prompt-security 开源项目推荐（92 个）

Tencent /AI-Infra-Guard

全栈 AI 红队测试平台，通过 Agent Scan、Skills Scan、MCP 扫描、AI Infra 扫描和 LLM 越狱评估来保障 AI 生态安全。

ZenGuard-AI /fast-llm-security-guardrails

The fastest Trust Layer for AI Agents

mega-edo /mega-security-leaderboard

Leaderboard Comparing LLM Agent Security on System Prompt Leakage and Attack Probes

prompt-armor /prompt-armor

Open-source prompt injection detector: 5 fused layers, offline, ~50ms. F1 82.9% in-sample / 84.2% ± 1.5% out-of-sample (leakage-audited). Apache 2.0

OthmanAdi /skill-deck

Lokale Werkbank zum Prüfen, Vergleichen und Härten von Skills für KI-Coding-Agenten.

Sevten /pi-data-masking

A Pi extension that replaces sensitive data with realistic placeholders before it reaches the LLM provider, then restores it locally for users and tools.

hermes-labs-ai /hermes-jailbench

Zero-LLM deterministic jailbreak regression benchmark: runs a repeatable battery of known-pattern attacks against an LLM endpoint and scores refusal/partial/compliance across runs. Catch when a model or prompt update silently got weaker. Single-turn, responsible-use. pip install hermes-jailbench

amirimi /Guardian

Extensible Go microkernel for LLM guardrails, quota enforcement and policy plugins.

upss-standard /universal-prompt-security-standard

Universal Prompt Security Standard (UPSS): A framework for externalizing, securing, and managing LLM prompts and genAI systems, inspired by and extending OWASP OPSS concepts for any organization or project.

AdirD /prompt-security-node

🚀 Unofficial Node.js SDK for Prompt Security's Protection API.

m4vic /promptxploit

LLM Penetration Testing Framework - Discover vulnerabilities in AI applications before attackers do. 100attacks + AI-powered adaptive mode.

KadirArslan /Mithra-Scanner

Mithra Scanner is an interactive API testing tool for prompt injection, refusal detection, and LLM security benchmarking. It supports YAML-based rule definitions, custom refusal lists, REST API integration, and provides detailed CLI output for security testing of language model endpoints.

Kushagratandon12 /cloakprompt-cli

CloakPrompt is a CLI tool that redacts secrets (passwords, API keys, credentials, etc.) before sending data to AI models.

jeslor /prompt-sanitizer

Privacy-first prompt sanitization. Fully local. Zero cloud calls. Fast, smart, and built for real-world AI workflows.

spectrayan /promptly

The enterprise AI governance and safety control plane for AI agent prompts

jouleka /aichu

Local Rust proxy that redacts secrets from prompts sent to AI coding agents (week-1 risk-validation scaffold)

yasir-naveed-riaz /secureai-guard

Lightweight AI security framework for prompt validation, output scanning, risk scoring, sensitive data detection, and Zero Trust policy enforcement.

benthompsondev /cloakscan

Local-first sanitizer for scripts, logs, prompts, and support text — clean secrets, hostnames, and org-specific terms on your device before sharing. PowerShell-aware Portfolio-code mode keeps sanitized code readable. Windows, Linux, and a live browser demo.

IulianVOStrut /ContextHound

Static analysis CLI that scans codebases for LLM prompt-injection, data-exfiltration, jailbreak, and unsafe agent/tool vulnerabilities. Runs fully offline, integrates with CI/CD, and outputs console, JSON, and SARIF reports.

anthfuller /F7-LAS

Fuller 7-Layer Agentic AI Security Framework (F7-LAS)