ai-bug-triage

Classify CI failures and draft tickets using fingerprinting and LLM analysis.

64|12|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/petrkindlmann/qa-skills --skill ai-bug-triage
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-bug-triage
Source: https://github.com/petrkindlmann/qa-skills/tree/main/skills/ai-bug-triage
Command: npx skills add https://github.com/petrkindlmann/qa-skills --skill ai-bug-triage

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Hybrid fingerprint + LLM-based pipeline for bug classification, deduplication, and ticket generation. Normalizes CI logs, creates stable fingerprints, clusters near-duplicates, then uses LLM for severity classification and ticket writing. Includes bug reporting templates and severity/priority matrix. Use when: "bug triage," "classify bugs," "failure analysis," "auto-classify," "CI failures," "bug report," "defect template."

Core Features & Use Cases

  • Deterministic fingerprinting for deduplication and clustering of similar failures
  • LLM-driven classification to assign severity, component, root cause hypothesis, and routing
  • Automated ticket drafting with structured description, evidence, and suggested assignees
  • Templates and guidance to ensure consistent triage and handoff

Quick Start

Run triage on recent CI logs to generate tickets for the latest P0 and P1 failures.

Frequently Asked Questions about ai-bug-triage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate bug triage and deduplication for CI failures and production logs?

Automated bug triage normalizes CI logs and stack traces into stable fingerprints, clusters near-duplicate failures, and uses LLM classification to assign severity, hypothesize root causes, and draft structured routing tickets for faster issue resolution.

What is deterministic fingerprinting for bug classification and how does it work?

Deterministic fingerprinting creates stable identifiers from normalized logs and stack traces to cluster near-duplicate bugs, ensuring consistent deduplication before an LLM assigns severity, component ownership, and root cause hypotheses for routing.

Do I need component ownership maps and ticketing system access to auto-classify bugs?

Yes, automated bug triage requires architecture context like component ownership maps and ticketing system access to accurately route classified issues, assign severity, and generate structured tickets with suggested assignees.

What's the best way to generate structured bug reports from test flakiness and stack traces?

Generating structured bug reports from test flakiness uses a pipeline that fingerprints stack traces, deduplicates failures, and leverages LLM classification to output category, severity, root cause hypotheses, and suggested ticket details.

Can I classify severity and priority for production bugs using LLM-based triage?

Yes, LLM-based triage classifies production bugs by assigning severity and priority using a defined matrix, combining deterministic deduplication with LLM analysis to output severity categories and route tickets to correct component owners.