Skip to main content
Bizzy lets you upload files — contracts, invoices, product sheets, runbooks — and makes them searchable by your agents. This page covers what happens between “I dropped a PDF in” and “my agent quoted the right paragraph back to me.”

​
What is a File?

A file is any artefact you’ve uploaded to your organization. Each file:
  • Is stored securely in Bizzy storage.
  • Belongs to an organization, optionally inside a folder.
  • Has metadata — filename, content type, size, tags, description.
  • Has an indexing status that tracks whether it’s searchable yet.
Today, the indexing pipeline reads these formats end-to-end via OCR: PDF, Word (.docx), PowerPoint (.pptx), and OpenDocument text and presentations (.odt, .odp), plus the images PNG, JPEG, and WebP — alongside inline Markdown and other plain-text formats. Spreadsheets and other formats can still be uploaded and shared via public links — they simply aren’t indexed for retrieval.

​
Folders

Files can be grouped into folders. Folders are useful for two reasons:
  1. Organisation. Same as folders anywhere — group what belongs together.
  2. Agent access control. Each file (and each folder) carries an Agent access flag. Turn the flag off on a folder and your agents will not see anything in it during retrieval, even if a query would otherwise match.
This matters more than it sounds: it’s how you separate, say, your public sales collateral from internal HR documents when both are in the same organization.

​
From Upload to Searchable

When a file is uploaded, an automatic pipeline takes it from raw bytes to retrievable knowledge:
  1. Store. The file is saved and appears in your file list immediately.
  2. Extract. PDFs, office documents (Word, PowerPoint, OpenDocument), and images are converted to clean, structured text via OCR. Markdown and other text files are read directly.
  3. Chunk. The text is split along its natural structure — sections, paragraphs, sentences — so each piece stays coherent on its own, with a little overlap between neighbouring pieces so context isn’t lost at boundaries.
  4. Index. Each piece is indexed for both semantic (meaning-based) search and keyword search.
A file’s indexing status walks through pending → indexing → indexed. If a file isn’t indexable (unsupported format, too large, or its text couldn’t be extracted) the status ends in skipped, excluded, or failed — visible on the file’s detail view so you know it’s not searchable.

​
How Agents Retrieve Files (RAG)

When an agent needs information from your files, Bizzy uses hybrid retrieval: a semantic search that is good at fuzzy, meaning-level matches, and a keyword search that is good at exact terms — run together, with the best-matching passages from both handed to the agent as context. You get relevant results whether you ask “what’s our refund policy?” or search for an exact part number. Two important guarantees are enforced on every retrieval:
  • Only content from the agent’s own organization is ever returned.
  • Files (and folders) with Agent access turned off, files that aren’t fully indexed, and files that have been deleted are silently excluded.
Your agent never sees what you’ve told it not to see.

​
What it Costs

File usage is metered through two of your account’s consumables:
  • Storage — the total bytes in your file library, counted against your tier’s storage allowance.
  • LLM Tokens — indexing a file consumes embedding tokens, which count toward the same token allowance your agents and automations use.
Both are included in your tier’s allowance and charged at the relevant overage rate beyond it. See the Credits page for how the meters and tiers fit together.

Upload Files

How to add files in the web app

RAG Indexing

Re-index files, exclude folders, troubleshoot

Agents

How agents use files at runtime

Credits & Billing

What storage and indexing cost
Last modified on June 24, 2026