Can it identify technical debt in my repository?

Yes, by clustering repeated struggles by file or folder, it identifies 'hotspots' where repository fragility is causing agent errors.

How does it improve my prompt engineering?

The skill scores opening requests on clarity, boundedness, and architectural specificity, highlighting missing ingredients that lead to rework.

What kind of sessions does this skill analyze?

It specifically analyzes AI-assisted coding sessions stored in the Antigravity brain directory, reviewing tasks, plans, and walkthroughs.

What is the Severity Score?

It is a 0–100 metric that prioritizes sessions needing attention based on completion failure, replanning intensity, and scope instability.

Project Root Cause Analyst

Name: Project Root Cause Analyst
Author: JantonioFC

byJantonioFC

•

Analytics & Monitoring

Conducts forensic analysis of AI coding sessions to identify root causes of friction and improve project health.

This skill performs a deep dive into session histories to uncover the underlying reasons for rework, scope creep, and plan failures. It evaluates prompt sufficiency, classifies scope changes (human vs. agent vs. necessary), and identifies specific file or subsystem hotspots where technical debt or complexity is slowing down development. By generating detailed reports and severity scores, it helps developers optimize their AI interactions and prioritize architectural improvements based on empirical session data.

Key Features

01Comprehensive session analysis reports with actionable triage recommendations

02Automated root cause classification including spec ambiguity and repo fragility

032 GitHub stars

04Subsystem friction mapping to identify problematic code areas

05Forensic session reconstruction from conversation artifacts and metadata

06Prompt sufficiency scoring to improve the quality of initial AI requests

Use Cases

01Identifying brittle codebase areas that frequently cause AI logic errors

02Auditing completion rates and scope stability in AI-assisted development workflows

03Optimizing prompt engineering strategies based on historical failure patterns

Key Features

01Comprehensive session analysis reports with actionable triage recommendations

02Automated root cause classification including spec ambiguity and repo fragility

032 GitHub stars

04Subsystem friction mapping to identify problematic code areas

05Forensic session reconstruction from conversation artifacts and metadata

06Prompt sufficiency scoring to improve the quality of initial AI requests

Use Cases

01Identifying brittle codebase areas that frequently cause AI logic errors

02Auditing completion rates and scope stability in AI-assisted development workflows

03Optimizing prompt engineering strategies based on historical failure patterns