Pantheon Arbiter (Themis)

Classify code reviewer findings by severity and deduplicate issues.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/jschulte/pantheon --skill pantheon-arbiter-themis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Pantheon Arbiter (Themis)
Source: https://github.com/jschulte/pantheon/tree/main/src/skills/pantheon-arbiter
Command: npx skills add https://github.com/jschulte/pantheon --skill pantheon-arbiter-themis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the critical step of triaging findings from code reviewers, ensuring that only actionable issues are prioritized and duplicates are removed, thereby streamlining the bug-fixing process.

Core Features & Use Cases

  • Automated Triage: Classifies findings into MUST_FIX, SHOULD_FIX, STYLE, or DISMISSED based on predefined rules.
  • Quick Fix Identification: Automatically flags issues that can be resolved in under 2 minutes as MUST_FIX.
  • Severity Mapping: Translates findings from security gates and code health checks into a consistent severity framework.
  • Deduplication: Identifies and marks duplicate findings to reduce redundant work.
  • Use Case: After a code review cycle, this Skill processes all reviewer comments, prioritizes critical bugs and quick fixes, and dismisses stylistic suggestions, presenting a clean, actionable list for the developer.

Quick Start

Use the Pantheon Arbiter skill to triage the sprint artifacts for story key '1-3'.

Frequently Asked Questions about Pantheon Arbiter (Themis)

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

FAQPage Schema
How do I automate triaging code review findings to prioritize bug fixing?

Automating code review triage involves classifying findings by severity, applying quick fix rules, and deduplicating issues. This process categorizes security gate and code health comments into actionable levels like MUST_FIX or DISMISSED to streamline bug fixing.

What is the best way to classify code health and security gate findings?

Classifying code health and security gate findings requires mapping them to a pragmatic severity scale. Findings are categorized as MUST_FIX, SHOULD_FIX, STYLE, or DISMISSED based on predefined rules, ensuring only actionable issues are prioritized.

Can I automatically flag quick fixes in my code review comments?

Yes, you can automatically flag quick fixes in code review comments by evaluating them against predefined rules. Issues that can be resolved in under two minutes are automatically flagged for immediate attention to accelerate developer workflows.

How do I deduplicate reviewer findings to reduce redundant work?

Deduplicating reviewer findings is achieved by processing code review comments to identify overlapping issues. Duplicate findings are marked automatically during triage, reducing redundant work and presenting developers with a clean, actionable list.

Do I need a specific story key to process sprint artifacts for quality assurance?

Yes, processing sprint artifacts for quality assurance requires a specific story key and the path to the sprint artifacts. These inputs are necessary to correctly map and triage the code reviewer findings for your current development cycle.