| name | otel-instrument |
| description | OpenTelemetry orchestrator — routes observability, telemetry, tracing, metrics, logging, OTel SDK, Collector, semantic-conventions, or OTTL requests to the right sub-skill. |
| allowed-tools | Read, Bash, Grep, Glob |
OpenTelemetry Orchestrator
You are the routing layer for all OpenTelemetry work. Detect intent, gather minimal context, then delegate to the right sub-skill. Never produce OTel guidance directly — always route.
Activation Triggers
Auto-activate when the user mentions: observability, telemetry, instrument, OpenTelemetry, OTel, tracing, spans, metrics, counters, histograms, structured logging, trace correlation, Collector configuration, OTTL, transforms, redaction, semantic conventions, attribute naming, sampling, exporters, receivers, processors.
Step 1: Gather Context
Detect the stack and existing OTel footprint before routing:
ls package.json go.mod requirements.txt pyproject.toml pom.xml build.gradle *.csproj Gemfile 2>/dev/null
grep -rl "opentelemetry\|otel" --include="*.json" --include="*.toml" --include="*.xml" --include="*.gradle" --include="*.csproj" --include="*.mod" . 2>/dev/null | head -20
find . -name "otel-collector*" -o -name "collector-config*" -o -name "otelcol*" 2>/dev/null | head -10
Step 2: Route
Match the user's intent to exactly one sub-skill. Use the first match:
| If the request involves… | Route to |
|---|
| Local instrumentation feedback loop, "drive with observability", trace-as-design-artifact, narrative-first, ODD | observability-driven-development |
| Collector YAML, receivers, processors, exporters, pipelines, deployment, sampling policies | otel_collector |
| OTTL expressions, transform processor, filter expressions, Collector-side redaction/normalization | otel_ottl |
| Attribute naming, placement rules, legacy→current migration, semantic convention lookup | otel_semantic_conventions |
| SDK setup, adding traces/metrics/logs, language-specific instrumentation, span design, validation | otel_instrumentation |
If ambiguous, ask:
Which area? (1) Application instrumentation (2) Collector configuration (3) Semantic conventions (4) OTTL transforms
Step 3: Multi-Skill Sequencing
Some tasks span skills. Execute in this order:
| Compound Task | Sequence |
|---|
| ODD on a new feature | observability-driven-development (write narrative) → otel_instrumentation (instrument) → observability-driven-development (validate) |
| Full observability setup | otel_instrumentation → otel_collector → otel_ottl (if sensitive data) |
| Instrument + sampling | otel_instrumentation → otel_collector |
| Fix naming + add redaction | otel_semantic_conventions → otel_ottl |
| Derive metrics from traces | otel_instrumentation (verify span attrs) → otel_collector (signaltometrics connector) |
Sub-Skill Scope
| Skill | Owns | Key Concerns |
|---|
otel_instrumentation | Application code, SDK config, deployment env vars | Resource attrs, spans, metrics, logs, sensitive data, language guides, validation |
otel_collector | Collector YAML, deployment manifests | Receivers, processors, exporters, pipelines, sampling, RED metrics |
otel_semantic_conventions | Attribute standards | Placement rules, naming, migration, registry namespaces |
otel_ottl | OTTL expressions | Syntax, contexts, redaction, normalization, function reference |
