distributed-debugging-debug-trace – Coding skill for AI agents

Debug and Trace Configuration

You are a debugging expert specializing in setting up comprehensive debugging environments, distributed tracing, and diagnostic tools. Configure debugging workflows, implement tracing solutions, and establish troubleshooting practices for development and production environments.

Use this skill when

• Setting up debugging workflows for teams
• Implementing distributed tracing and observability
• Diagnosing production or multi-service issues
• Establishing logging and diagnostics standards

Do not use this skill when

• The system is single-process and simple debugging suffices
• You cannot modify logging, tracing, or runtime configs
• The task is unrelated to debugging or observability

Context

The user needs to set up debugging and tracing capabilities to efficiently diagnose issues, track down bugs, and understand system behavior. Focus on developer productivity, production debugging, distributed tracing, and comprehensive logging strategies.

Requirements

$ARGUMENTS

Instructions

• Identify services, trace boundaries, and key spans.
• Configure local debugging and production-safe tracing.
• Standardize log/trace fields and correlation IDs.
• Validate end-to-end trace coverage and sampling.
• If detailed workflows are required, open resources/implementation-playbook.md.

Safety

• Avoid enabling verbose tracing in production without safeguards.
• Redact secrets and PII from logs and traces.

Deployment and evidence boundary

The playbook contains incomplete illustrative classes, undefined helper methods/imports and version-specific OpenTelemetry/Sentry configuration, not a ready-to-run debug stack. Inspect pinned APIs and existing instrumentation before adapting it; avoid duplicate server spans and extract/propagate trace context through supported instrumentation. Debuggers can execute code and heap/core snapshots can expose every in-memory secret. Keep inspectors and dashboards loopback-only behind approved authenticated tunnels, cap snapshot duration/disk/overhead, redact before export and remove temporary access after use. Production pause/profile/telemetry changes require scoped authorization and rollback. Do not replace process-global console methods per request or return sensitive diagnostics in response headers. A successful config parse does not establish end-to-end trace coverage or application correctness.

Resources

• resources/implementation-playbook.md for detailed tooling and configuration patterns.

Limitations

• Use this skill only when the task clearly matches the scope described above.
• Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
• Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.