When should I use the Error Recovery skill?

Use this skill when your error rate exceeds 5%, when errors are recurring, or when you need a systematic framework to reduce Mean Time To Resolve (MTTR).

What categories of errors does this skill cover?

It covers 13 major categories including file operations, API calls, data validation, concurrency, and resource management, providing over 95% coverage of common issues.

How does this skill improve Claude's performance?

It provides a 'meta-cognition' layer that allows Claude to analyze its own session history, identify bottlenecks, and apply specific recovery patterns rather than guessing at fixes.

Is this skill specific to a particular programming language?

No, the taxonomy and diagnostic workflows are 85-90% transferable across all major languages and platforms because error patterns are largely universal in software development.

Can this skill prevent errors before they happen?

Yes, it includes automation tools for file path, size, and state validation that can prevent approximately 23.7% of common operational errors before they occur.

Error Recovery & Workflow Optimization

Name: Error Recovery & Workflow Optimization
Author: yaleh

byyaleh

•

Analytics & Monitoring

Implements a systematic error-handling methodology using a 13-category taxonomy to diagnose, recover from, and prevent session failures.

This skill transforms error handling from an ad-hoc process into a data-driven engineering discipline by utilizing a taxonomy based on over 1,300 historical errors. It provides Claude with structured diagnostic workflows and recovery patterns—such as exponential backoff and circuit breakers—while introducing automation tools to prevent nearly 24% of common failures. Ideal for complex development sessions where error rates exceed 5% or recovery time is too high, it ensures your AI workflows remain resilient, efficient, and capable of self-correction across any programming language.

Key Features

0113-category error taxonomy covering 95.4% of common developer failures

0215 GitHub stars

03Eight specialized diagnostic workflows for file operations, API calls, and resources

04Performance tracking for MTTD (Mean Time To Detect) and error distribution

05Automated prevention tools for file path validation and read-before-write checks

06Five proven recovery patterns including exponential backoff and circuit breakers

Use Cases

01Building robust error-handling infrastructure for production-grade projects

02Reducing high error rates in complex automation and long-running sessions

03Performing root cause analysis on recurring or systematic workflow failures

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yaleh/meta-cc error-recovery

For use in Claude.ai and ChatGPT

Download Skill