harvest-all

Scan a codebase and convert undocumented knowledge into maintainable documentation.

1|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/hirokimry/vibecorp --skill harvest-all
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harvest-all
Source: https://github.com/hirokimry/vibecorp/tree/main/skills/harvest-all
Command: npx skills add https://github.com/hirokimry/vibecorp --skill harvest-all

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill finds undocumented implementation details, recurring conventions, and hidden edge cases across a repository so they can be captured before they are lost.

Core Features & Use Cases

It scans the codebase, compares findings against existing docs, and classifies new knowledge into architecture notes, coding rules, or operational cautions. It is designed for initial repository onboarding, periodic technical audits, and repository-wide cleanup of tribal knowledge into docs, rules, and knowledge files. It includes user approval gates and fail-secure protections so only explicitly approved findings are written back.

Quick Start

Use /vibecorp:harvest-all to scan the repository, review the discovered knowledge items, and apply only the updates you approve.

Frequently Asked Questions about harvest-all

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

FAQPage Schema
How do I extract undocumented knowledge from a legacy codebase?

To extract undocumented knowledge from a legacy codebase, scan the repository to uncover hidden implementation details, recurring conventions, and edge cases, then classify findings into architecture notes, coding rules, or operational cautions.

What is the best way to audit tribal knowledge in a repository?

Auditing tribal knowledge involves scanning the codebase to find undocumented edge cases and conventions, comparing findings against existing docs, and converting approved items into maintainable documentation files and coding rules.

How do I generate onboarding documentation from existing source code?

To generate onboarding documentation from existing source code, run a repository-wide scan to identify implementation details, classify the discovered knowledge, and apply user-approved updates directly to docs and knowledge files.

Can I automatically write discovered coding rules to my documentation files?

You cannot automatically write discovered coding rules without explicit consent; the process uses user approval gates and fail-secure protections so only manually approved findings are written back to documentation.

Does the codebase audit process detect duplicate knowledge entries?

Yes, the codebase audit process includes duplicate detection during scoped traversal, ensuring that newly discovered knowledge is compared against existing docs to prevent redundant entries before applying updates.