
more than you think.

Ask questions straight from your document backups — PDFs, Word, Excel, PowerPoint, images, Markdown, or any other text format — without touching production.

Data GovernanceAgentic RAGAgent HarnessNever Disrupts Production
HyperFileLens Overview dashboard showing the data protection pipeline from production source to isolated target storage to recovery drill, with 452 sources and 428.7 TB protected, all systems healthy, and recovery verification ready.

Use Cases

Answers, not just backup search results.

Customer support Q&A

Answer customer questions straight from your product docs, PDFs, and slide decks — multimodal understanding reads across formats, so nothing needs to be reformatted first.

Company knowledge base

Turn scattered specs, policies, and decisions — engineering or company-wide — into one knowledge base every team can actually query, not just search and summarize.

Source-grounded root cause analysis

Not keyword matching, not log scraping — the agent reads and reasons through your real source code, the way a senior engineer would, until it can confirm the actual root cause.

Still a rock-solid backup tool

Any host, plus NAS

Windows, Linux, and Mac — servers and workstations alike — plus NAS shares, all protected under one policy, not three different tools.

No storage lock-in

Object storage or local storage, any S3-compatible provider — your backups aren't tied to one storage vendor.

Single-file recovery

Need one file back right now? Browse any snapshot and restore just that file in seconds — no full-volume restore required.

How It Works

Your files stay put. The engine comes to them.

HyperFileLens backs up your documents and code into a safe, isolated copy — then SourceLens, our open source Agentic RAG engine, reasons directly over that copy, never touching production.

Your documents (Word, PDF, Excel, PowerPoint, images, code, markdown) flow into HyperFileLens for open source backup, protection, and flexible storage. HyperFileLens produces an isolated safe copy with no impact on production, which the SourceLens open source AI agent reasons over directly — reading, searching, navigating, and reasoning with no pre-built index — to produce AI insights: customer support Q&A, a company knowledge base, source-grounded root cause analysis, and actionable insights. The whole pipeline is governed and secure: access control, data stays in your control, audit and compliance, enterprise ready. Open source, Apache 2.0.

Community

Open source.
One-command install.

Install the latest Community tag on an Ubuntu host with Docker. The installer verifies its images and assets before changing the host.

Community is free and open source, with S3-compatible storage and an AI model or API key. Enterprise capabilities will be available in a later release.

HyperFileLens is currently in public beta.

CommunityLatest tag

Run on an Ubuntu host

# Install the latest Community tag
$ curl -fsSL \
  https://raw.githubusercontent.com/oneprolabs/hyperfilelens/main/deploy/online/install.sh \
  | sudo bash -s -- --mirror global --yes
Installer handles the restEnvironment check · Image pull · Service startup

Blog

Latest writing

Give Kopia an AI Brain

Kopia's dedup and encryption are solid, but its CLI runs on memorized flags and KopiaUI isn't much better — once something's backed up, it disappears into a black box. HyperFileLens builds its Agent directly on Kopia, collapses setup into a few clicks, and wires in an AI engine (SourceLens) that reads snapshots directly. A full hands-on walkthrough, from deployment to asking AI questions about your own backups.

Ray SunSep 9, 2026

Help AI Understand Your Business Data, Starting With Your Backups

From recovering data to rediscovering its value — the open-source HyperFileLens skips Embedding, Vector DB, and pre-indexing entirely, letting an agent search, read, and reason over raw files directly. It's already running in production across HyperBDR and AGIOne's global sales support, product support, and engineering analysis.

Ray SunSep 7, 2026

View all posts

Contact

Contact us

Enterprise deployment

Planning a private, self-hosted deployment for your organization? We can help you scope it.

Email us

Technical support

Running into an issue or found a bug? File it on GitHub and we'll take it from there.

Open an issue