Skip to Oxygen content

Analysts8 min read

Org Corpus vs Workspace Mounts: Two Ways to Bring Files into a Local AI Studio

Serious local AI needs two modes: durable institutional memory, and “open this project folder and work.” Oxygen supports both — sticky org corpus uploads into on-device storage, and workspace mounts that bind a folder from your machine into the sandbox.

Org corpus: sticky memory for the team ritual

Upload CSVs, PDFs, and docs into organization-scoped storage. Files land on your machine — native data directory or local Docker volumes, depending on the build — and are parsed into advanced storage — including graph and vector layers — so chat, Studio agents, and flows can query the same memory later. This is for handbooks, recurring MLS pulls, board packs, SOP libraries: things you want to keep around.

Workspace mounts: project folder as a volume

When the job is a live project — a repo, a research folder, a client packet on disk — mount that folder. Oxygen loads it into the sandbox so agents can read and write against the tree. It feels like a lightweight built-in IDE path: browse files on the left, chat in the center, open editors when you need to inspect or tweak.

  • Best for active analysis folders and small builds.
  • Keeps the source of truth on your filesystem.
  • Pairs with sandbox packages and the script runner for exploratory work.

Which mode should you start with?

  1. Recurring role playbook (CMA, Monday metrics, policy Q&A) → start with org corpus uploads.
  2. One-off deep dive on a folder you already have → mount the workspace.
  3. Many teams use both: sticky reference docs in corpus, active exports in a mounted week folder.

Privacy posture stays the same

Both modes stay on-machine in the sandboxed local stack. Optional hosted models receive prompt/context you send — not a silent sync of every file to a public chat product.

FAQ

Are org uploads stored in the cloud?
Org corpus uploads stay on your machine in Oxygen's data plane — native OS-sandboxed storage or local Docker volumes, not a generic cloud drive for the files themselves.
Can agents edit my mounted folder?
Workspace mounts are bound into the sandbox environment so agents can work against that tree. Treat it like granting a local tool access to a project folder — mount only what you intend to work on.