skills.rest
abagames
abagamesCOMMUNITY
Explore all skills by this author →
Website𝕏@abagamesGitHub

refining-workflows-from-artifacts

Refine agent workflows by classifying issues and proposing evidence-based changes.

11|2|Updated May 7, 2026
Search Tags:#agent skill#workflow improvement#workflow refinement#problem classification#execution analysis
One-click install
npx skills add https://github.com/abagames/agentic-gamedev-skills --skill refining-workflows-from-artifacts
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: refining-workflows-from-artifacts
Source: https://github.com/abagames/agentic-gamedev-skills/tree/main/.agents/skills/refining-workflows-from-artifacts
Command: npx skills add https://github.com/abagames/agentic-gamedev-skills --skill refining-workflows-from-artifacts

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps refine reusable agent workflows based on actual execution results, reviews, failures, or simulations, improving reliability and efficiency.

Core Features & Use Cases

  • Evidence-Based Refinement: Analyze execution results to propose the smallest workflow changes.
  • Problem Classification: Classify issues as task, execution, workflow, evaluation, capability-boundary, output-format, or overconstraint problems.
  • Role Execution Model: Use subagents for classification, workflow diff editing, and deletion review.
  • Companion Skills: Integrates with skills like empirical-prompt-tuning and extracting-agent-skills.

Quick Start

Run the refining-workflows-from-artifacts skill to refine your agent workflow based on the results of your latest task execution.

Frequently Asked Questions about refining-workflows-from-artifacts

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

FAQPage Schema
How do I improve agent workflow reliability using execution results?▼

Agent workflow reliability is improved by analyzing execution artifacts to classify issues and propose the smallest evidence-based changes to the workflow.

What is the best way to classify agent workflow failures for refinement?▼

Workflow failures are classified into task, execution, evaluation, capability-boundary, output-format, or overconstraint problems to target specific workflow refinement actions.

How do I refine an agent workflow after a failed task execution?▼

You refine the workflow by running the skill with your execution artifacts, which triggers subagents to classify the failure, edit the workflow diff, and review deletions.

Can I use execution reviews to make small changes to my agent workflows?▼

Yes, execution reviews and simulation results are used to propose the smallest possible evidence-based workflow changes, ensuring adjustments are targeted and reliable.

Do I need subagents to classify workflow problems and edit agent workflows?▼

The skill uses a role execution model with subagents to handle problem classification, workflow diff editing, and deletion review, automating the refinement process.

What are the limitations of evidence-based workflow refinement for agents?▼

Refinement focuses on small, evidence-based changes derived from execution artifacts; it requires classified problem data and integrates with companion skills rather than acting as a standalone solution.

Related Skills

Explore More Software Engineering Skills→
catlog22catlog22
511

maestro-coordinate

Coordinate multi-agent pipelines that map intents to ordered skill chains.

Community
Advanced
ProyectoG007ProyectoG007

openclaw-agent-optimize

Audit OpenClaw agent configurations for cost, routing, and context management.

Community
Advanced
mizchimizchi
320

waxa-eval

Coordinate empirical evaluation loops for skill prompts using the waxa CLI.

Community
Advanced
saschadudkin-sketchsaschadudkin-sketch

code-review

Automate code review for rezidence4/DomHub repository changes.

Community
Advanced