root-cause

Name: root-cause
Author: Avtr99

Pending

Catch and fix AI-written patchwork code before it ships, in any language or stack.

0stars

0forks

Added 8/20/2026

Installation

# Add to your Claude Code skills
git clone https://github.com/Avtr99/root-cause

Getting Started

Guides for using ai agents skills like root-cause.

Caveman: Cut Claude Token Use by 65%
How agent-side prompt compression works, when to use it, and when not to.
What is an AI Skills Marketplace?
Definitions, how marketplaces work, and how to choose between them in 2026.
Getting Started with AI Skills

README.md

Frequently Asked Questions

What is root-cause?

root-cause is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Avtr99. Catch and fix AI-written patchwork code before it ships, in any language or stack. It has 0 GitHub stars.

Is root-cause safe to use?

root-cause's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.

How do I install root-cause?

Clone the repository with "git clone https://github.com/Avtr99/root-cause" and add it to your Claude Code skills directory (see the Installation section above).

Are there alternatives to root-cause?

Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh root-cause against similar tools.

Agentic AI for Beginners

Build your first AI agent from scratch - tool use, ReAct pattern, memory, deployment

41 minBeginner

Comments (0)

to leave a comment.

No comments yet. Be the first to share your thoughts!

Related Skills

ECC

by affaan-m

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

241,213

review-it scholar-sidekick-skills

⚠️ Third-Party Software Notice

This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.

Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.

Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.

Read the Terms of Service

root-cause

Catch and fix AI-written patchwork code before it ships, in any language or stack.

AI agents patch symptoms by default — a bug appears, they add a guard, the bug returns on the next code path that forgets the guard. root-cause steers the agent to the structural fix instead: a stricter type, a deleted code path, a shared implementation, occasionally a library. The bug becomes impossible, not just currently prevented.

The skill prompt and its rules live in skills/root-cause/SKILL.md. This README covers why you'd install it, the benchmark results, and how to install it.

Benchmark

The skill was tested with the skill-eval-harness across five models. Each test gives the model bugged code and asks it to fix the root cause; a script oracle compiles and runs the output to check that the fix is structural, not a patch. Four cases, 2 runs per arm. Pass rate = share of runs that passed the oracle.

Benchmark results: structural-fix pass rate by model, with vs without the root-cause skill

Model	With skill	Without skill	Improvement
Claude Sonnet 5	56%	0%	+56 points
GLM 5.2	56%	0%	+56 points
moonshotai/kimi-k3	50%	0%	+50 points
Claude Haiku 4.5	50%	31%	+19 points
tencent/hy3	44%	0%	+44 points

Four of the five models passed zero runs without the skill — they patch symptoms by default. With the skill, they find the structural fix instead. Full methodology and reproduction commands: docs/benchmark.md.

Install

This skill follows the open Agent Skills standard. It works in Claude Code, Codex, Cursor, Devin, Cline, GitHub Copilot, and other popular AI coding platforms.

Universal — 75+ agents (recommended)

npx skills add Avtr99/root-cause

This command detects every supported agent on your machine and installs the skill into each agent's native skills directory.

npx skills add Avtr99/root-cause --global          # all projects, not just this one
npx skills add Avtr99/root-cause --skill root-cause # install only this skill
npx skills add Avtr99/root-cause --list             # preview without installing

Claude Code

claude plugin marketplace add Avtr99/root-cause
claude plugin install root-cause@avtr-root-cause

Codex / ChatGPT

codex plugin marketplace add Avtr99/root-cause
codex plugin install root-cause@avtr-root-cause

Then install from the Plugins directory in ChatGPT Work or the Codex desktop app.

GitHub Copilot CLI

copilot plugin marketplace add Avtr99/root-cause
copilot plugin install root-cause@avtr-root-cause

Cursor

cursor plugin install Avtr99/root-cause

Or search for "root-cause" in the Cursor plugin marketplace.

Devin

devin plugins install Avtr99/root-cause

Verify with devin plugins list. Plugins load at session start, so start a new Devin session after installing.

Grok Build CLI

grok plugin marketplace add Avtr99/root-cause
grok plugin install root-cause@avtr-root-cause

Or install directly: grok plugin install Avtr99/root-cause.

Manual

Copy skills/root-cause/ into the skills directory of your agent:

Agent	Path
Claude Code	`.claude/skills/root-cause/`
Codex	`.agents/skills/root-cause/`
Cursor	`.cursor/skills/root-cause/`
Devin	`.devin/skills/root-cause/`
Generic / cross-agent	`.agents/skills/root-cause/`

Usage

The skill activates on its own when the agent writes or reviews code. You can also invoke it by name:

Slash command (Claude Code, Codex, Cursor, Devin): /root-cause
Natural language: "use root-cause", "audit for overcomplication", "find the root cause"

It has two modes — fix (one bug) and audit (a whole codebase) — described in SKILL.md.

What is inside

File	Purpose
`skills/root-cause/SKILL.md`	The skill — the prompt that the agent reads
`skills/root-cause/README.md`	Human-facing docs for the skill
`plugin.json`	Root plugin manifest (Devin, Antigravity, and other root-native agents)
`.claude-plugin/marketplace.json`	Claude Code plugin marketplace manifest
`.claude-plugin/plugin.json`	Claude Code plugin manifest
`.codex-plugin/plugin.json`	Codex / ChatGPT plugin manifest
`.agents/plugins/marketplace.json`	Codex / ChatGPT plugin marketplace manifest
`.github/plugin/marketplace.json`	GitHub Copilot CLI plugin marketplace manifest
`.cursor-plugin/plugin.json`	Cursor plugin manifest
`.devin-plugin/plugin.json`	Devin plugin manifest
`.grok-plugin/plugin.json`	Grok Build plugin manifest
`.grok-plugin/marketplace.json`	Grok Build plugin marketplace manifest
`docs/agent-portability.md`	Which adapter each host uses
`docs/benchmark.md`	Evaluation methodology and results

Compatibility

root-cause is a pure-prompt skill. It has no scripts, no network calls, and no dependencies. It works in any agent that reads the SKILL.md format. The adapter manifests in this repo (.claude-plugin/, .codex-plugin/, .cursor-plugin/, .devin-plugin/, .grok-plugin/, .github/plugin/) add native install support on top.

License

MIT — see LICENSE.