Observability Tracing

Overview

Use this skill to make critical request flows traceable end-to-end for latency and failure diagnosis.

Scope Boundaries

• Use this skill when the task matches the trigger condition described in description.
• Do not use this skill when the primary task falls outside this skill’s domain.

Shared References

• Span attribute and sampling rules:
  • references/span-attribute-and-sampling-rules.md

Templates And Assets

• Trace coverage map template:
  • assets/trace-coverage-map-template.md
• Trace quality checklist:
  • assets/trace-quality-checklist.md

Inputs To Gather

• Critical cross-service user flows.
• Current instrumentation and propagation gaps.
• Sampling constraints and storage budget.
• Incident diagnosis requirements.

Deliverables

• Trace coverage map for critical paths.
• Span attribute conventions and sampling policy.
• Validation evidence for trace queryability.

Workflow

1. Define coverage targets in assets/trace-coverage-map-template.md.
2. Apply span/sampling policy from references/span-attribute-and-sampling-rules.md.
3. Validate quality via assets/trace-quality-checklist.md.
4. Fix missing spans/attributes on critical paths.
5. Publish tracing baseline and ownership.

Quality Standard

• Critical flows are traceable end-to-end.
• Span attributes are consistent for diagnosis queries.
• Sampling preserves incident usability.

Failure Conditions

• Stop when critical paths remain untraceable.
• Stop when trace context propagation is inconsistent.
• Escalate when tracing gaps block root-cause isolation.