skill-create-repo

Extract coding conventions from git commit history into a SKILL.md draft.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/sujithatzackriya/skills --skill skill-create-repo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-create-repo
Source: https://github.com/sujithatzackriya/skills/tree/main/skills/skill-create-repo
Command: npx skills add https://github.com/sujithatzackriya/skills --skill skill-create-repo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Creates a Claude Code skill tailored to the conventions and workflows of an existing repository by mining its git history, so you don’t have to manually author everything from scratch.

Core Features & Use Cases

  • Git-history pattern extraction: Analyzes commits, file co-changes, workflow sequences, and architecture conventions.
  • Skill draft generation: Outputs a {repo-name}-patterns SKILL.md capturing detected conventions (and optional instincts for continuous-learning-v2).
  • Safety-first generation: Flags adversarial or injection-like content in commit messages/comments and requires user review before adoption.

Quick Start

Run the command /skill-create-repo to analyze your current repo’s recent commits and generate a {repo-name}-patterns skill draft for review.

Frequently Asked Questions about skill-create-repo

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a Claude Code skill from existing git history and repository conventions?

Git-history pattern extraction analyzes commit messages and file co-change sequences to identify recurring coding conventions and internal architecture workflows, producing a repository-specific skill draft for adoption.

Can I automatically extract testing patterns and commit conventions from a repository?

Yes, you can automatically extract testing patterns and commit conventions by parsing git commit logs and detecting recurring file co-change sequences across commits to capture the repository's established workflows.

What is the best way to learn internal architecture conventions from an existing codebase?

The best way to learn internal architecture conventions is to mine the git repository's commit history for file co-change sequences and workflow patterns, generating a structured skill draft that captures these conventions for reuse.

Does generating a skill from git history protect against malicious commit message injection?

Generating a skill from git history treats commit messages as untrusted input, applying injection-detection guardrails to flag adversarial content and requiring user review before adopting the generated SKILL.md draft.

What limitations exist when mining git history for continuous-learning instincts?

Limitations include treating git history as untrusted input, requiring manual user review before adopting the generated patterns, and needing sufficient commit history depth to accurately detect recurring file co-change sequences and architecture conventions.