Can I focus the reflection on a specific area?

Yes. If you provide focus text when invoking the command, the skill will prioritize its analysis around your specific observations while still scanning for general friction points.

What is a 'trace-aware' reflection?

It means the reflection is grounded in the actual history of the session—analyzing exactly what commands were run, what errors occurred, and which patterns led to success.

How do I trigger the reflection process?

You can invoke the skill by typing '/compound:reflect' or asking Claude to 'reflect on this session', 'capture learnings', or 'debrief this work'.

Where are the captured learnings stored?

If the Linear MCP is available, it creates issues in your workspace. Otherwise, it generates local markdown files in a './compound-learnings/' directory.

What makes a learning 'frontier-shifting'?

A frontier-shifting learning improves the underlying capability so that both the volume and quality of work can improve simultaneously, rather than just trading one for the other.

Reflection Craft

Name: Reflection Craft
Author: mberto10

bymberto10

Productivity & Workflow

Transforms work sessions into structured, actionable learnings and testable rules to improve development workflows.

About

Reflection Craft is a meta-development skill designed to capture and codify knowledge from active coding sessions. By performing trace-aware analysis of successes and friction points, it distills complex developer experiences into one-line testable rules and actionable issues. Whether documenting workarounds or identifying missing capabilities, this skill ensures that every session contributes to a compounding library of organizational knowledge, preventing the repetition of past mistakes and systematically shifting your development frontier.

Key Features

Linear MCP integration for automated issue tracking and labeling
Automatic generation of local markdown learning artifacts
Distillation of friction into specific, one-line testable heuristics
Heuristic filtering to prioritize frontier-shifting improvements over temporary fixes
0 GitHub stars
Trace-aware session debriefing grounded in actual activity

Use Cases

Converting developer friction into structured feature requests or bug reports
Conducting post-session debriefs to capture architectural decisions and patterns
Creating a compounding library of 'rules of thumb' for specific codebases

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mberto10/mberto-compound reflection-craft

For use in Claude.ai and ChatGPT

Download Skill

GitHub