Route each security questionnaire question to the right source
Each question in an incoming security questionnaire is sorted by a decision model into answerable from the library, needs an engineer, needs legal, or not applicable, so the first draft starts with the easy ones done.
An example pattern written by Agents Lookup. We have not yet found a public build of this exact job.
The use case
Security questionnaires arrive as long spreadsheets, and most of the time goes into working out who should answer what. Jev reads each question with a short description of your answer library and picks where the answer should come from. Questions classified with a confidence of 0.85 or higher are tagged and sent to that route; the library ones go to Claude to draft from your approved answers. Lower-confidence questions are tagged for the person running the questionnaire. Jev does not answer any question itself or decide what to disclose. The owner reviews every drafted answer, and engineers and legal sign off on the questions routed to them. Test the routes on a past questionnaire you have already answered, and keep the threshold high until the routes match what your team did.
Example workflow
- The questionnaire spreadsheet is uploaded to Google Drive and split into one row per question.
- Each question is sent to Jev with a short summary of the topics your answer library covers.
- Jev answers a Choice question, where should this answer come from, with the allowed answers answer library, engineering, legal or compliance, not applicable, and unsure.
- Questions routed to the answer library are sent to Claude with the matching approved answers to draft a reply.
- Engineering and legal questions are assigned in a Google Sheet with an owner and due date, and unsure ones go to the questionnaire owner.
- The questionnaire owner reviews every answer, gets sign-off where needed, and sends the final file.
Tools
- Jev (TypeSafe AI)
- Google Sheets
- Google Drive
- Claude
- Slack
- Zapier or Make
Risks and guardrails
- A question routed to the library may need a current answer — guardrail: answers in the library carry a last-reviewed date, and stale ones are routed to engineering.
- Not applicable can hide a real gap — guardrail: every not applicable tag is checked by the owner before sending.
- Drafts could overstate controls — guardrail: Claude drafts only from approved answers, and a person signs off on the final file.
Related recipe
Security Questionnaire First-Draft Agent with Cited Sources: An agent drafts security questionnaire answers from your own trust documents and approved answer library, cites every source, and marks gaps NEEDS OWNER for a responsible person to sign.
Want it set up for you? Request a quote.
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- Google Drive (official page): developers.google.com/workspace/drive (checked 2026-11-19)
- Claude (official page): claude.ai (checked 2026-11-09)
- Slack MCP server (official page): docs.slack.dev/ai/slack-mcp-server (checked 2026-11-09)
- Zapier MCP (official page): zapier.com/mcp (checked 2026-11-09)
- Make MCP server (official page): developers.make.com/mcp-server (checked 2026-11-09)