01
Upload the file you were sent
XLSX, CSV or PDF — devguard parses it into individual questions and shows which columns it read. If the wrong columns were picked, choose them manually and rescan; there’s no need to re-upload.
Upload the XLSX, CSV or PDF you received and devguard reads it into questions, pre-fills what your approved answer library already knows, and lets you review, approve and export the rest. Every answer you approve joins the library, so each questionnaire makes the next one faster.
The file you received becomes a question-by-question review, pre-filled from your library, and goes back out the way it came in.
XLSX, CSV or PDF — devguard parses it into individual questions and shows which columns it read. If the wrong columns were picked, choose them manually and rescan; there’s no need to re-upload.
Each question is matched against your answer library and pre-filled where a confident match exists. Every answer carries its provenance (library, AI or hand-written), and an attention flag marks anything you should read closely before approving.
Optionally, AI auto-fill drafts answers for the remaining questions from your organization’s own data. Drafts are flagged as AI, flagged for attention when unsupported, and never approved on your behalf — the loop works fully without them.
Work the rail question by question — editing an answer withdraws its approval until you approve it again, so what you export is always vetted. Export as XLSX or CSV to send back, and every approved answer joins the library for next time.
Four choices behind how questionnaires work here — each one something you can check, not an adjective.
Every answer you approve is promoted into the answer library, and the next questionnaire that asks the same thing is pre-filled. The tenth questionnaire starts mostly answered.
Each answer shows where it came from: your library, AI, or written by hand. An attention flag marks low-confidence matches and undrafted claims, so you know exactly what to read closely.
Upload, parse, match, review and export never require generation. AI auto-fill is an explicit, optional accelerator for the remaining questions — never a silent author.
Export the completed questionnaire as XLSX or CSV and return it the way it arrived. Your customer doesn’t need an account, a portal or a login to receive your answers.
Hosted in Switzerland by default, in German and English, with on-premise possible. Your data and evidence are yours and exportable in full at any time, with no lock-in.
A questionnaire answer is a claim about your ISMS — the policies, evidence and vendor work it describes live in the same workspace.
The reverse direction — the questionnaires you send out.
Opt-in drafts for unanswered questions, from your own data.
The source of truth your answers describe.
The proof behind the claims your answers make.
Upload XLSX, CSV or PDF. After parsing, devguard shows which columns it read the questions and answers from — if it picked wrong, choose the columns yourself and rescan without re-uploading. Completed questionnaires export as XLSX or CSV.
No. The core loop of upload, parse, match, review, approve and export runs entirely without AI. AI auto-fill is opt-in and drafts answers for unmatched questions from your organization’s own data; anything it drafts is flagged for your review before approval.
Two ways. Every answer you approve on a live questionnaire is promoted into the library. And when you upload a questionnaire you’ve already answered elsewhere, its answers can seed the library directly — so your first live questionnaire already has matches.
That’s a separate module, kept apart on purpose. Vendor questionnaires are the ones you send out as part of vendor assessments, with builder, scoring context and external invite links. Security questionnaires are the ones your customers send you. This page is the inbound side.
Swiss-hosted by default, in German and English, with on-premise possible. Your questionnaires and answer library are yours and exportable in full at any time, with no lock-in.
Upload the next questionnaire a customer sends, review what the library already answered, and export it back — with an answer library that compounds instead of a copy-paste doc that rots.