error-diagnostics-smart-debug

Error Diagnostics Smart Debug - Intelligent Error Diagnosis and Debugging Assistant

Skill Overview

An AI-driven debugging expert that helps developers quickly analyze errors, identify root causes, and generate repair plans. It supports end-to-end debugging across all scenarios—from local development to production environments.

Use Cases

Production Incident Triage: When errors occur in production, data is collected through observability platforms (Sentry, DataDog, etc.). The AI analyzes error patterns, generates hypotheses, and recommends debugging strategies.

Stack Trace and Error Analysis: When faced with complex stack traces or error messages, the AI automatically detects error patterns, analyzes component dependencies, assesses severity, and provides possible root-cause hypotheses.

Intermittent and Performance Issues: For hard-to-reproduce intermittent errors or performance problems, advanced methods such as chaos engineering and statistical analysis are used together with AI to intelligently instrument and locate issues.

Core Features

AI-Driven Initial Analysis: Automatically parses error messages, stack traces, reproduction steps, and other information to generate 3–5 root-cause hypotheses ranked by probability, along with targeted debugging strategies.

Observability Data Collection and Analysis: Integrates mainstream APM, log aggregation, and distributed tracing platforms to query key metrics such as error frequency, affected users, and environment-specific patterns.

Intelligent Instrumentation and Safe Production Debugging: The AI recommends optimal breakpoint and logging locations, and supports production-safe techniques such as dynamic instrumentation, feature-flag-based debugging, and sampling analysis.

Frequently Asked Questions

When should this debugging skill be used?

Use it whenever you need to diagnose an error—whether it’s a bug in local development, anomalies in a testing environment, or incidents in production. It is especially suitable for scenarios such as complex stack trace analysis, intermittent error investigation, and performance issue troubleshooting.

Which observability platforms are supported?

It supports popular error tracking and monitoring platforms, including Sentry, Rollbar, Bugsnag (error tracking); DataDog, New Relic, Dynatrace (APM); Jaeger, Zipkin, Honeycomb (distributed tracing); ELK, Splunk, Loki (log aggregation), and more.

How do you ensure the safety of debugging in production?

It provides multiple production-safe debugging techniques: dynamic instrumentation (OpenTelemetry), conditional logs via feature flags, sampling analysis (minimal overhead), read-only debug endpoints (with authentication and rate limiting), and gradual traffic switching (verify with 10% traffic first). All debugging actions undergo risk assessment and come with rollback strategies.