Halaman ini diterjemahkan oleh mesin. Baca versi asli dalam bahasa Inggris.

Component

core/security/nemo-guardrails

NVIDIA NeMo Guardrails' server as a guard in a model's path: a service speaking the OpenAI chat-completions API that checks each request against its input rail, forwards what it admits to the upstream model API in infra.upstream_endpoint, and returns the upstream's answer, streamed or not. A request the rail refuses is answered by the guard itself with an ordinary completion, and nothing of it reaches the upstream. The rail sends the latest user message to the detector wired on the detector socket and refuses the request when the detector reports one of values.blocked_labels at or above values.threshold; a detector that cannot be reached, or answers with an error, refuses the request too. It judges the latest user message alone. Its verdict is a probabilistic classifier's: it misses some attacks, refuses some legitimate requests and is known to be evadable. Only a flow in its configuration, redefining the product's own, keeps a request body from switching the rail off, and that closure has to be proved again for any image or flow it was not proved on. NVIDIA's NemoGuard models add rails on the content_safety and jailbreak_detection sockets; the jailbreak rail admits a request when its classifier is unreachable. It authenticates no caller, and with no credential attached to its upstream_key slot a caller can reach the upstream on a credential of its own. The Release supplies the image in infra.image and the catalog names none: TrustyAI's published build carries NLTK's punkt tokenizer model, whose licence is unresolved upstream, and is published only under a floating tag, whose superseded digests stop resolving. Nothing renders until the Release sets `infra.image`.

Asal-usul

ResmiDitandatangani oleh identitas penandatangan yang diakui katalog ini sebagai penerbit resmi.
Digest konten
sha256:f871e845d92afd93e2d6d8c7fc6a9b33f86abbcfe1730cc89aab9d44fb3f2721
Dipublikasikan
2026-11-10T10:13:39Z
Identitas penandatangan
https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main

Tanda tangan

  • Resmiapplication/vnd.dev.sigstore.bundle.v0.3+json

    https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main

    sha256:4a2b09d9b13c5c2efc29928cc9bd5abc2304b72043e199a4b9ffddab6eb8b06e

Referensi pada versi ini

oci://hub.aggrega.ai/component/core/security/nemo-guardrails:1.0.0

Parameter pengembang

Nilai yang ditetapkan pengembang di Aggregate.

NamaTipeBawaanWajibDeskripsi
thresholdnumber (min 0, max 1)0.5tidakProbability at or above which a label the detector reports refuses the request. Measured with the detector's default model, which scores most texts close to 0 or close to 1: at 0.5 the guard refuses plain attempts to override a model's instructions; at 0.001 it also refuses an override respelled in leetspeak, which the model scores near 0.008, and with it some ordinary requests the model scores low but not near zero. At either value it refuses the ordinary requests the model scores close to 1, and serves attacks it scores close to 0. Those two numbers hold for that model only, and are measured again for any other.
blocked_labelsarray (min 1)INJECTIONtidakLabels of the detector's classifier that refuse a request when reported at or above the threshold. The default is the label the detector's default model gives an attempt to override a model's instructions; another model labels by its own vocabulary, which this list follows. It may not be empty, because a guard refusing on no label forwards everything.

Penentuan ukuran oleh operator

Ukuran infrastruktur yang disesuaikan operator untuk setiap lingkungan.

NamaTipeBawaanWajibDeskripsi
portinteger (min 1024, max 65535)8000tidakPort the guard listens on and its Service accepts callers on. Most operators should omit this and let the Recipe assign the default (8000), the port the published image listens on unless told otherwise.
upstream_endpointstring—yaBase URL of the OpenAI-compatible model API the guard forwards the requests it admits to, with its API path, for example http://model-server:8000/v1. A deterministic pointer string the guard resolves when it forwards; the compiler never contacts it. The request is forwarded under the model name the caller asked for.
imagestring—yaImage the guard runs, supplied by the Release; the catalog names none. The Recipes run the published build's own start-up and paths, so the image is TrustyAI's NeMo Guardrails server or a copy of it. That build carries a data package whose licence is unresolved upstream and is tagged only as latest, under which the registry stops serving a build once the tag moves on, so a Release that must keep one build mirrors it and names the copy here. The guard's refusal of a request-body switch was proved on one build, and is proved again for any other image.
computeobject—tidak—

Kontrak keluaran

Nilai yang dipublikasikan komponen ini kepada konsumen melalui topologi uses.

