External Documents

Let people outside your organisation contribute source documents into a project's or DRP's AI Knowledgebase, via a passphrase-protected link — with or without your review.

Two ways to contribute

External documents feed the AI Knowledgebase, not the requirements list directly — a document becomes context the AI can draw on:

  • Upload link — a Lead Analyst generates a shareable, passphrase-protected link scoped to one project or DRP. Anyone with the link and passphrase can upload straight into that Knowledgebase — no account needed. The Lead Analyst chooses, per link, whether submissions go straight in or wait for review (see below).
  • Manual upload on someone's behalf — a Lead Analyst or Analyst uploads material they received outside Requio (an email attachment, something from a meeting) from the External Documents Triage screen. This always goes through the review queue.

Both paths require the Pro plan or above.

Manual upload and triage

A Lead Analyst or Analyst can upload files on a stakeholder's behalf from the External Documents Triage screen, for material that arrived outside Requio entirely. Multiple files can be selected and submitted in one go; each shows up in the triage list right away, but stays malware-scanning until clean — review actions (AI summary, approve, reject) aren't available until the scan clears.

Each pending upload shows an AI-generated summary of what the document actually contains, to help decide whether to approve it, reject it, and where it belongs.

  • Approve — choose where it goes: the project itself, or a specific DRP within it (the Target DRP list only appears once you pick a DRP). "Approve & Add to AI KB" then asks you to confirm before the document is added to that Knowledgebase.
  • Reject — provide a reason, then confirm. The document is not added anywhere, and the reason is kept on record and shown against the upload.

Both actions ask first because both matter: an approved document starts shaping what the AI writes for everyone on that project, and a rejection is recorded against the person who sent it. Afterwards the screen says plainly what happened.

If AI-Knowledgebase indexing fails after approval (a transient error), a "Retry AI KB indexing" action appears on the approved upload — approval itself is not undone.