learn

Captures session learnings into persistent project memory via parallel sub-agents.

Updated Dec 17, 2025
One-click install
npx skills add https://github.com/eprouveze/HealthPulse --skill learn-eprouveze
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learn
Source: https://github.com/eprouveze/HealthPulse/tree/main/.agent/skills/learn
Command: npx skills add https://github.com/eprouveze/HealthPulse --skill learn-eprouveze

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Captures session learnings into persistent project memory to accelerate future work.

Core Features & Use Cases

  • Spawns parallel sub-agents to analyze context, extract solutions, find related docs, and classify learnings.
  • Stores solutions in docs/solutions/ for future reference.
  • Use after completing a feature, fixing a tricky bug, or when the user says "what did we learn", "capture this", or "remember this".
  • Supports multi-LLM mode via Gemini CLI to preserve token limits.

Quick Start

After finishing a feature or bug fix, say "what did we learn" to trigger the learn process.

Frequently Asked Questions about learn

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

FAQPage Schema
How do I capture session learnings into persistent project memory?

Trigger the learning capture process by saying "what did we learn" or "capture this" after completing a feature or fixing a bug. The system spawns parallel sub-agents to analyze context and store extracted solutions in docs/solutions/ for future reference.

When should I apply session analysis to extract project knowledge?

Apply session analysis after completing a feature, fixing a tricky bug, or when you need to classify learnings. It extracts solutions and finds related docs to accelerate future work by preserving project knowledge.

Can I use multi-LLM mode to preserve token limits during knowledge capture?

Yes, knowledge capture supports multi-LLM mode via Gemini CLI to preserve token limits. This allows parallel sub-agents to analyze context and extract solutions efficiently without exceeding token constraints.

What is the best way to store project solutions for future reference?

The best way to store project solutions is using automated knowledge capture. It classifies learnings and stores extracted solutions directly in the docs/solutions/ directory, creating a persistent project memory for future reference.

Does capturing project knowledge require any external dependencies?

No external dependencies are required for capturing project knowledge. The system operates independently to spawn sub-agents that analyze context, find related docs, and store solutions in your project directory.