PageSourceSearch

https://xpharma.ai/assets/data-room-BeyZMAgC.js

js xpharma.ai collected 2026-10-05 14:11:30 UTC 2,249 bytes, 43 lines download raw bytes

1const e=`## What the data room is
2
3Every published asset has a data room — a structured set of documents organized by category, with confidentiality controls and AI-driven extraction. Buyers see the data room as part of the asset listing and can request access to confidential documents.
4
5Suppliers manage the data room from the **Asset Workspace**. Uploads, categorization, and confidentiality controls all live there.
6
7## How AI extraction works
8
9When you upload a document, xPharma reads it twice:
10
111. **Categorization** — assigns the document to the right category in the pharma-specific taxonomy.
122. **Extraction** — pulls structured information (target, MOA, preclinical results, IP claims, clinical stage, and more) into the asset's profile.
13
14Each extracted field shows a confidence indicator and a link back to the source document, so you can always trace a value back to where it came from. Manual edits you make are never overwritten by re-extraction.
15
16## Confidentiality controls
17
18Every document has a confidentiality flag.
19
20- **Non-confidential** documents are visible to buyers as part of the asset listing.
21- **Confidential** documents are hidden from buyer views entirely. Buyers see that confidential documents exist (count and category breakdown) but cannot read them.
22
23Set the flag at upload time, or update it later from the document row. There is no "watermarked preview" — confidential means confidential.
24
25## Document statuses
26
27Each document moves through a few statuses:
28
29- **Processing** — upload in progress or extraction running.
30- **Categorized** — auto-categorized successfully.
31- **Manual** — you've overridden the category.
32
33Processing usually takes under a minute for a typical PDF; very large files can take longer.
34
35## Storage and access
36
37Documents are stored privately. Every read — by the supplier, by an authorized buyer, by an admin — produces a short-lived link rather than a public URL. Access is enforced by your role and the asset's status:
38
39- The supplier can read all documents on their own assets.
40- Buyers can read non-confidential documents on published assets only.
41- Admins can read everything, with their actions logged.
42- Anyone else gets an access denied.
43`;export{e as default};

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.