tune-sampling

Select and configure head and tail sampling strategies for tracing costs.

7|3|Updated Nov 26, 2025

Search Tags:#performance #observability #trace #sampling #cost-management

Ready to run

npx skills add https://github.com/jagreehal/autotel --skill tune-sampling

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: tune-sampling
Source: https://github.com/jagreehal/autotel/tree/main/packages/autotel/skills/tune-sampling
Command: npx skills add https://github.com/jagreehal/autotel --skill tune-sampling

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides strategies to manage tracing costs and visibility by selecting appropriate head and tail sampling methods, ensuring critical data is retained without incurring excessive expenses.

Core Features & Use Cases

Sampling Strategy Selection: Choose between head and tail sampling based on cost, performance, and data importance.
AI-aware and Customer-controlled Sampling: Incorporate AI-generated spans and customer debug headers into sampling decisions.
Use Case: Adjust sampling rates for high-volume API services to balance observability with budget constraints, or enable customers to request full trace capture when needed.

Quick Start

Use the tune-sampling skill to determine the most effective sampling approach for your observability needs, balancing cost and data retention.

Frequently Asked Questions about tune-sampling

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

FAQPage Schema

What is the difference between head and tail sampling for distributed tracing?▼

Head sampling makes early trace collection decisions to save resources, while tail sampling evaluates complete traces to retain critical data. This Skill provides strategies to balance tracing cost and data quality across diverse application scenarios using both methods.

How do I reduce observability costs for high-volume API services without losing critical trace data?▼

You can reduce observability costs by applying intelligent sampling strategies to high-volume API services. This Skill guides you in configuring head and tail sampling methods to balance trace visibility with your specific budget constraints.

Can I use customer-driven debug headers to trigger full trace capture?▼

Yes, customer-controlled sampling allows you to incorporate customer debug headers into sampling decisions. This Skill explains how to enable customers to request full trace capture on demand while maintaining standard cost-management sampling for general traffic.

When should I use AI-aware sampling strategies for trace collection?▼

AI-aware sampling is useful when you need to incorporate AI-generated spans into your trace collection decisions. This Skill covers how to apply these sampling methods in distributed systems to ensure meaningful observability data is retained.

How do I choose the best tracing sampling strategy for a distributed system?▼

Choosing the best tracing sampling strategy requires evaluating cost, performance, and data importance. This Skill helps you select and configure appropriate head or tail sampling approaches to optimize data quality and manage resource expenses in distributed systems.