How does the multi-agent review process work?

The skill spawns four specialized AI subagents in parallel: a Code-First Reviewer, a Testing Reviewer, a Skeptical Reviewer for edge cases, and a Big-Picture Reviewer to ensure the PR meets high-level goals.

What is the 'fail-first' testing requirement?

When fixing a bug, the skill requires writing a test that fails without the fix and passes with it, ensuring the test actually reproduces the specific issue and prevents future regressions.

Can I use the code-simplifier independently?

Yes, the skill encourages running the code-simplifier subagent before reviews to remove redundancy and verbose patterns, making the final code easier for humans and agents to audit.

Why does this skill mandate using Git worktrees?

Worktrees prevent developers from working in a stale main directory, ensuring every PR is based on the latest origin/main and avoiding common environment or merge conflict issues.

Freenet PR Quality Standards

Name: Freenet PR Quality Standards
Author: freenet

byfreenet

•

Collaboration Tools

Enforces rigorous quality standards and multi-agent review processes for Freenet pull requests.

The freenet-pr-creation skill provides a comprehensive framework for contributing high-quality code to the Freenet ecosystem. It guides developers through a disciplined workflow including mandatory Git worktree management, Conventional Commits, and a 'fail-first' regression testing philosophy. Its standout feature is the orchestration of four specialized AI subagents—Code-First, Testing, Skeptical, and Big-Picture—that perform parallel reviews to eliminate technical debt and ensure architectural integrity before any code is merged.

Key Features

01Integrated code-simplifier tool to clean and optimize logic before the final review phase.

02Automated Git worktree management to isolate feature branches and prevent environment pollution.

03Comprehensive testing requirements including unit, integration, and E2E network simulations.

04Orchestration of four parallel AI review subagents for multi-perspective code auditing.

051 GitHub stars

06Enforcement of Conventional Commits and structured PR descriptions focusing on the 'Why' over the 'What'.

Use Cases

01Automating multi-layered code reviews to identify race conditions and architectural misalignments.

02Standardizing contributions to freenet-core, freenet-stdlib, and distributed network repositories.

03Implementing rigorous regression testing for complex peer-to-peer network bug fixes.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add freenet/freenet-agent-skills pr-creation

For use in Claude.ai and ChatGPT

Download Skill