nemo-relay-start

작성자: nvidia

Help application developers pick a NeMo Relay binding and get to a first working scope, tool call, and LLM call

npx skills add https://github.com/nvidia/nemo-relay --skill nemo-relay-start

Get Started With NeMo Relay

Use this skill for first-time users who want the shortest path to a working example. Rust, Python, and Node.js are the primary quick-start and hosted-docs paths. Go and the raw FFI surface are source-first advanced paths.

Default Path

  • Pick the user's host language first: Rust, Python, or Node.js. If they are using Go or raw FFI, verify names against tracked source and tests.
  • Prefer the managed execution APIs over manual lifecycle APIs.
  • Start with one scope, one tool call, and one LLM call.
  • Add observability only after the basic flow works.

Guidance

  • Rust: use nemo_relay::api::scope::{push_scope, pop_scope, event} with builder params, then nemo_relay::api::tool::tool_call_execute(...) and nemo_relay::api::llm::llm_call_execute(...)
  • Python: uv sync, then use nemo_relay.scope.scope(...), nemo_relay.tools.execute(...), and nemo_relay.llm.execute(...)
  • Node.js: build the addon, then use withScope(...), toolCallExecute(...), and llmCallExecute(...)
  • Go: use source-first wrappers such as scope.Push(...), tools.Execute(...), llm.Execute(...), or top-level PushScope(...), ToolCallExecute(...), and LlmCallExecute(...)
  • FFI: recommend only for binding or embedding work; verify C names such as nemo_relay_push_scope, nemo_relay_tool_call_execute, and nemo_relay_llm_call_execute in the current header

Common Pitfalls

  • Calling execute APIs without an active scope
  • Skipping the build step for Rust, Python, or Node.js
  • Assuming source-first Go/FFI bindings have the same hosted-doc coverage as Rust, Python, and Node.js
  • Mixing manual lifecycle APIs into a first example

Embedded Quick-Start Notes

  • Install from packages when building a consumer app: Rust uses cargo add nemo-relay, Python uses uv add nemo-relay, and Node.js uses npm install nemo-relay-node.
  • Use repository setup commands when working from a checkout: Rust builds the workspace, Python rebuilds the virtual environment and native extension with uv sync, and Node.js installs and builds the native addon before tests or examples run.
  • A first example should register a short-lived subscriber, open an agent scope, emit one mark event, run one managed tool call, run one managed LLM call, then deregister the subscriber. In Python use nemo_relay.subscribers; in Node.js use root exports such as registerSubscriber; in Rust use nemo_relay::api::subscriber.
  • Success means the app emits scope start/end events plus tool and LLM lifecycle events, and the application result remains the provider or tool result.
  • Scope handles are explicit in Rust and optional in higher-level Python and Node.js helpers when the active scope is already correct. Pass the handle when the surrounding framework makes parentage ambiguous.

Related Skills

  • nemo-relay-instrument-calls
  • nemo-relay-setup-observability
  • nemo-relay-debug-runtime-integration

nvidia의 다른 스킬

compileiq-debug
nvidia
Use when something is wrong: Search() hangs, all evaluations return INVALID_SCORE, scores aren't improving, every config returns the same number, ptxas errors…
official
create-github-pr
nvidia
gh CLI를 사용하여 GitHub 풀 리퀘스트를 생성합니다. 사용자가 새 PR을 만들거나, 코드 리뷰를 제출하거나, 풀 리퀘스트를 열고자 할 때 사용합니다. 트리거 키워드 -…
official
diagnose-perf
nvidia
First-responder performance triage for Isaac Sim and Isaac Lab. Identifies bottleneck category (GPU-bound, CPU-bound, VRAM, loading) using nvidia-smi and…
official
eagle3-review-logs
nvidia
Review EAGLE3 pipeline experiment logs from the launcher's experiments/ directory. Summarizes pass/fail status for all 4 tasks, diagnoses failures with root…
official
nemoclaw-maintainer-cross-issue-sweep
nvidia
다른 열린 이슈들을 스캔하여 주어진 PR이 함께 수정하거나 실수로 망가뜨릴 수 있는 이슈를 찾습니다. 인접 수정 기회와 모순 위험을 file:line…과 함께 출력합니다.
official
karpathy-guidelines
nvidia
일반적인 LLM 코딩 실수를 줄이기 위한 행동 지침입니다. 코드 작성, 검토 또는 리팩토링 시 과도한 복잡성을 피하고 정밀한 변경을 위해 사용하세요.
official
fhir-basics
nvidia
에이전트에게 FHIR R4 API의 작동 방식, 사용 가능한 리소스, 검색 매개변수를 사용한 쿼리 방법, 모든 응답 형식을 올바르게 파싱하는 방법을 가르칩니다…
official
underdeclared-agent
nvidia
A helpful assistant agent
official