nvidia/nemo-relay-plugin-observability
Use this skill when choosing or configuring NeMo Relay 0.6 or 0.7 observability through the built-in plugin, subscribers, or exporters, including raw ATOF events, ATIF trajectories, OpenTelemetry, OpenInference, or custom event handling.
npx skills add https://github.com/NVIDIA/skills --skill nemo-relay-plugin-observability
Start with one exporter managed by the built-in Observability plugin. This is
the default for reusable process configuration and the best first plugin for
most users because it makes Relay's captured activity visible.
Choose one proof output before layering additional telemetry destinations.
Use manual subscriber or exporter APIs only when a test, script, or application
needs direct control over registration names, collection windows, or flush
timing. Both paths consume the same canonical event stream.
Determine whether the application uses NeMo Relay 0.6 or 0.7 before proposing
configuration or binding APIs. Prefer the installed package version, lockfile,
or manifest. Ask the user when the version cannot be established; do not mix
the two surfaces in one example.
OpenInference are separate exporters with OpenTelemetryConfig /
OpenTelemetrySubscriber and OpenInferenceConfig /
OpenInferenceSubscriber.
exporter provides full, gen_ai, and openinference projections.
Select the output that best matches the user's immediate inspection target:
Use a manual subscriber for short-lived in-process inspection.
Use ATOF JSONL; read references/atof.md.
Use ATIF; read references/atif.md.
For 0.6, choose the separate OpenTelemetry or OpenInference exporter. For
0.7, choose a typed OpenTelemetry endpoint (full, gen_ai, or
openinference). Read references/opentelemetry.md and, for an
OpenInference-aware backend, references/openinference.md.
Choose one output first and verify it before adding another. ATOF is the
default local proof because it preserves the raw event stream with the least
translation. Use synthetic, non-sensitive payloads for the first proof. Add and
verify sanitization before exporters receive production payloads, and never
display complete event records while validating an exporter.
Use this model when explaining how capture and export relate:
calls, managed LLM calls, middleware, and manual lifecycle APIs.
subscribers can observe the same stream for logging, export, analytics, or
diagnostics.
scope closes.
components.
into ATIF or the version-appropriate OpenTelemetry/OpenInference form.
use managed helpers or manual lifecycle params provide those fields.
compaction mark refreshes it.starts retain system instructions, the latest user message, and every
following assistant or tool message.
event-only projection to provider-shaped event input without changing
provider execution.
data field,typed profile data such as model_name and tool_call_id, and codec-provided
annotated LLM request/response data for in-process subscribers and exporters.
SKILL.mdautomatically emit skill.load marks under the tool span. The payload
contains only skill_name; metadata records the load source and tool name.
Partial reads do not count, and ambiguous slash-command expansions use the
separate skill.load.inferred name. The eager mark remains present if tool
execution later fails.
Use the names exported by the selected language binding and Relay version:
nemo_relay.subscribers.register(...), AtofExporter,AtifExporter, OpenTelemetrySubscriber, and OpenInferenceSubscriber
OpenTelemetrySubscriber for all three typed projections
nemo_relay::api::subscriber and nemo_relay::observability::*lifecycle methods
Load only the reference required by the selected output:
references/atof.md for raw JSONL events used in local debugging oroffline inspection.
references/atif.md for ATIF trajectories.references/opentelemetry.md for OTLP/OpenTelemetry traces.references/openinference.md for the standalone 0.6 OpenInferenceexporter or the 0.7 openinference OpenTelemetry projection.
Choose another skill when the task belongs to an adjacent workflow:
nemo-relay-plugin-build to package subscriber-based export behavior asa reusable plugin.
nemo-relay-get-started or nemo-relay-instrument-calls when no scope,tool call, or LLM call has been instrumented.
nemo-relay-debug-runtime-integration to diagnose missing telemetry.Use these skills for adjacent workflows:
nemo-relay-instrument-calls.nemo-relay-instrument-typed-wrappers.nemo-relay-plugin-build.nemo-relay-debug-runtime-integration.Take nvidia/nemo-relay-plugin-observability from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.