skills.rest
mshaheerz
mshaheerzCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

skill-improver

Analyze execution logs and artifacts to identify skill failure modes and improvements.

1|Updated Feb 25, 2026
Search Tags:#debugging#AI agent#retrospective#workflow analysis#performance tuning#skill improvement
One-click install
npx skills add https://github.com/mshaheerz/hive-mind --skill skill-improver-mshaheerz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-improver
Source: https://github.com/mshaheerz/hive-mind/tree/main/agent-skills/skill-improver
Command: npx skills add https://github.com/mshaheerz/hive-mind --skill skill-improver-mshaheerz

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill addresses the challenge of improving the reliability and efficiency of AI skills by analyzing past execution data, identifying failure points, and suggesting concrete improvements.

Core Features & Use Cases

  • Failure Analysis: Pinpoints errors, bottlenecks, and manual intervention points in skill execution logs.
  • Contract Validation: Checks for consistent artifact handling and path usage.
  • Improvement Suggestions: Generates actionable recommendations for prompt optimization, I/O contract refinement, and error handling.
  • Use Case: After a multi-agent workflow fails, use this skill to analyze the runs/<workflow>/active/<run_id>/ directory to understand why it failed and get specific suggestions on how to fix the problematic skill's prompt or logic.

Quick Start

Analyze the run artifacts in the directory 'runs/my-workflow/active/run-123/' to identify areas for skill improvement.

Frequently Asked Questions about skill-improver

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

FAQPage Schema
How do I analyze AI agent workflow failures from execution logs?▼

Analyze AI agent workflow failures by processing run artifacts and execution logs to pinpoint errors, bottlenecks, and manual intervention points. This skill examines past execution data to identify failure modes and validate I/O contracts.

What is the best way to debug inconsistent I/O contracts in AI agent skills?▼

Debug inconsistent I/O contracts by checking execution artifacts for consistent path usage and artifact handling. The analysis validates input and output contracts against execution logs, highlighting mismatches and suggesting concrete edits for skill stability.

Can I use run artifacts to improve prompt clarity and skill performance tuning?▼

Yes, you can use run artifacts to improve prompt clarity and skill performance tuning. By analyzing previous run logs, the skill identifies unclear prompts and generates actionable recommendations for prompt optimization and enhanced error handling.

How do I perform a retrospective on a failed multi-agent workflow run?▼

Perform a retrospective on a failed multi-agent workflow by analyzing the active run directory containing execution logs and artifacts. This identifies why the workflow failed and provides specific suggestions to fix problematic skill logic.

Why does my AI agent skill keep failing at manual intervention points?▼

Your AI agent skill fails at manual intervention points due to identified bottlenecks or prompt clarity issues within the execution logic. Analyzing run artifacts reveals these failure modes and suggests concrete edits to enhance effectiveness and reduce manual overrides.

Related Skills

Explore More Software Engineering Skills→
intel-agencyintel-agency

forensic-analysis-report

Produce structured forensic markdown reports from GitHub Actions workflow evidence.

Official
Advanced
dzackgarzadzackgarza
1

python

Optimize Python code using performance patterns for I/O, memory, and concurrency.

Community
Advanced
m2ai-portfoliom2ai-portfolio
4

dynamic-workflow-orchestration

Decompose large tasks into coordinated multi-agent batch workflows with parallel analysis.

Official
Advanced
B4sanB4san

test-generator

Generate unit, integration, and end-to-end test suites from requirements and architectural designs.

Community
Intermediate