metrics-queries
How to query OpenTelemetry metrics datasets in Honeycomb correctly. Metrics datasets follow different rules from trace/event datasets — many operations (bare COUNT, RATE_SUM, RATE_AVG, RATE_MAX, CONCURRENCY) are forbidden, temporal aggregation is automatic, and each metric has its own attributes. Use this skill when querying a metrics dataset (gauges, counters, histograms, sums), asking about temporal aggregation (RATE, INCREASE, SUMMARIZE, LAST), finding the metrics dataset or discovering metric names and attributes, debugging unexpected metrics query results, or querying infrastructure metrics like CPU, memory, disk I/O, or network stats. Do NOT use for instrumenting metrics (use otel-instrumentation), querying event datasets with "metrics" in their name, or conceptual questions (use observability-fundamentals).
94
92%
Does it follow best practices?
Impact
96%
1.68xAverage score across 3 eval scenarios
Passed
No known issues
Evaluation results
89%
11%Kubernetes Cluster Health Dashboard
Correct metrics dataset identification and forbidden operation avoidance
Dataset identification method
0%
75%
No dataset guessing by name
100%
100%
Attribute discovery step
0%
20%
No bare COUNT on metrics
100%
100%
No RATE_SUM / RATE_AVG / RATE_MAX
100%
100%
CPU query uses AVG or MAX
100%
100%
Memory query uses AVG or MAX
100%
100%
Correct breakdowns per query
100%
100%
Three queries produced
100%
100%
No CONCURRENCY operation
100%
100%
100%
61%Per-Second Request Throughput and Error Rate Analysis
Temporal aggregation override with RATE calculated fields and query math
RATE calculated field for request rate
0%
100%
No RATE_SUM or RATE_AVG on query 1
0%
100%
Spatial aggregation on rate field
0%
100%
range_interval_seconds used for smoothing
0%
100%
RATE calculated fields for error query
0%
100%
Formula for error percentage
100%
100%
Named calculations for formula reference
100%
100%
No nesting of temporal functions
100%
100%
Service breakdown on query 1
100%
100%
approach.md documents temporal override
8%
100%
100%
44%HTTP Latency Analysis for a Latency Regression Investigation
Histogram parent column vs sub-fields and tail ratio query math
Parent column for HEATMAP
100%
100%
Parent column for P50 and P99
0%
100%
No AVG on .p99 sub-field
100%
100%
Throughput uses .count sub-field
30%
100%
Tail ratio via named calculations
41%
100%
Service breakdown on per-service queries
100%
100%
methodology.md explains parent vs sub-field
25%
100%
No bare COUNT on metrics
100%
100%
No RATE_SUM / RATE_AVG / RATE_MAX
0%
100%
Order clause on per-service query
100%
100%
- Repository
- honeycombio/agent-skill
- Commit
53e6bb8
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.