LogoClawIndex
CasesSkillsAbout
LogoClawIndex

sentinel-prompt-protection - Sentinel: SOTA Model for Prompt Injection Protection

Deploy binary classifier using ModernBERT to detect prompt injection attacks

Tags

Updated: 2026-05-09
securityprompt-injectionLLM-safetyclassificationModernBERT

Capabilities

classify promptsdetect injectionspredict malicious probabilityprotect LLM inputs

Typical Inputs

text promptclassification thresholdtrained model

Typical Outputs

classification labelconfidence probabilityprediction result

What this skill does

  • classify prompts
  • detect injections
  • predict malicious probability
  • protect LLM inputs

Inputs

  • text prompt
  • classification threshold
  • trained model
  • tokenizer

Outputs

  • classification label
  • confidence probability
  • prediction result

Requirements

  • ModernBERT model
  • CUDA-enabled GPU
  • PyTorch framework
  • transformers library

Source

  • Spec: SKILL.md

ClawIndex

OpenClaw Skills & Use Case Index

ClawIndex is an ecosystem-driven index of OpenClaw skills and real-world use cases.

Index

Skills·
Cases

Meta

About·
Disclaimer·
Email·
GitHub
© 2026 ClawIndex All Rights Reserved.