LLM Firewall Defense LLM Tools

Tools for real-time detection and blocking of prompt injection, jailbreaks, and malicious tool calls in LLM applications. Does NOT include vulnerability research, fuzzing frameworks, or governance/compliance tooling.

There are 92 llm firewall defense tools tracked. 2 score above 70 (verified tier). The highest-rated is ethz-spylab/agentdojo at 75/100 with 471 stars and 15,222 monthly downloads. 2 of the top 10 are actively maintained.

Get all 92 projects as JSON

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=llm-tools&subcategory=llm-firewall-defense&limit=20"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

#	Tool	Score	Tier	Stars	Language
1	ethz-spylab/agentdojo A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents.	75	Verified	471	Python
2	guardrails-ai/guardrails Adding guardrails to large language models.	70	Verified	6,534	Python
3	JasonLovesDoggo/caddy-defender Caddy module to block or manipulate requests originating from AIs or cloud...	46	Emerging	508	Go
4	inkdust2021/VibeGuard Uses just 1% memory while protecting 99% of your personal privacy.	42	Emerging	130	Go
5	Heiberg-Industries/designbrief Design guardrails, not templates. A library of UI design direction files...	41	Emerging	9	JavaScript
6	ankitlade12/AgentArmor The full-stack safety layer for AI agents. Budget limits, prompt injection...	39	Emerging	2	Python
7	AmenRa/GuardBench A Python library for guardrail models evaluation.	39	Emerging	34	Python
8	mguard-ai/mguard Memory defense for AI agents — stops MINJA, AgentPoison, and MemoryGraft...	37	Emerging	1	TypeScript
9	deadbits/vigil-llm ⚡ Vigil ⚡ Detect prompt injections, jailbreaks, and other potentially risky...	37	Emerging	464	Python
10	whitecircle-ai/circle-guard-bench First-of-its-kind AI benchmark for evaluating the protection capabilities of...	37	Emerging	51	Python
11	hexitlabs/vigil 🛡️ Open-source safety guardrail for AI agent tool calls. <2ms, zero dependencies.	35	Emerging	5	TypeScript
12	turbot/guardrails-lib-ai A unified interface for multiple AI language model providers, tailored for...	35	Emerging	1	JavaScript
13	dropbox/llm-security Dropbox LLM Security research code and results	35	Emerging	256	Python
14	lukeslp/ux-oss-safeguard Content safety evaluator built on OpenAI's gpt-oss-safeguard-20b — zero...	35	Emerging	1	HTML
15	mthamil107/prompt-shield Self-learning prompt injection detection engine that gets smarter with every...	34	Emerging	4	Python
16	pampanic/pam_panic A PAM module that protects sensitive data and provides a panic function for...	33	Emerging	60	C
17	tyoung1996/guardrail-layer Guardrail Layer: Open-source AI data privacy firewall — redact, audit, and...	31	Emerging	5	TypeScript
18	kdunee/intentguard A Python library for verifying code properties using natural language assertions.	29	Experimental	34	Python
19	henchiyb/breaker-ai Breaker AI - Security check for your LLM prompts	28	Experimental	8	TypeScript
20	North-Shore-AI/LlmGuard AI Firewall and guardrails for LLM-based Elixir applications	28	Experimental	7	Elixir
21	Zierax/Basic-ML-prompt-injections llm attacks basic payloads	28	Experimental	9	—
22	yihedeng9/DuoGuard DuoGuard: A Two-Player RL-Driven Framework for Multilingual LLM Guardrails	27	Experimental	32	Python
23	upss-standard/universal-prompt-security-standard Universal Prompt Security Standard (UPSS): A framework for externalizing,...	26	Experimental	5	Python
24	automorphic-ai/aegis Self-hardening firewall for large language models	26	Experimental	267	Python
25	recurprotocol/recur-protocol Self-evolving recursive AI sentinels for real-time IP protection and prompt...	25	Experimental	3	JavaScript
26	sleeepeer/PIArena PIArena: A Platform for Prompt Injection Evaluation	25	Experimental	4	Python
27	chenglin1112/AgentTrust Real-time trustworthiness evaluation and safety interception for AI agents....	25	Experimental	3	Python
28	k14uz/PhishNet PhishNet is an experimental research project implementing Reinforced...	24	Experimental	4	Python
29	Agentic-AI-Risk-Mitigation/Janus System-level security for LLM agents: fine-grained policy enforcement on...	24	Experimental	2	Python
