githuman_skill

This skill helps you review AI-generated code changes before committing using GitHuman, enabling thorough inline comments, suggestions, and todo tracking.

5

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill mcollina/githuman-skills --skill githuman

SKILL.md1.3 KB

Overview

This skill helps you review AI-generated code changes with GitHuman before committing. It streamlines visual code review, inline commenting, suggestion management, and exporting review artifacts so you can confidently accept, adjust, or reject AI edits. Use it to catch mistakes, enforce style, and track follow-ups created by the AI agent.

How this skill works

The skill inspects staged diffs and files targeted by the AI agent, presenting them in a web interface with inline comments, suggestions, and todo items. It supports staging files for review, adding and resolving inline comments, converting comments to actionable todos, running predefined checks (linters/tests), and exporting the review for documentation or handoff. CLI commands let you open reviews, stage changes, and export results programmatically.

When to use it

Before committing any AI-generated changes to ensure correctness and style compliance
When creating a formal code review for AI-produced diffs or pull requests
To audit what an AI agent wrote to detect logic bugs, security issues, or regressions
When preparing a commit and you want an organized list of follow-up todos
Any time you mention “review”, “GitHuman”, or “before commit” in workflow automation

Best practices

Stage focused, small sets of files to keep reviews manageable and focused
Run linters and unit tests locally before or during review to surface obvious issues
Turn unclear comments into todos so they are tracked and assigned for follow-up
Use inline suggestions for minimal edits and comment threads for design concerns
Export reviews for documentation or asynchronous handoff to collaborators

Example use cases

Review an AI agent’s feature implementation before merging to main
Create a documented review with todos to hand off remediation to teammates
Audit AI-generated security-sensitive changes and block unsafe code paths
Stage only modified files for a quick sanity check before committing
Export a review to attach to a ticket or release notes for compliance

FAQ

No. It surfaces AI-generated diffs and lets you add inline suggestions or apply edits manually. You control which edits are committed.

Can I export reviews for sharing or compliance?

Yes. Reviews and their todo lists can be exported for documentation, audits, or asynchronous handoff.