find-traces

Analyzes OpenTelemetry traces to identify errors, latency hotspots, and trace-level patterns using queries on trace_id, service, duration, and error status via authenticated Axiom CLI sessions.

59|11|Updated Oct 19, 2020
One-click install
npx skills add https://github.com/axiomhq/cli --skill find-traces
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: find-traces
Source: https://github.com/axiomhq/cli/tree/main/skills/find-traces
Command: npx skills add https://github.com/axiomhq/cli --skill find-traces

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables rapid investigation of OpenTelemetry distributed traces by locating and analyzing traces that match a given trace ID or search criteria (errors, latency, service) to diagnose issues in distributed systems.

Core Features & Use Cases

  • Trace discovery across datasets containing traces (e.g., otel-*, traces, spans)
  • Query traces by trace_id, service, duration, and error status to surface relevant traces quickly
  • Analyze traces to identify errors, latency hotspots, and potential root causes
  • Use case: When investigating a slow transaction, run a trace search by trace_id to surface related spans and timing, then correlate with service maps and dashboards

Quick Start

Invoke the skill with a trace ID, for example /find-traces abc123def456 to retrieve matching traces. Then refine results with dataset discovery and service filters as needed.

Frequently Asked Questions about find-traces

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

FAQPage Schema
How do I query OpenTelemetry traces to find errors and latency hotspots?

To query OpenTelemetry traces, you search across trace datasets by trace_id, service, duration, and error status. This analyzes distributed traces to identify errors, latency hotspots, and potential root causes quickly.

How do I search for a specific trace_id in my distributed traces?

You can search for a specific trace_id by invoking a trace search query. This surfaces related spans and timing data for that trace_id, allowing you to correlate the results with service maps and dashboards to diagnose slow transactions.

Do I need an authenticated Axiom CLI session to analyze distributed traces?

Yes, analyzing distributed traces requires an authenticated Axiom CLI session. This session provides the necessary access to query trace datasets containing OpenTelemetry field conventions like trace_id, span_id, and duration.

What datasets can I use for trace discovery and analysis?

Trace discovery works across datasets containing OpenTelemetry traces, such as otel-*, traces, and spans. Querying these datasets by service, duration, and error status surfaces relevant traces to locate issues in distributed systems.

Can I investigate slow transactions by analyzing OpenTelemetry spans?

Yes, you can investigate slow transactions by running a trace search to surface related spans and timing. Analyzing these OpenTelemetry spans identifies latency hotspots and helps diagnose the root cause of slow distributed transactions.