learn-add

Route and log skill-context learning feedback into categorized LEARN.md files.

25|1|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/parhumm/jaan-to --skill learn-add
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learn-add
Source: https://github.com/parhumm/jaan-to/tree/main/adapters/codex/skillpack/skills/learn-add
Command: npx skills add https://github.com/parhumm/jaan-to --skill learn-add

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Add and route learning feedback into LEARN.md files to preserve and organize workflow lessons across skills and contexts.

Core Features & Use Cases

  • Route feedback to the appropriate LEARN.md file (skill-specific, template, or context).
  • Create or update LEARN.md with categorized lessons under Better Questions, Edge Cases, Workflow, and Common Mistakes.
  • Enforce a two-phase review workflow (hard stop for human confirmation) and optionally commit changes for traceability.

Quick Start

Provide a target skill or context and the lesson text, and the system will prompt for confirmation before updating LEARN.md.

Frequently Asked Questions about learn-add

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

FAQPage Schema
How do I log workflow lessons and feedback into a markdown file?

To log workflow lessons, you route feedback into a LEARN.md file by defining a target skill or context and providing the lesson text. The system validates inputs and categorizes them under sections like Workflow, Edge Cases, and Common Mistakes.

Can I track learning feedback changes with version control?

Yes, you can optionally commit LEARN.md updates for traceability. After the two-phase review workflow prompts for human confirmation, the system can commit the categorized lessons to maintain a version-controlled history of feedback.

How does the two-phase review workflow handle lesson updates?

The two-phase review workflow enforces a hard stop for human confirmation before updating LEARN.md. After you provide the target context and lesson text, the system validates the inputs and pauses to require your explicit approval before writing or committing changes.

What is the best way to preserve edge cases and common mistakes across skill contexts?

Preserving edge cases and common mistakes is done by routing feedback into context-specific LEARN.md files. You provide the lesson text and target context, and the system categorizes and logs the entry to ensure actionable knowledge is captured for future reference.