learn

Organize and surface project learnings in learnings.jsonl with search and pruning.

Updated Apr 6, 2026
One-click install
npx skills add https://github.com/JeremiahEllington/NoahEllington --skill learn-jeremiahellington
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learn
Source: https://github.com/JeremiahEllington/NoahEllington/tree/main/.claude/skills/learn
Command: npx skills add https://github.com/JeremiahEllington/NoahEllington --skill learn-jeremiahellington

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Capture, organize, search, prune, and export learnings from project work to avoid repeating mistakes and preserve institutional knowledge.

Core Features & Use Cases

  • Centralized learnings store across projects and sessions.
  • Quick search and filtering by key, type, or text to surface relevant insights.
  • Prune stale or conflicting entries to keep knowledge current.
  • Export learnings to Markdown or CLAUDE.md for documentation and onboarding.
  • Proactively surface patterns and recommendations based on recent activity.

Quick Start

Use /learn to view recent learnings, or /learn search <query> to find specific insights.

Frequently Asked Questions about learn

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

FAQPage Schema
How do I capture and organize project learnings across multiple sessions?

To capture project learnings across sessions, you can append new entries to a centralized learnings.jsonl file. The skill automatically deduplicates and stores insights so institutional knowledge is preserved and searchable for future reviews.

What is the best way to search past project knowledge and filter by text or type?

Searching past project knowledge is done by querying the centralized learnings store with specific text, keys, or types. This surfaces relevant historical insights quickly without manually digging through past session logs.

Can I export project learnings to Markdown or CLAUDE.md for documentation?

You can export project learnings to Markdown or CLAUDE.md files. This creates formatted documentation outputs useful for team onboarding and sharing institutional knowledge across workflows.

How do I prune stale or conflicting entries from my learnings knowledge base?

Pruning stale entries from your learnings knowledge base involves removing outdated or conflicting items from the learnings.jsonl file. This keeps the stored project knowledge current and prevents repeating past mistakes.

Do I need a specific format to store project insights before searching or pruning?

You need to use the JSONL format, specifically a learnings.jsonl file, to store project insights. The skill reads and writes directly to this file, automatically deduplicating entries when appending or pruning knowledge.

Does this approach work for surfacing patterns during project retrospectives?

This approach works for retrospectives by proactively surfacing patterns and recommendations based on recent activity. It organizes session insights, enabling quick search and filtering during reviews to find relevant historical context.