TrustDesk
About this demo

TrustDesk — customer security questionnaire automation

A portfolio demonstration by Trevor Fischer · Fischer Product Lab — secure AI systems for trust, risk, and enterprise execution. Live demo.

Why TrustDesk

Every enterprise deal now arrives with a security questionnaire — encryption, access control, incident response, privacy, AI usage. Security, Sales, Legal, and Privacy answer the same questions over and over, by hand, under deal pressure. Answers drift, stale evidence slips through, and leadership has no view of how much revenue is waiting on the security queue.

TrustDesk turns customer trust into a governed, measurable workflow: it classifies each question, maps it to approved control evidence, drafts a response, scores confidence, and routes anything sensitive or stale-backed to a human — with SLA and revenue visibility for leadership.

How confidence works

Every verdict is a pure, deterministic function — no AI calls in V1 — which is what makes it explainable and auditable.

  1. 1Five weighted factors score each answer 0–100 — including evidence freshness, which is derived live from the mapped controls so it can never be faked.
  2. 2The score maps to a band: Auto-Approved (85+), Suggested (65–84), Needs Review (45–64), or Escalate (<45).
  3. 3Hard governance gates override the band: expired or missing evidence caps the verdict, and privacy- or AI-governance questions always require human sign-off — regardless of score.

The thesis in one line: automation must never outrun the evidence or the governance behind it — and the gates encode that judgment so a high score can't bury a stale fact or a legal question.

Security posture

  • Synthetic data only — no real customer, employer, or personal data.
  • Read-only — no write endpoints, forms, sending, or admin surface.
  • No client-side secrets; zero environment variables required.
  • Deterministic scoring — explainable, auditable, no black box.
  • Documented STRIDE threat model and security policy.
  • Hardened supply chain — Dependabot, secret scanning, CodeQL.

Built with

  • Next.js 16
  • React 19
  • TypeScript (strict)
  • Tailwind CSS v4
  • Recharts
  • Zod