skills.rest
Mosaic-agent
Mosaic-agentOFFICIAL
Explore all skills by this author →
GitHub

caveman-review

Generate one-line code-review comments with line references from PR diffs.

36|7|Updated Feb 22, 2026
Search Tags:#automation#comments#diff#pr#code-review#review-automation
One-click install
npx skills add https://github.com/Mosaic-agent/Mosaic-fund-agent --skill caveman-review-mosaic-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: caveman-review
Source: https://github.com/Mosaic-agent/Mosaic-fund-agent/tree/main/.agents/skills/caveman-review
Command: npx skills add https://github.com/Mosaic-agent/Mosaic-fund-agent --skill caveman-review-mosaic-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reduces PR review noise by generating concise, one-line findings that pinpoint exact locations and fixes.

Core Features & Use Cases

  • One-line findings: Output concise, actionable comments in the format L<line>: <problem>. <fix>.
  • Preserves context: Keeps exact line numbers and code references, even in multi-file diffs.
  • Use Case: When reviewing a PR, you can automatically generate a tight set of review comments to speed up feedback.

Quick Start

Run caveman-review on the PR diff to generate the inline comments immediately.

Frequently Asked Questions about caveman-review

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

FAQPage Schema
How do I generate concise PR review comments automatically from a diff?▼

To generate concise PR review comments automatically, run the automation on the pull request diff. It outputs terse, one-line findings with precise line references in the format L<line>: <problem>. <fix>, reducing review noise instantly.

Can I automate code review for multi-file pull request changes?▼

Yes, you can automate code review for multi-file pull request changes. The automation processes multi-file diffs and preserves exact line numbers and code references across the entire pull request to output reviewer-ready findings.

What is the best way to reduce pull request review noise?▼

The best way to reduce pull request review noise is generating terse, one-line findings that pinpoint exact locations and fixes. This format enforces concise, actionable comments with precise line references instead of verbose explanations.

Does PR review automation preserve exact code identifiers and line numbers?▼

Yes, PR review automation preserves exact code identifiers and line numbers. Even when processing multi-file diffs, it maintains precise line references and keeps exact code identifiers to ensure reviewer-ready output.

What format do automated code review comments follow for pull requests?▼

Automated code review comments follow the format L<line>: <problem>. <fix> with optional severity. This enforces concise, one-line findings that pinpoint exact locations and actionable fixes for pull request review.

Related Skills

Explore More Software Engineering Skills→
dynatrace-ossdynatrace-oss
10

pr-reviewer

Review pull request diffs and comments to generate a structured review brief.

Official
Intermediate
hyperlane-xyzhyperlane-xyz
97

inline-pr-comments

Post line-specific inline comments to GitHub pull request diffs.

Official
Intermediate
inbharataiinbharatai
22

code-reviewer

Analyzes code for bugs, security vulnerabilities, performance bottlenecks, and style violations.

Community
Advanced
hamzaPixlhamzaPixl
2

caveman-review

Generate one-line code review comments with line numbers and severity markers.

Community
Basic