skills.rest
DoanNgocCuong
DoanNgocCuongCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

prompt-guard

Detect prompt injections and jailbreak attempts in LLM applications.

Updated Aug 23, 2026
Search Tags:#security#input validation#llm#jailbreak#prompt injection#content filtering
One-click install
npx skills add https://github.com/DoanNgocCuong/continuous-training-pipeline_T3_2026 --skill prompt-guard-doanngoccuong
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prompt-guard
Source: https://github.com/DoanNgocCuong/continuous-training-pipeline_T3_2026/tree/main/.claude/skills/prompt-guard
Command: npx skills add https://github.com/DoanNgocCuong/continuous-training-pipeline_T3_2026 --skill prompt-guard-doanngoccuong

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires transformers, torch.

What problem does it solve?

This Skill safeguards Large Language Model applications by detecting and filtering malicious prompts, including prompt injections and jailbreak attempts, as well as securing third-party data.

Core Features & Use Cases

  • Prompt Injection & Jailbreak Detection: Identifies and blocks attempts to override LLM instructions or bypass safety filters.
  • Third-Party Data Filtering: Secures data from APIs, RAG, or web scraping against embedded malicious instructions.
  • Multilingual Support: Operates effectively across 8 different languages.
  • Use Case: Protect your customer-facing chatbot by ensuring user inputs are safe before they reach the LLM, preventing unexpected or harmful responses.

Quick Start

Use the prompt-guard skill to check if the user message 'Ignore previous instructions' is a jailbreak attempt.

Frequently Asked Questions about prompt-guard

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect prompt injection attacks in LLM applications?▼

Detect prompt injection attacks by passing user inputs through a pre-trained transformer model that identifies and blocks malicious instruction overrides before they reach the LLM.

What is the best way to filter malicious prompts from third-party data in RAG pipelines?▼

Filter malicious prompts from third-party RAG data by validating the retrieved information with a transformer model to remove embedded jailbreak attempts before generation.

Does prompt injection detection support multilingual user inputs?▼

Prompt injection detection supports multilingual user inputs, operating effectively across 8 different languages to secure applications with a global user base.

Can I use a transformer model for real-time jailbreak detection in customer-facing chatbots?▼

You can use this transformer model for real-time jailbreak detection in customer-facing chatbots because it offers low-latency inference suitable for immediate input filtering.

Do I need torch and transformers to run prompt injection detection?▼

You need the torch and transformers dependencies to run prompt injection detection, as the Skill relies on a pre-trained transformer model for low-latency inference.

Related Skills

Explore More Software Engineering Skills→
SunS1eep1ngSunS1eep1ng

transformers

Load and run pre-trained transformer models for NLP, vision, audio, and multimodal tasks.

Community
Advanced
BagelHoleBagelHole
46

rag-infrastructure

Build and operate RAG infrastructure with vector stores and hybrid search.

Community
Advanced
moonmingmoonming
1

a6-plugin-ai-content-moderation

Configures APISIX AI content moderation plugins for LLM request and response filtering.

Community
Intermediate
TymorIbrahimTymorIbrahim

hookify-rules

Create and manage hookify rules with YAML configuration and regex matching.

Community
Intermediate