System Requirements Document for quedote

1. Introduction

The "Questionnaire Copilot" is a production-ready B2B SaaS web application designed to automate the process of filling out security questionnaires for small to mid-sized B2B SaaS companies. The application leverages AI to extract questions from uploaded questionnaires, search company knowledge bases, generate evidence-backed answers, and facilitate human review and export of completed questionnaires. The target audience includes companies with 10–100 employees that regularly receive security questionnaires.

2. System Overview

The system provides a streamlined workflow for handling security questionnaires, focusing on efficiency and accuracy. It supports document uploads, AI-driven question answering, human review, and export functionalities. The application ensures secure multi-tenant data isolation and offers a demo mode for immediate testing. The system excludes integrations and complex features in its initial version, focusing on core functionalities.

2a. Product Interpretation and Delivery Boundary

The application is delivered as a web-based platform accessible via a browser. It requires user authentication for accessing private workspaces and documents. The system is designed to handle sensitive data securely, with features like row-level security and private file storage. The initial version excludes integrations with external services like Slack or AWS and focuses on the core questionnaire processing workflow.

2b. Source Content Inventory

Not applicable as no content source directive was provided.

2c. Page Content and Component Coverage

Landing Page

• Information: Product overview, target audience, headline, subheading, and CTA.
• Primary Actions: Start Free, Learn More.
• Components: Hero section with split-screen diagrammatic workflow and CTA.

Login

• Information: User authentication for returning users.
• Primary Actions: Sign in, Forgot Password.
• Components: Login form, password recovery.

Dashboard

• Information: Recent questionnaires, total questions, answered, needs review, insufficient evidence, completion percentage.
• Primary Actions: View Questionnaire, Access Demo.
• Components: Summary cards, progress bars, demo entry point.

Knowledge Base

• Information: Upload and manage company documents.
• Primary Actions: Upload Document, Search.
• Components: Document list, upload form, search bar.

Questionnaires

• Information: Browse and upload questionnaires.
• Primary Actions: Upload Questionnaire, View Details.
• Components: Questionnaire list, upload form, question count display.

Questionnaire Workspace

• Information: Review and edit AI-generated answers.
• Primary Actions: Approve, Edit, Regenerate, Mark N/A.
• Components: Question display, AI answer, confidence indicator, evidence/source, filters, categories.

Answer Memory

• Information: Stored approved answers for semantic reuse.
• Primary Actions: View Answer, Edit.
• Components: Answer list, semantic matching.

Export

• Information: Export completed questionnaires.
• Primary Actions: Export as XLSX, CSV, DOCX.
• Components: Export options, format selection, evidence inclusion toggle.

3. Functional Requirements

1. As a Workspace User, I should be able to sign up and log in to access my company workspace.

   • Provenance: explicit
   • Observable Acceptance: Successful authentication and workspace access.

2. As a Workspace User, I should be able to upload knowledge-base documents in various formats and have them processed for searchability.

   • Provenance: explicit
   • Observable Acceptance: Documents are indexed and searchable.

3. As a Workspace User, I should be able to upload questionnaires and have questions automatically extracted.

   • Provenance: explicit
   • Observable Acceptance: Display of detected question count.

4. As a Workspace User, I should be able to view a dashboard with recent questionnaires and progress metrics.

   • Provenance: explicit
   • Observable Acceptance: Accurate display of metrics and recent activity.

5. As a Workspace User, I should be able to review AI-generated answers, inspect evidence, and approve or edit answers.

   • Provenance: explicit
   • Observable Acceptance: Ability to approve, edit, or regenerate answers.

6. As a Workspace User, I should be able to export completed questionnaires in multiple formats while preserving the original structure.

   • Provenance: explicit
   • Observable Acceptance: Successful export with correct format and structure.

7. As a Workspace User, I should be able to filter questions by status and category.

   • Provenance: explicit
   • Observable Acceptance: Correct filtering and categorization of questions.