30	EctoSpace/EctoLedger The dashcam and emergency brake for AI agents. A security proxy that...	24	Experimental	2	Rust
31	willshacklett/gvai-safety-systems Runtime AI safety & security infrastructure for monitoring constraint strain...	24	Experimental	2	Python
32	microsoft/llmail-inject-challenge-analysis Data Analysis of the results of llmail-inject challenge	23	Experimental	10	Jupyter Notebook
33	dfghdrtawD/TrustLayer--Security-Control-Plane-For-LLM-AI 🛡️ Secure AI agents with TrustLayer's LLM Firewall, block prompt injection,...	23	Experimental	1	—
34	Buddafest/wonderwallai WonderwallAi — Open-source AI firewall SDK for LLM applications. Prompt...	23	Experimental	1	Python
35	safe-ai-factory/saifctl Safety harness for autonomous AI agents: Spec-driven AI factory. Use with...	23	Experimental	1	TypeScript
36	acebot712/promptguard-node PromptGuard Node.js SDK — Drop-in security for AI applications	23	Experimental	1	TypeScript
37	danielmaddaleno/llm-guardrails-toolkit Pluggable guardrails pipeline for LLM apps – PII redaction, prompt...	22	Experimental	—	Python
38	zjdtm/secure-prompt-spring-boot-starter Spring Boot LLM Prompt Injection Defense Library	22	Experimental	—	Java
39	ogulcanaydogan/Prompt-Injection-Firewall Real-time prompt injection detection and prevention middleware for LLM...	22	Experimental	—	Go
40	MoeinAlvandi/sovereign-vault 🛡️ Automate your data backups with Sovereign Vault, ensuring secure 3-2-1...	22	Experimental	—	Shell
41	heymumford/cognilateral-trust AI that tells you when it's guessing. Confidence tiers + accountability for...	22	Experimental	—	Python
42	AUTHENSOR/prompt-injection-benchmark Standardized benchmark for testing AI safety scanners. Run your scanner, get...	22	Experimental	—	TypeScript
43	studiomeyer-io/ai-shield LLM security toolkit — prompt injection detection, PII masking, cost...	22	Experimental	—	TypeScript
44	marcin-jasinski/outlier-vanguard Prompt injection detection via novelty scoring. Securing agentic AI systems...	22	Experimental	—	—
45	mrSamDev/llm-moat TypeScript toolkit for prompt injection detection, sanitization, and LLM...	22	Experimental	—	TypeScript
46	isartor-ai/Isartor Pure-Rust Prompt Firewall that eliminates unnecessary LLM cloud calls....	22	Experimental	—	Rust
47	michusSq/configguard Detect network misconfigurations using AI to ensure compliance with security...	22	Experimental	—	Python
48	WardLink/TrustLayer--Security-Control-Plane-For-LLM-AI TrustLayer is an API-first security control plane for LLM apps and AI...	22	Experimental	4	—
49	AdirD/prompt-security-node 🚀 Unofficial Node.js SDK for Prompt Security's Protection API.	22	Experimental	7	TypeScript
50	sammm0308/bonklm Enforce security guardrails for large language models in Node.js...	22	Experimental	—	TypeScript
51	GaBySuGy/agent-guardrails 🛡️ Enforce AI rules with Agent Guardrails, preventing breaches and ensuring...	22	Experimental	—	Shell
52	logicbunchhq/ai_guardrails 🛡️ The safety & validation layer for LLMs in Ruby. Prevents JSON errors,...	19	Experimental	6	Ruby
53	maro-style/SlangShield SlangShield: Because ‘hey bro’ doesn’t belong in Q4 stakeholder updates (IT/EN).	19	Experimental	—	JavaScript
54	Rami8612/signguard-ai Protect yourself and your team before signing — designed for multi-signature...	19	Experimental	—	JavaScript
55	zentinelproxy/zentinel-agent-ai-gateway AI Gateway agent for Zentinel proxy - prompt injection, PII, jailbreak detection	19	Experimental	—	Rust
56	christopherpaquin/Guardrails-AI Reliable AI Infrastructure: Input/Output validation, structured data...	19	Experimental	—	Python
57	voodooEntity/ghost_trap Multi‑surface anti‑scraping: GitHub Action appends a README trap;...	18	Experimental	3	JavaScript
58	new-world-coder/AgentShield Developer tool for detecting and testing loopholes in agentic AI applications	18	Experimental	3	JavaScript
59	maltyxx/guardix An autonomous Web Application Firewall (WAF) that uses a Large Language...	17	Experimental	2	Rust
60	DrPwner/PromptSniffer PromptSniffer is a security auditing tool designed for authorized...	16	Experimental	1	Python
61	urcuqui/PhishAwareBot PhishAwareBot is a web application that generates and displays two types of...	16	Experimental	7	Python
