triage-findings

Classify code review findings in a browser and update report.json.

1|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/lklimek/claudius --skill triage-findings
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: triage-findings
Source: https://github.com/lklimek/claudius/tree/main/skills/triage-findings
Command: npx skills add https://github.com/lklimek/claudius --skill triage-findings

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3-jsonschema, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the code review process by allowing interactive classification and management of findings, ensuring that only relevant issues are addressed.

Core Features & Use Cases

  • Interactive Triage UI: Classify findings (Fix, Accept Risk, Defer, False Positive) in a browser.
  • Automated Report Updates: Decisions are written back to the report.json file.
  • Targeted Fixes: Facilitates applying fixes, adding TODO comments, or marking risks as intentional.
  • Use Case: After an automated code review generates a report.json, use this Skill to quickly go through each finding, decide on the appropriate action, and have the system update the code or comments accordingly.

Quick Start

Use the triage-findings skill to interactively triage the findings in the file 'report.json'.

Frequently Asked Questions about triage-findings

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

FAQPage Schema
How do I interactively triage code review findings from a JSON report?

To triage code review findings interactively, you provide a JSON report file to the Skill, which validates it with Python scripts and serves a browser-based UI for classifying each finding as Fix, Accept Risk, Defer, or False Positive.

What is the best way to classify automated code review findings before applying fixes?

The best way to classify automated code review findings is using an interactive triage UI that lets you review each issue and write decisions directly back to the report.json file, ensuring only relevant issues are addressed.

Do I need Python and jsonschema to run an interactive code review triage interface?

Yes, you need Python along with the python3-jsonschema dependency to validate the JSON report file and serve the browser-based triage interface for managing your code review findings.

Can I defer false positives and automatically update the report file during code review triage?

Yes, you can defer false positives during code review triage, and the system automatically updates the report.json file with your decisions to facilitate targeted fixes or mark risks as intentional.

What actions can I take when triaging findings in a code review report?

When triaging findings in a code review report, you can classify each issue as Fix, Accept Risk, Defer, or False Positive, enabling the system to apply fixes, add TODO comments, or document intentional risks.

Why does triage of code review findings require manual classification before code modification?

Triage of code review findings requires manual classification before code modification to filter out false positives and ensure developers only spend time applying fixes or adding TODO comments to relevant, actionable issues.