NamaTipeBawaanWajibDeskripsi
hoststring—yaIn-cluster DNS name of the Service fronting the guard. Every caller reaching this name can send requests through the guard, so it is an address for the component routing model traffic through it rather than one to publish.
portinteger—yaTCP port that Service accepts callers on. It follows the port an operator set in the Release, so a consumer reads it rather than assuming a default.
endpointstring—yaBase URL of the OpenAI-compatible API the guard serves, with the host, the port and the API path already composed. A consumer hands it to a client library as its base URL, and sends chat completions to it.
protocolstring—tidakURL scheme to reach `host` and `port` over, for a consumer composing its own address rather than reading `endpoint`.
thresholdnumber—yavalues.threshold, the probability at or above which a label the detector reports refuses the request.

Soket uses

Soket dependensi yang menjadi jalur koneksi komponen ini. Target setiap koneksi yang dihubungkan melalui sebuah soket harus memenuhi selektor labelnya, dan setiap koneksi dapat membawa lampiran pada slot yang dideklarasikan soket tersebut. Slot wajib harus diisi oleh setiap koneksi yang dihubungkan oleh Aggregate atau Release melalui soket tersebut.

detector

Selektor
security.aggrega.ai/text-detection=detector
Injeksi
—

Soket ini tidak mendeklarasikan slot lampiran, sehingga koneksi melaluinya tidak membawa lampiran.

content_safety

Selektor
component.aggrega.ai/category=ai-model, model.aggrega.ai/api=openai
Injeksi
—

Soket ini tidak mendeklarasikan slot lampiran, sehingga koneksi melaluinya tidak membawa lampiran.

jailbreak_detection

Selektor
component.aggrega.ai/category=security
Injeksi
—

Soket ini tidak mendeklarasikan slot lampiran, sehingga koneksi melaluinya tidak membawa lampiran.

Slot lampiran

Titik pengikatan bernama tempat sebuah Feature dilampirkan ke komponen ini. Selektor sebuah slot membatasi Feature mana yang dapat mengisinya, dan daftar injeksinya menyebutkan cara materi Feature dapat dikirimkan. Slot wajib harus terikat pada setiap instans, melalui lampiran pada instans tersebut atau pada cakupan di atasnya, dan slot sensitif dideklarasikan untuk menangani materi rahasia. Sebuah Feature juga dapat dilampirkan dengan nama yang tidak dideklarasikan oleh slot mana pun.

SlotSelektorWajibInjeksiSensitifDeskripsi
upstream_keyfeature.aggrega.ai/kind=secrettidakenvyaOptional credential the guard presents to the upstream model API as a bearer token. Attached, every forwarded request carries it, whatever the caller sends. Unattached, the guard holds no upstream credential, which suits an upstream that asks for none: the product then sends a fixed placeholder in its place, which an upstream requiring a key refuses, or passes a caller's X-Authorization header on as the credential when the caller sends one.

Daftar izin injeksi mandiri

Tidak ada pipa injeksi mandiri yang dideklarasikan.

Relasi

Sisi bertipe ke dan dari entitas lain yang dipublikasikan. Referensi tanpa target yang dipublikasikan ditampilkan sebagai sisi yang terputus, bukan sebagai tautan.

Menggunakan

Digunakan oleh

Referensi konsep Component →

Katalog Aggrega — indeks terbuka berisi komponen, fitur, resep, dan plugin yang merendernya.

Dibangun dengan sistem desain Aggrega yang digunakan bersama.

Privasi