Documentation

VirtualBoard is a GitHub-native feature management platform that transforms your .virtualboard/ folders into powerful workspaces for managing feature specifications.

Instead of scattered docs, wikis, or project management tools, VirtualBoard keeps your feature specs right where they belong—in your repository, version-controlled and synchronized with your code.

• •
  Workspaces: Each connected GitHub repository becomes a workspace
• •
  Features: Markdown files with frontmatter metadata tracking specs
• •
  Status Workflow: Features move through backlog → in-progress → review → done
• •
  Active Features: Features in backlog, in-progress, review, or blocked status (done features don't count toward limits)

VirtualBoard integrates deeply with GitHub through OAuth and the GitHub App. Here's what happens behind the scenes:

1. Authentication: You sign in with GitHub OAuth to establish your identity
2. Repository Access: Install the VirtualBoard GitHub App to grant repository access
3. Discovery: VirtualBoard scans for .virtualboard/ folders
4. Indexing: Feature files are parsed and indexed into the VirtualBoard database
5. Sync: Webhooks keep VirtualBoard updated when you push changes to GitHub

VirtualBoard performs the following GitHub operations:

• ✓
  Read Operations: List repositories, read file contents, check commit SHAs, view branches
• ✓
  Write Operations: Update files in .virtualboard/ folder, create commits when you edit specs
• ✓
  Webhook Events: Receive push events to keep workspace in sync

Every feature follows a standard workflow:

Each feature file includes YAML frontmatter with structured metadata:

---
id: FTR-0042
title: User Authentication
status: in-progress
owner: backend-team
priority: P1
complexity: M
labels: [auth, security, backend]
dependencies: [FTR-0001, FTR-0023]
---

# Feature Spec: User Authentication

## Summary
Implement OAuth 2.0 authentication...

VirtualBoard uses AI to help you write better feature specifications through an interactive, multi-step process:

1. 1.
   Analysis: AI analyzes your current spec and identifies gaps, unclear requirements, or missing details
2. 2.
   Questions: AI generates targeted questions to clarify ambiguities and gather more context
3. 3.
   Your Input: You answer the questions, providing the AI with necessary context
4. 4.
   Enhancement: AI generates an improved version of your spec with better clarity and completeness
5. 5.
   Review: You review the enhanced spec and can iterate further if needed

For every AI enhancement, VirtualBoard shows you the complete thought process including:

• •
  Analysis Reasoning: Why the AI identified certain gaps or issues
• •
  Question Rationale: Why specific questions were asked
• •
  Enhancement Strategy: How the AI decided what to improve and add
• •
  Step-by-step Progress: Each phase of the enhancement process is tracked

VirtualBoard provides complete transparency into AI usage costs. For every AI operation, you can see:

Track your AI usage across your entire workspace:

• •
  Per-Feature Breakdown: See costs for each feature enhancement
• •
  Workspace Total: Aggregate AI spend across all features
• •
  Model Information: Which AI model was used (e.g., claude-3-5-sonnet)
• •
  Duration Tracking: How long each request took to complete

Your plan limits are based on active features, which include:

• ✓
  Features in backlog status
• ✓
  Features in in-progress status
• ✓
  Features in review status
• ✓
  Features in blocked status
• ×
  Features in done status (don't count!)