Skip to main content
Rather than asking you to paste excerpts into a chat window, ParleHub gives the AI direct access to the documents stored in your project. When you ask the AI a question that requires reading a contract, scanning a balance sheet, or summarizing a presentation, it reaches into the project’s file store, runs the appropriate tool call, and grounds its answer in the actual source material. You get a traceable, auditable response — not a hallucination built from text you copied in manually.

​
Supported file types

Upload any of the following file types to a project and the AI can work with them immediately. Conversion to searchable markdown happens automatically in the background — you do not need to pre-process or reformat anything.
Scanned PDFs that contain only images are not yet supported for text extraction. Use a PDF that has a text layer, or run the document through OCR before uploading.

​
What the AI can do with files

The AI interacts with your project’s files through four discrete tool capabilities, each of which runs as a real, auditable tool call — not a background prompt injection.

Search Documents

Ask the AI to find every clause, paragraph, or data point in a file that is relevant to your question. The search runs against the converted markdown, so it covers the full document even if it spans hundreds of pages.

Read & Summarize

Direct the AI to read a specific section — a contract clause, an exhibit, a cover memo — and summarize or analyze it. Responses include the source file name and section so you can verify the reference.

Query Spreadsheets

Ask questions against rows and columns in natural language: “What is the total revenue for Q3?” or “Which line items exceed the contingency budget?” The AI translates your question into a structured query against the spreadsheet data.

Convert to Markdown

PowerPoint decks and formatted Word documents are converted to clean, searchable markdown automatically on upload. You can also ask the AI to surface the converted text of a specific file so you can copy and reuse it downstream.

​
How to add files to a project

1

Open the project's Files tab

Navigate to your project in the sidebar and click the Files tab at the top of the project view.
2

Upload your documents

Drag files into the upload area or click Upload Files to browse. You can upload multiple files at once. Supported formats are processed automatically — a status indicator shows when conversion is complete.
3

Confirm the AI can see the files

Once a file shows a green Ready badge, it is available for the AI to use in any conversation in the project. Start a conversation and ask the AI a question that references the document to confirm.
4

Enterprise: connect cloud storage

On Enterprise plans, files can be sourced directly from your organization’s SharePoint or Google Drive site. Connect your cloud storage in Project Settings → Integrations, and the AI can query those files without requiring a manual upload step.

​
Grounded answers

Because the AI works by calling file tools — not by consuming a blob of pasted text — every answer it gives is traceable back to a specific source document. Each response that draws from a file includes a citation block showing the file name, the page or section referenced, and the tool call that retrieved it. This traceability matters in professional services contexts where the answer is only as good as the source. When you take an AI-assisted conclusion into a client deliverable or a legal brief, you need to be able to point to exactly where it came from.
No code execution runs near your client data. File tools are read-only operations — the AI can search, read, and query your documents, but it cannot modify, delete, or export them.
After the AI gives you an answer based on a file, ask it to quote the exact text it drew from. For example: “Show me the verbatim language from the contract that supports that conclusion.” This gives you a defensible reference you can drop directly into a memo or client communication.