Skip to content

Features Overview

BatraXPankaj edited this page Nov 13, 2025 · 1 revision

Features Overview

Smart Issue Analyzer provides 12+ AI-powered capabilities to streamline your issue management workflow.

Core Analysis Features

1. Duplicate Detection

What it does: Identifies if a new issue duplicates existing issues
How it works: Compares against all repository issues using semantic similarity
Action taken: Auto-closes duplicate with reference to original

Example:

Issue #23: "Login not working on mobile"
→ Detected duplicate of #12
→ Auto-closed with comment: "Duplicate of #12"

Learn more


2. Auto-Labeling by Topic

What it does: Applies relevant topic labels automatically
Labels applied:

  • bug - Defects and errors
  • enhancement - New features
  • documentation - Documentation updates
  • security - Security concerns
  • performance - Performance issues
  • ui/ux - Interface improvements

Example:

"The dashboard loads slowly for 100+ items"
→ Labels: performance, ui/ux

Learn more


3. Priority Scoring

What it does: Assigns priority levels (P0-P3) based on impact and urgency

Priority Levels:

  • P0 - Critical: Production down, security breach, data loss
  • P1 - High: Major feature broken, significant user impact
  • P2 - Medium: Minor bugs, non-critical enhancements
  • P3 - Low: Nice-to-have features, minor improvements

Example:

"Users cannot checkout - payment system down"
→ Priority: P0 (payment critical for business)

Learn more


4. Size Estimation

What it does: Estimates development effort using t-shirt sizes

Size Scale:

  • XS (1-2 hours): Simple config change, typo fix
  • S (2-4 hours): Small bug fix, minor feature
  • M (1-2 days): Moderate feature, complex bug
  • L (3-5 days): Large feature, architectural change
  • XL (1+ week): Epic-level work, major refactor

Example:

"Add dark mode theme support"
→ Size: L (CSS overhaul, state management, persistence)

Learn more


5. Sentiment Analysis

What it does: Detects urgency, frustration, or critical tone
Flags applied: needs-attention for negative sentiment

Example:

"This is EXTREMELY frustrating! Lost all my work!!!"
→ Sentiment: Negative
→ Flag: needs-attention
→ Priority: Elevated

Learn more


6. Missing Information Detection

What it does: Identifies gaps in issue description
Prompts for:

  • Screenshots
  • Reproduction steps
  • Error messages/logs
  • Environment details (browser, OS, version)

Example:

"App crashes when I click submit"
→ Missing: error logs, browser version, repro steps

Learn more


7. Epic Detection

What it does: Identifies large initiatives that should be broken down
Suggests: Sub-issues for better tracking

Example:

"Rebuild entire authentication system with OAuth2, SSO, MFA"
→ Type: epic
→ Suggested sub-issues:
  - Implement OAuth2 provider integration
  - Add SSO support for enterprise
  - Build MFA enrollment flow
  - Update security documentation

Learn more


8. Dependency Detection

What it does: Identifies issue dependencies and blockers
Detects:

  • Blocks: Issues that must be completed first
  • Blocked by: Issues preventing progress

Example:

"Add user profile page (requires #45 user API endpoint)"
→ Blocked by: #45

Learn more


9. Historical Pattern Matching

What it does: Links to similar past issues
Benefits:

  • Reuse solutions from previous work
  • Avoid duplicate effort
  • Learn from past mistakes

Example:

"Database migration failing on startup"
→ Similar to: #67, #82, #103
→ Solution found in: #103

Learn more


10. Auto-Assignment Suggestions

What it does: Recommends team members based on expertise
Considers:

  • Issue type and labels
  • Team member specializations
  • Workload distribution

Example:

"PostgreSQL query timeout on reports page"
→ Suggested assignee: @database-team

Learn more


11. Spanish Translation

What it does: Provides Spanish translation of title and description
Use case: International teams, accessibility

Example:

Original: "Login button not responsive on mobile"
Spanish: "Botón de inicio de sesión no responde en móvil"

Learn more


12. Merge Proposal (for duplicates)

What it does: When duplicates found, suggests consolidating discussions

Example:

Issue #50 duplicates #30 (has 12 comments)
→ Proposal: Close #50, move discussion to #30
→ Action: Auto-comment with link to original

Performance

  • 3-5 seconds average analysis time
  • 4 parallel LLM calls for speed
  • 99%+ success rate on valid issues

How It All Works Together

New Issue Created
     ↓
[Fetch All Issues Context]
     ↓
┌────────────────────────────────┐
│  4 Parallel LLM Analyses:      │
│  1. Duplicate Detection        │
│  2. Classification             │
│  3. Context Analysis           │
│  4. Spanish Translation        │
└────────────────────────────────┘
     ↓
[Apply Labels & Priority]
     ↓
[Post Comprehensive Comment]
     ↓
[Close if Duplicate]

Next Steps

Clone this wiki locally