Skip to content
Back to the index

Shield Small

huggingface.coInfrastructure

PulseGate's liveness check found it on 3 Oct 2026; it has been in the index since 3 Oct 2026. How this is checked

Shield is a 118M-parameter ONNX text-classification model for identifying prompt injections and jailbreak attempts. Developers can download and run the open model locally in AI security and guardrail pipelines.

Inferred · not functionally tested

Open SourceWebCLISelf-hostedAPI
Shield Small preview
Visit huggingface.co

Overview

5 features

Purpose: Detecting prompt injections and jailbreak attempts in AI inputs.

Inferred · not functionally tested

Audience: AI developers and security engineers

Inferred · not functionally tested

Functions: Unknown

Interfaces: API: indicated (inferred, not tested) · MCP: unknown · CLI: indicated (inferred, not tested) · Self-hosting: indicated (inferred, not tested)

Recorded constraints: pricing: open_source · license: Open Source · platforms: CLI, WEB · deployment: browser, cli, self_hosted, api_only

Constraint provenance is unknown; confirm requirements with the publisher.

Record sources: huggingface.co. These links do not verify the individual claims.

In the AI security & guardrails space, Shield Small takes a focused approach. Inferred · not functionally tested: It focuses on detecting prompt injections and jailbreak attempts in AI inputs. Inferred · not functionally tested: Shield Small is an open-source project aimed at AI developers and security engineers. Basis unknown · not verified: Shield Small is open source under the Open Source license. Basis unknown · not verified: It runs on the web, the command line, and API, and it can be self-hosted.

It is developed by ZeroLeaks. Inferred · not functionally tested: Among its 5 catalogued features are Prompt Injection Detection, Jailbreak Detection, and Text Classification.

Summary written by a language model from the project’s public pages.

Tasks: Inferred · not functionally tested

  • Prompt Injection Detection
  • Jailbreak Detection
  • Text Classification
  • ONNX Runtime
  • Local Inference

Topics: Inferred · not functionally tested

Tags
prompt-injection-detectionjailbreak-detectiononnx-modelllm-security
AI capabilities
TextStructured
Inference: LocalWeights: Open

JSON profile · Text profile · Access guide

Built with & integrations

Hosting
AWS
Runs on
BrowserCLISelf-hostedAPI-only
Detected from
AWS
x-amz-cf-id header · x-amz-cf-pop header · via header

Trust & compliance

License
Open Source
Public signals
HTTPSPrivacy PolicyTerms of ServiceOpen SourceFree tier
Legal
Privacy Policy Terms of Service

Indexing history

1

What PulseGate has recorded for this listing

  1. Indexed3 Oct · 00:28 UTC
    Shield: A 118M model for detecting prompt injections and jailbreaks seen via Hacker News firehose (Algolia)
    Source: Hacker News firehose (Algolia) · Open

Frequently asked questions about Shield Small

What is Shield Small?
Inferred · not functionally tested: Shield Small focuses on detecting prompt injections and jailbreak attempts in AI inputs. It is catalogued under AI security & guardrails on PulseGate.
Who is Shield Small for?
Inferred · not functionally tested: Shield Small is an open-source project built for AI developers and security engineers.
Is Shield Small free?
Basis unknown · not verified: Yes — Shield Small is open source under the Open Source license and free to use.
What platforms does Shield Small run on?
Basis unknown · not verified: Shield Small runs on the web, the command line, and API. It can also be self-hosted.
Is Shield Small still active?
PulseGate's liveness check found it on 3 Oct 2026.
What projects are similar to Shield Small?
Similar projects tracked by PulseGate include VITNA, cloakroom, and Privacy Policy.VITNAcloakroomPrivacy Policy
Who makes Shield Small?
Shield Small is developed by ZeroLeaks.
Is Shield Small open source?
Basis unknown · not verified: Yes — Shield Small is open source under the Open Source license.

Also in AI security & guardrails

Same category — not a similarity match