skill-improver

Guide structured reflection on skill execution to identify and codify concrete improvements.

52|7|Updated Jul 5, 2025
One-click install
npx skills add https://github.com/bfollington/terma --skill skill-improver
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-improver
Source: https://github.com/bfollington/terma/tree/main/skills/skill-improver
Command: npx skills add https://github.com/bfollington/terma --skill skill-improver

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Users often encounter friction, confusion, or repeated failures when using AI, or discover better ways to do things that aren't captured. This skill provides a structured framework for reflection and improvement, preventing bloat and ensuring AI workflows become smoother and more efficient over time.

Core Features & Use Cases

  • Structured Reflection: Guides you through a systematic analysis of past AI interactions to pinpoint issues.
  • Pattern Identification: Helps recognize recurring problems and suggests proven solutions from a catalog of common issues.
  • Actionable Improvements: Formulates concrete, specific proposals for enhancing existing skills or creating new ones.
  • Use Case: After a complex task where Claude struggled with file permissions, use this skill to identify the root cause (missing script, incomplete error handling) and propose creating a reusable script and updating documentation, making future tasks effortless.

Quick Start

Reflect on the recent interaction where I struggled to extract data from the PDF. Help me identify what went wrong and how to improve the process.

Frequently Asked Questions about skill-improver

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

FAQPage Schema
How do I improve my AI workflow after encountering friction or repeated errors?

Structured workflow improvement involves analyzing past AI interactions at natural checkpoints—after complex tasks, session end, or recurring issues—to identify root causes and extract actionable patterns. This skill guides systematic reflection through a framework that documents the skill/process, issue observed, root cause, proposed change, and implementation steps to prevent future friction.

What's the best way to identify patterns in AI task failures?

Pattern identification extracts recurring problems from past interactions by examining documented issues against a catalog of common patterns. This skill applies a reference framework to recognize root causes—such as missing scripts or incomplete error handling—and suggests proven solutions from established improvement patterns.

How do I turn observed problems into concrete improvements for my AI processes?

Convert observations into improvements by formulating specific, actionable proposals that address identified root causes. This skill structures the process with defined outputs: Skill/Process name, Issue Observed, Root Cause analysis, Proposed Change, Impact assessment, and Implementation guidance tied to documented reference materials.

When should I apply workflow optimization to my AI tasks?

Apply workflow optimization at natural checkpoints: immediately after completing complex tasks where friction occurred, at the end of work sessions to capture learnings, or when the same problem recurs. This timing captures insights while context is fresh and prevents issues from compounding across future interactions.

Can I use this reflection framework to prevent future AI task failures?

Yes. Structured reflection codifies lessons learned into reusable improvements—new skills, updated documentation, or modified processes—that make future tasks smoother. By documenting root causes and solutions, the framework transforms one-off friction into systematic prevention.

What information do I need to document when reflecting on a failed AI task?

Document: the skill or process used, what issue occurred, the root cause analysis, your proposed change, expected impact, and implementation steps. This structured output ensures improvements are concrete, traceable, and actionable rather than vague observations.