New Feature

Reputation Cog

The Reputation cog automatically tracks and scores helper performance in support communities. It analyzes conversations to identify helpful responses and builds reputation scores for community members.

Features

Automated Help Tracking

  • Detects help-seeking posts (problem statements) in channels
  • Identifies users providing helpful responses
  • Tracks complete help sessions until solution
  • Cross-post detection prevents duplicate crediting

Reputation Scoring

  • Automatic points for confirmed help contributions
  • Score decay over time
  • Leaderboard rankings for top helpers
  • Session-based scoring system

Session Management

  • Auto-detection of problem statements
  • Thread-based session organization
  • OCR-based spoiler detection
  • Automated session closure on solution detection

Configuration

  • Customizable scoring parameters
  • Per-channel and per-role settings
  • Flexible session duration settings

How It Works

The reputation system monitors designated channels for help sessions:

  1. User posts problem statement (detected by keywords/format)
  2. Other users respond with solutions
  3. Problem is marked solved (by reporter or automated detection)
  4. Helper receives reputation points
  5. Sessions tracked in database with timestamps

Commands

Reputation is tracked automatically. View your standing with:

  • View leaderboards and statistics via dashboard or bot commands
  • Reputation adjustments are automatic
  • Moderators can review sessions manually

Configuration

Reputation system requires complex database-driven configuration. Consult server administrators for:

  • Help channels to monitor
  • Scoring parameters
  • Role-based bonuses
  • Session timeout settings
  • Tiers: XP bonuses for helpers
  • Economy: Economic rewards for reputation
  • Tags: Helper resource access

Best Practices

  1. Channel Selection: Monitor appropriate help channels
  2. Score Calibration: Adjust thresholds based on community
  3. Review Process: Regularly audit reputation assignments
  4. Encouragement: Use reputation to motivate helping
  5. Fairness: Ensure algorithm treats users fairly

Technical Details

Session States

  • Active: Currently being helped
  • Solved: Problem resolved
  • Abandoned: No solution found
  • Ghost: OP left server

Confidence System

  • High Confidence: Clear problem resolution
  • Medium Confidence: Likely solution
  • Low Confidence: Uncertain outcome

Troubleshooting

  • False Positives: Adjust detection parameters
  • Score Disputes: Manual review process
  • Performance Impact: Monitor system resources
  • Accuracy Issues: Fine-tune algorithms