code-tour

Generates CodeStar tour files with verified file and line anchors.

2|Updated May 11, 2026

Search Tags:#onboarding #root cause analysis #pr review #codetour #codebase walkthrough #architecture tour

Ready to run

npx skills add https://github.com/himanshu231204/AI_Research_agent --skill code-tour-himanshu231204

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: code-tour
Source: https://github.com/himanshu231204/AI_Research_agent/tree/main/.opencode/skills/code-tour
Command: npx skills add https://github.com/himanshu231204/AI_Research_agent --skill code-tour-himanshu231204

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the inefficiency of one-off, non-reusable code explanations that waste engineering time during new hire onboarding, pull request reviews, and root cause analysis of production issues.

Core Features & Use Cases

Standardized Tour Generation: Creates valid CodeTour tour files with verified file, line, and directory anchors for interactive, in-IDE walkthroughs.
Persona-Adapted Content: Automatically calibrates tour depth and focus for specific readers including new joiners, PR reviewers, architects, and RCA investigators.
Use Case Example: Generate a 12-step architecture tour for this AI Research Agent codebase that walks a new backend engineer through the LangGraph orchestration flow, Celery worker pools, and API gateway in order.

Quick Start

Use the code-tour skill to generate a 10-step onboarding tour for new backend engineers covering the core request path of this AI Research Agent repository.

Frequently Asked Questions about code-tour

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

FAQPage Schema

How do I create a codebase walkthrough for new hire onboarding?▼

You can create a codebase walkthrough by generating standardized CodeTour files with verified file and line anchors. This replaces ad-hoc explanations by producing interactive, in-IDE tours calibrated specifically for new joiners.

What is the best way to document root cause analysis for production failures?▼

Documenting root cause analysis is best done by generating CodeTour files that anchor directly to the relevant code paths and line numbers. This creates reusable, verifiable walkthroughs for RCA investigators instead of disposable notes.

Can I generate a pull request review walkthrough that points to specific code lines?▼

Yes, you can generate a pull request review walkthrough using CodeTour files. The skill validates all referenced file paths and line numbers against the target repository to ensure PR reviewers see accurate, interactive anchors.

How do I build an architecture tour for a specific code module?▼

You can build an architecture tour by requesting a persona-adapted CodeTour for architects. The skill generates a multi-step walkthrough that deep dives into specific code modules while calibrating the depth for the intended reader.

Does the generated code tour validate file paths against the target repository?▼

Yes, generated code tours validate all referenced file paths, directory anchors, and line numbers against the target repository. This strict adherence to the CodeTour JSON schema ensures interactive walkthroughs do not contain broken links.

When should I not use ad-hoc codebase explanations instead of CodeTour files?▼

You should avoid ad-hoc codebase explanations during onboarding, PR reviews, and RCA when you need reusable documentation. CodeTour files replace one-off explanations with standardized, persona-calibrated walkthroughs that save engineering time.