retain-learning

Classify new learnings and store them in Hindsight's retain tool.

1|Updated Mar 3, 2026
One-click install
npx skills add https://github.com/AndreJorgeLopes/devflow --skill retain-learning
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: retain-learning
Source: https://github.com/AndreJorgeLopes/devflow/tree/main/skills/retain-learning
Command: npx skills add https://github.com/AndreJorgeLopes/devflow --skill retain-learning

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps prevent valuable decisions, discoveries, and hard-won lessons from being lost between AI agent sessions.

Core Features & Use Cases

  • Learning capture: Converts a user-provided note into a structured memory when you don’t already have a full write-up.
  • Automatic classification: Tags the learning as a Mental Model, Hard Rule, Gotcha, Decision, Technique, or Discovery for better retrieval.
  • Hindsight retention: Stores the structured payload into Hindsight’s long-term memory using the Hindsight retain tool.

Quick Start

Use retain-learning to store a specific lesson for future sessions by telling your agent what you learned and including any relevant context or rationale.

Frequently Asked Questions about retain-learning

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

FAQPage Schema
How do I retain learning insights for future AI agent sessions?

To retain learning insights for future AI agent sessions, you capture the discovery or decision rationale and store it as a structured memory payload using a long-term memory tool. This prevents valuable decisions from being lost between sessions.

What is the best way to classify agent memory for better retrieval?

The best way to classify agent memory for better retrieval is to automatically tag the learning payload as a Mental Model, Hard Rule, Gotcha, Decision, Technique, or Discovery. This categorization structures the data for efficient future recall.

Can I preserve decision rationale and hard rules between agent workflows?

Yes, you can preserve decision rationale and hard rules between agent workflows by converting user-provided notes into a structured memory payload. This ensures hard-won lessons and specific gotchas are stored for later reuse.

How do you structure a memory payload from a raw note?

To structure a memory payload from a raw note, you parse the input arguments, classify the specific learning into a defined category, and invoke a memory retention tool to store the structured payload for future recall.

When do I need to store agent memory in Hindsight?

You need to store agent memory in Hindsight when you have a discovery, hard rule, decision rationale, technique, mental model, or gotcha to preserve. This is required to prevent valuable insights from being lost between AI agent sessions.