distilling-prs

Name: distilling-prs
Availability: InStock
Author: axiomantic

Distills pull requests into categorized review signals for human attention.

8|5|Updated Dec 30, 2025

Search Tags:#heuristics #analysis #ai #review #diff #pr #pull-requests

Ready to run

npx skills add https://github.com/axiomantic/spellbook --skill distilling-prs

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: distilling-prs
Source: https://github.com/axiomantic/spellbook/tree/main/skills/distilling-prs
Command: npx skills add https://github.com/axiomantic/spellbook --skill distilling-prs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Distills pull requests into categorized review signals to identify changes that require human attention.

Core Features & Use Cases

Deterministic heuristic matching to surface obvious changes early.
AI-assisted analysis for unmatched changes to categorize risk and effort.
Structured reports with contextual diffs to preserve review context.

Quick Start

Invoke the skill with /distilling-prs <pr> to fetch PR data, analyze changes, and generate a categorized review report.

Frequently Asked Questions about distilling-prs

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

FAQPage Schema

How do I review large pull request diffs more efficiently?▼

You can distill pull requests into categorized review signals to surface changes requiring human attention. This two-phase analysis applies deterministic heuristics and AI-assisted categorization to large diffs, generating structured reports with contextual changes.

What is the best way to identify high-risk changes in multi-file pull requests?▼

Identifying high-risk changes in multi-file pull requests is done by distilling PRs into categorized review signals. The analysis uses pattern matching and AI-assisted evaluation to categorize risk and effort for unmatched changes.

How do I generate a human-readable code review report from a PR?▼

To generate a human-readable code review report, invoke the skill with the PR reference to fetch data, analyze changes, and output a structured report. This report preserves review context by including contextual diffs alongside the categorized signals.

Can I use heuristic analysis for pull request reviews without AI?▼

Yes, the pull request analysis uses deterministic heuristic matching to surface obvious changes early in the first phase. AI-assisted analysis is then applied only to unmatched changes to complete the categorization.

Does distilling PRs work for code-review workflows with massive diffs?▼

Yes, distilling PRs is specifically designed for code-review workflows with large diffs and multi-file changes. It processes these complex PRs by applying pattern matching and generating contextual, human-friendly review reports.