tracing

Enforce OpenTelemetry naming and instrumentation conventions in Ghost codebase.

5|Updated Feb 24, 2026
One-click install
npx skills add https://github.com/mrtolkien/GHOST --skill tracing
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tracing
Source: https://github.com/mrtolkien/GHOST/tree/main/.agents/skills/tracing
Command: npx skills add https://github.com/mrtolkien/GHOST --skill tracing

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ensures consistent tracing, observability, and instrumentation practices across the Ghost codebase to produce reliable telemetry and easier debugging.

Core Features & Use Cases

  • Define and enforce span naming conventions to maintain low-cardinality names.
  • Provide OpenTelemetry (OTel) conventions for provider calls, tool executions, and entry points.
  • Offer guidance for instrumenting code with meaningful boundaries and attributes, reducing noisy instrumentation.
  • Use across development workflows to harmonize tracing spans, logs, and metrics during debugging and performance investigations.

Quick Start

Read this guide and apply these tracing conventions whenever you add or modify instrumentation in Ghost.

Frequently Asked Questions about tracing

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

FAQPage Schema
How do I enforce consistent OpenTelemetry tracing conventions in my codebase?

To enforce consistent tracing, apply standardized span naming patterns and semantic conventions across provider calls, tool executions, and critical entry points. This ensures reliable telemetry and improves debuggability during development.

What is the best way to reduce span cardinality when instrumenting telemetry?

To reduce span cardinality, enforce low-cardinality span names and attach explicit span attributes during instrumentation. This minimizes noisy telemetry and produces meaningful boundaries for performance investigations and debugging.

How do I instrument gen_ai fields using OpenTelemetry semantic conventions?

Instrumenting gen_ai fields requires adherence to OTel semantic conventions, including explicit span attributes and proper error handling. This applies when introducing or modifying instrumentation points across critical code entry points.

When should I add tracing spans and log statements during development?

Add tracing spans and log statements whenever you introduce or modify instrumentation in code. This is applicable during development and code changes to harmonize tracing spans, logs, and metrics for performance investigations.

Does this tracing guidance work for provider calls and tool executions?

Yes, this tracing guidance provides specific OpenTelemetry conventions for provider calls, tool executions, and critical entry points. It requires adherence to OTel naming patterns, instrument scopes, and semantic conventions to ensure telemetry quality.

Why does my observability instrumentation produce noisy telemetry data?

Noisy telemetry often results from inconsistent span boundaries and missing explicit span attributes. Enforcing consistent OTel naming patterns and semantic conventions reduces noisy instrumentation and produces reliable, debuggable telemetry.