원클릭으로 Manus에서 모든 스킬 실행

시작하기

nemo-relay-export-openinference

스타73

포크40

업데이트2026년 7월 1일 15:57

Configure and use NeMo Relay OpenInference export for OTLP backends that understand OpenInference semantics

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

NVIDIA

NVIDIA/NeMo-Relay

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

Export OpenInference Traces

Use this skill when the destination expects OpenInference semantic conventions, for example Arize Phoenix or another OpenInference-aware OTLP backend.

Default Path

Build the binding-specific OpenInferenceConfig
Set endpoint, transport, service metadata, and headers
Construct and register the subscriber
Run instrumented scoped work
Deregister, flush, and shut down when done

Embedded OpenInference Semantics

OpenInference export is for OTLP backends that understand model-centric OpenInference semantic conventions.
Configure transport, endpoint, service_name, optional namespace and version, instrumentation scope, headers, resource attributes, and timeout.
Start with http_binary transport and an OTLP/HTTP traces endpoint. Use grpc only when a Tokio runtime is active.
Scope, tool, and LLM start inputs become input.value.
Scope, tool, and LLM end outputs become output.value.
LLM usage metadata maps token counters when provider responses include usage.
Use explicit config fields for endpoint, headers, resource attributes, and service identity in application code.
Validate export by checking construction logs, collector traffic, and spans from the same root_uuid in the tracing backend.

Important Semantics

Spans include OpenInference semantic attributes
LLM spans derive input.value from request content, not request headers
Scope types map to OpenInference span kinds
Orphan mark events still export as zero-duration spans

Troubleshooting Focus

No spans in the OpenInference-aware backend
Expected semantic attributes missing
Wrong scope types or no active scope
Wrong OTLP transport for the chosen binding or target

Related Skills

nemo-relay-setup-observability
nemo-relay-export-otel
nemo-relay-typed-wrappers-codecs

이 저장소의 다른 Skills

같은 저장소

draft-release-notes

NVIDIA/NeMo-Relay

Compare NeMo Relay release branches and draft or update the documentation-site release notes. Use when preparing a release-notes update under docs/about-nemo-relay/release-notes, reviewing release-to-release changes, or gathering evidence for a current release summary.

2026-07-0873

contribute-docs

NVIDIA/NeMo-Relay

Contribute documentation or example changes that stay aligned with NeMo Relay public behavior

2026-07-0273

nemo-relay-build-plugin

NVIDIA/NeMo-Relay

Build and package reusable NeMo Relay runtime behavior as a config-activated plugin with validation and rollback-safe registration

2026-07-0273

add-binding-feature

NVIDIA/NeMo-Relay

Add or change a public NeMo Relay API surface across the core runtime and every affected binding

2026-07-0173

maintain-observability

NVIDIA/NeMo-Relay

Maintain or extend NeMo Relay observability surfaces across ATIF, OpenTelemetry, and OpenInference

2026-07-0173

maintain-optimizer

NVIDIA/NeMo-Relay

Maintain or extend the NeMo Relay adaptive surface across config, plugins, docs, and bindings; use this when users still say optimizer

2026-07-0173

name	nemo-relay-export-openinference
description	Configure and use NeMo Relay OpenInference export for OTLP backends that understand OpenInference semantics
author	NVIDIA Corporation and Affiliates
license	Apache-2.0

Export OpenInference Traces

Use this skill when the destination expects OpenInference semantic conventions, for example Arize Phoenix or another OpenInference-aware OTLP backend.

Default Path

Build the binding-specific OpenInferenceConfig
Set endpoint, transport, service metadata, and headers
Construct and register the subscriber
Run instrumented scoped work
Deregister, flush, and shut down when done

Embedded OpenInference Semantics

OpenInference export is for OTLP backends that understand model-centric OpenInference semantic conventions.
Configure transport, endpoint, service_name, optional namespace and version, instrumentation scope, headers, resource attributes, and timeout.
Start with http_binary transport and an OTLP/HTTP traces endpoint. Use grpc only when a Tokio runtime is active.
Scope, tool, and LLM start inputs become input.value.
Scope, tool, and LLM end outputs become output.value.
LLM usage metadata maps token counters when provider responses include usage.
Use explicit config fields for endpoint, headers, resource attributes, and service identity in application code.
Validate export by checking construction logs, collector traffic, and spans from the same root_uuid in the tracing backend.

Important Semantics

Spans include OpenInference semantic attributes
LLM spans derive input.value from request content, not request headers
Scope types map to OpenInference span kinds
Orphan mark events still export as zero-duration spans

Troubleshooting Focus

No spans in the OpenInference-aware backend
Expected semantic attributes missing
Wrong scope types or no active scope
Wrong OTLP transport for the chosen binding or target

Related Skills

nemo-relay-setup-observability
nemo-relay-export-otel
nemo-relay-typed-wrappers-codecs