8. As a Workspace User, I should be able to use approved answers for future similar questions through semantic matching.

   • Provenance: explicit
   • Observable Acceptance: Reuse of relevant approved answers.

9. As a Workspace User, I should be able to access a demo mode with sample data for testing.

   • Provenance: explicit
   • Observable Acceptance: Immediate access to demo content.

10. As a Workspace User, I should be able to manage my subscription plan and usage limits.

    • Provenance: explicit
    • Observable Acceptance: Accurate billing and usage tracking.

4. User Personas

• Workspace User: Responsible for uploading documents and questionnaires, reviewing AI-generated answers, approving answers, and exporting completed questionnaires.

5. Core User Flows

Flow 1: User Authentication and Workspace Access

1. User navigates to the login page.
2. User enters credentials and logs in.
3. User is redirected to the dashboard.

Flow 2: Uploading and Processing Knowledge Documents

1. User accesses the Knowledge Base page.
2. User uploads a document in a supported format.
3. System processes and indexes the document for searchability.

Flow 3: Uploading and Extracting Questionnaire Questions

1. User navigates to the Questionnaires page.
2. User uploads a questionnaire file.
3. System extracts questions and displays the count.

Flow 4: Reviewing and Approving AI-Generated Answers

1. User accesses the Questionnaire Workspace.
2. User reviews AI-generated answers and evidence.
3. User approves, edits, or regenerates answers as needed.

Flow 5: Exporting Completed Questionnaires

1. User navigates to the Export page.
2. User selects export format and options.
3. System exports the completed questionnaire.

Flow 6: Using Demo Mode

1. User accesses the dashboard.
2. User selects the demo mode.
3. System provides access to sample data and questionnaires.

6. Visuals Colors and Theme

• Muse: Dieter Rams
• Palette (light mode):
  • Background: #F7F7F7
  • Surface: #FFFFFF
  • Text: #333333
  • Primary: #FF6F00
  • Accent: #009688
  • Muted: #BDBDBD
• Typography:
  • Headings: Archivo, Medium weight, all caps
  • Body: IBM Plex Sans
  • Scale: 1.25 modular
• Shape Language: Rounded-rectangle controls
• Layout: Strict grid layout
• Imagery: Diagrammatic line art

7. Signature Design Concept

The landing page features a split-screen hero with a diagrammatic representation of the workflow on one side and a call-to-action on the other. The primary orange color is used for emphasis, and rounded-rectangle cards display questions and answers. The dashboard layout follows a strict grid with small caps labels for filters and categories.

8. Interaction Model & Motion Direction

• Interaction Model: Static
• Motion Tempo: Restrained
• Hero Dimensionality: Flat
• Landing Hero Motion Brief: The hero section features a composed split-screen with a diagrammatic workflow on one side and a CTA on the other. The primary orange color highlights the CTA, providing instant feedback with mechanical precision.

9. Non-Functional Requirements

• Security: Secure authentication, multi-tenant isolation, row-level security, private file storage, server-side API keys, and audit logs.
• Performance: The system should handle concurrent uploads and processing efficiently.
• Scalability: The application should support scaling to accommodate more users and data as needed.

10. Tech Stack

• Frontend: Next.js, TypeScript, Tailwind, shadcn/ui
• Backend: Next.js API/server actions
• Database: Supabase PostgreSQL
• Storage: Supabase Storage
• AI: OpenAI API with structured outputs and retrieval/RAG
• Payments: Stripe
• Deployment: Vercel, Supabase

11. Assumptions and Constraints

• The application will not include integrations with Slack, AWS, or other external services in V1.
• The system will not support mobile apps or browser extensions in V1.
• The application will focus on the core workflow of questionnaire processing without additional features like full GRC or SOC 2 management.

12. Glossary

• AI Security Questionnaire Autofiller: The core product functionality that automates the filling of security questionnaires using AI.
• Workspace User: The primary user persona responsible for interacting with the application.
• Semantic Matching: The process of using AI to identify and reuse relevant approved answers for similar questions in future questionnaires.