62	Nibir1/VaultSim An event-driven AI social engineering simulator. Execute prompt injection...	16	Experimental	—	Python
63	personal-vault/pvp Personal Vault Project	15	Experimental	17	PHP
64	GPierce9/Vault-Protocol-v2.6-Safer-AI-by-Design Vault Protocol is an integrated, trauma-informed safety architecture built...	15	Experimental	—	—
65	SH-Nihil-Mukkesh-25/Vaultify Vaultify is an intelligent security system combining ESP32 hardware with...	15	Experimental	—	C++
66	57karakalkan/MetaSafe-Guardian- 🛡️ Monitor virtual spaces with AI to detect harmful behavior, manage safe...	15	Experimental	1	JavaScript
67	darkmailr/darkmailr darkmailr - Offline Phishing Simulation Tool	15	Experimental	7	HTML
68	dakshaladia/lost-in-the-middle-prompt-injection Research study on context-window analysis of LLMs	15	Experimental	—	Python
69	arihantprasad07/guni AI agent security middleware — detect prompt injection, phishing and goal...	15	Experimental	1	HTML
70	JuanAirala/freedom 🔒 Protect your data with Freedom, an encrypted storage app using AES-256-GCM...	15	Experimental	1	HTML
71	ajutamangdev/PromptShield PromptShield is an open-source LLM firewall intended to inspect prompts for...	15	Experimental	1	Go
72	Zikodenomics/cipher-vault 🔐 Manage your passwords securely with Cipher Vault, a simple Python CLI that...	14	Experimental	—	Python
73	pavani-n-hash/guardianmesh AI agent security gateway that applies config‑driven rules and LLM‑based...	14	Experimental	—	JavaScript
74	Tearfullnex/SpecGuard 🛡️ Enforce AI behavior guidelines with SpecGuard, a tool that turns policies...	14	Experimental	—	Python
75	RahulR767/defender-acl-blocker 🔒 Block Microsoft Defender's user space with easy ACL adjustments to enhance...	14	Experimental	—	Go
76	rizkycsv/PromptGuard 🔒 Safeguard LLM behavior with PromptGuard to detect unseen regressions and...	14	Experimental	—	Python
77	vartulzeroshieldai/AIGuardX Loss prevention and policy enforcement for generative AI tools (ChatGPT,...	14	Experimental	—	—
78	shaqir/healthcare-ai-safety-demo- Healthcare AI safety demo — 7-layer architecture with prompt injection...	14	Experimental	—	JavaScript
79	shaqir/healthcare-ai-safety Healthcare AI safety system — 7-layer architecture with prompt injection...	14	Experimental	—	JavaScript
80	augustyatuhsexpeimentation/agentshield Security firewall for AI agents — block prompt injection, data exfiltration,...	14	Experimental	—	Python
81	satrijan/LLM-PROMPT-INJECTION-PAYLOAD-S 🛡️ Explore and test prompt injection techniques safely for AI applications,...	14	Experimental	—	—
82	mhsn1/ghostshield AI-powered LLM security scanner, real prompt injection attacks.	14	Experimental	—	HTML
83	montanaflynn/AdversarialBench Adversarial prompt-injection benchmark for LLMs	14	Experimental	—	TypeScript
84	kourgeorge/prompt-sentinel Python library designed to protect sensitive data when interacting with...	14	Experimental	3	Python
85	cybertechajju/LLM-PROMPT-INJECTION-PAYLOAD-S Unlock safe, high-signal prompt workflows for ethical hacking and AI red-teaming	13	Experimental	22	—
86	metawake/puppetry-detector Puppetry Detector is a modular engine for detecting structured and...	13	Experimental	—	Python
87	yashcanbuild/social_shield 🛡️ AI-powered email security platform that detects phishing attempts,...	12	Experimental	1	JavaScript
88	wwa/FIMjector FIMjector is an exploit for OpenAI GPT models based on Fill-In-the-Middle...	12	Experimental	4	—
89	rohilrg/CatchPromptInjection This repo focus on how to deal with prompt injection problem faced by LLMs	11	Experimental	2	Python
90	ilicitos/agent-creds 🔐 Inject API credentials securely into unmodified code via a transparent...	11	Experimental	—	—
91	juyterman1000/llm-safety Stop prompt injections in 20ms. The safety toolkit every LLM app needs. No...	11	Experimental	—	Python
92	miksto/danger-llm_validator Danger plugin for write rules in natural language, and let an LLM ensure...	10	Experimental	1	Ruby

Comparisons in this category

guardrails and circle-guard-bench (70 vs 37) guardrails and GuardBench (70 vs 39) guardrails and guardrails-lib-ai (70 vs 35)