Overview

This PR implements a label-based point system for the leaderboard, replacing the previous flat 10-points-per-PR calculation with a level-based scoring mechanism that better reflects the complexity and value of contributions.

Changes

Previous Behavior

All merged PRs received a fixed 10 points regardless of their complexity, difficulty, or scope.

New Behavior

PRs now earn points based on their labels:

• Must have recode label to earn any points
• Points are assigned based on difficulty level:
  • level 1 → 10 points
  • level 2 → 30 points
  • level 3 → 50 points
• PRs without proper labels receive 0 points

Implementation Details

1. Added label support to the PullRequestItem interface to capture labels from the GitHub API
2. Created calculatePointsForPR() function that:
   • Validates presence of recode label
   • Checks for level labels and assigns appropriate points
   • Handles missing or incomplete labels gracefully
   • Performs case-insensitive label matching
3. Updated point calculation logic to sum individual PR points instead of using flat multiplication
4. Added points field to PRDetails interface to store calculated points per PR

Example

// PR with labels: ["recode", "level 2", "bug"]
// Earns: 30 points

// PR with labels: ["recode", "level 3", "feature"] 
// Earns: 50 points

// PR with labels: ["bug", "documentation"]
// Earns: 0 points (missing recode label)

Benefits

✅ Fairer representation of contribution value
✅ Incentivizes proper PR labeling by maintainers
✅ Rewards challenging work appropriately
✅ Backward compatible - no breaking changes
✅ Minimal code changes - surgical implementation in a single file

Technical Notes

• Label data is already included in GitHub API responses (no extra API calls required)
• All existing time filtering and leaderboard features remain functional
• Type-safe TypeScript implementation with proper error handling

Testing

Verified with comprehensive test cases covering:

• Various label combinations
• Missing label scenarios
• Case-insensitive matching
• Edge cases and boundary conditions

Closes #[issue_number]

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.