skill-harvester

Review live Codex threads to identify skill gaps and recommend updates.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/lc0rp/luke-agent-scripts --skill skill-harvester-lc0rp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-harvester
Source: https://github.com/lc0rp/luke-agent-scripts/tree/main/skills/skill-harvester
Command: npx skills add https://github.com/lc0rp/luke-agent-scripts --skill skill-harvester-lc0rp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Harvester identifies gaps and opportunities in the current Codex session, surfacing reusable skill work and guiding decisions to create new skills or update existing ones.

Core Features & Use Cases

  • Review the live thread to detect repeated workflows, failing guardrails, or gaps in skill coverage.
  • Compare observed workflows against existing skills in the target root and recommend update or new skill; include a compact audit plan.
  • Transition recommendations into the skill-creator workflow to implement and validate chosen work.
  • Use cases include transforming session learnings into new skills or refining existing ones based on live-thread evidence.

Quick Start

Run the Harvester on the current session to generate a prioritized plan for new or updated skills and hand off to skill-creator for implementation.

Frequently Asked Questions about skill-harvester

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

FAQPage Schema
How do I extract reusable workflow learnings from a live session thread?

To harvest reusable learnings from a live session thread, review the active workflow to detect repeated patterns, failing guardrails, or coverage gaps, then generate a prioritized audit plan to create or update skills.

When should I update existing skills instead of creating new ones from session governance audits?

Update existing skills when the session reveals gaps or failing guardrails in current coverage; create new skills when the live thread exposes entirely new workflows or learnings not addressed by the current skill set.

What is the best way to identify skill coverage gaps during a live thread review?

The best way to identify skill coverage gaps is comparing observed live-thread workflows against existing skills in the target root, generating a compact audit plan that recommends targeted updates or new skill creation.

How do I transition harvested session learnings into a validated skill-creator workflow?

Transition harvested learnings by generating a prioritized plan from the session audit and handing it off to the skill-creator workflow for implementation and validation steps, including quick_validate.

Do I need existing skills in the target root before running a live session audit?

No, existing skills are not strictly required; the harvester compares live-thread workflows against the current target root to recommend updates to current skills or craft new ones for uncovered learnings.