Coordinate auditable code review workflows across multiple LLMs or agents.
npx skills add https://github.com/vikingmute/review-forge --skill review-forge
Please help me install this Agent Skill. Skill: review-forge Source: https://github.com/vikingmute/review-forge/tree/main Command: npx skills add https://github.com/vikingmute/review-forge --skill review-forge
Orchestrates end-to-end, auditable code review workflows across multiple LLMs or agents, turning scattered feedback into a single coherent process.
Invoke the skill to orchestrate a multi-model review workflow on the current feature branch.
High-intent search queries and answers about installing and using this skill.
Multi-agent code review works by applying multiple LLMs to your feature branches and commits, then synthesizing their scattered feedback into a single unified checklist to apply fixes and track status.
You synthesize multi-agent feedback by orchestrating the review workflow across LLMs, which automatically consolidates their varied perspectives into a single coherent checklist for applying fixes and running tests.
Yes, the multi-model code review workflow enforces strict artifact isolation under the code_review/ directory, keeping your product code clean while maintaining full traceability for audit purposes.
Yes, you can run tests automatically after applying fixes, as the workflow orchestrates the sequence from review synthesis to fix application and final verification to track overall status.
The best way to track status and traceability for automated code reviews is to orchestrate the process end-to-end, isolating artifacts in code_review/ while tracking synthesis, fixes, and verification across teams.
You do not need to provide optional scripts or references to use the workflow, as it supports optional inputs like scripts/, references/, and assets/ only as needed for specific review scenarios.
praveenpereraOrganize JJ-based version control workflows for stacked and independent PRs.
dr-codeRun pre-release verification checks and write a versioned release assessment report.
giljaeAutomate web app testing with Puppeteer or Selenium to identify bugs and verify fixes.
TeamSPWKOrchestrate 3-way jury evaluations for code review, plan review, and security diagnosis.
