CommunityCoding & Developmentgithub.com

chips-lxm/deep-project-review

Performs an independent, deep secondary review of an existing checkable version of a project, involving reviews by multiple real sub-agents, high-level Astra verification, followed by authorized repair scheduling and independent acceptance. This skill is only triggered by explicit user invocation or through a configured entry point that explicitly requests an in-depth independent re-review; it is not started by normal checks, project completion, discussions, or the creation of this skill.

What is deep-project-review?

This skill, chips-lxm/deep-project-review, provides a rigorous, independent secondary review process for projects that already have a checkable version. It orchestrates a multi-layered review involving several genuine sub-agents (which could be distinct AI instances or simulated agents) and a high-level verification step using 'Astra' for extremely thorough re-examination. The skill's workflow includes scheduling authorized fixes and performing an independent acceptance check after repairs, ensuring a high-quality outcome. Crucially, this skill is designed to be invoked only through explicit user commands or via pre-configured entry points that specifically request an in-depth independent re-review; it does not trigger automatically during routine tasks like normal project checks, completion events, discussions, or the process of creating the skill itself. This makes it a powerful, deliberate quality assurance tool for developers using AI agent platforms like Cursor or Codex.

Works with~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/chips-lxm/deep-project-review/tree/main

Installed? Explore more Coding & Development skills: steipete/bluebubbles, steipete/eightctl, steipete/blucli · View all 6 →

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

What does deep-project-review do?

Performs an independent, deep secondary review of an existing checkable version of a project, involving reviews by multiple real sub-agents, high-level Astra verification, followed by authorized repair scheduling and independent acceptance. This skill is only triggered by explicit user invocation or through a configured entry point that explicitly requests an in-depth independent re-review; it is not started by normal checks, project completion, discussions, or the creation of this skill.

Related Skills