GitHub topics

llm-security

Mostly in Security Tools 27%, LLM Infrastructure 16%

Repositories

18

Total stars

8,026

Trending

xalgorix

Open-source AI penetration testing agent that exploit-verifies every finding. Autonomous 22-phase pentests of web apps & APIs with CVSS-scored PDF reports. Self-hosted, bring your own LLM, or run fully local with Ollama.

Created Mar 2026

ai-hacking ai-penetration-testing ai-pentesting ai-security api-security
Total 1,375
30 days +162
aashish254

Runtime permission firewall for AI agents - YAML policies, one-tap approvals (macOS popup/Telegram), tamper-evident signed audit log. Benchmarked: 103/103 attacks blocked, 8/8 normal dev ops allowed.

Created Sep 2026

agent-security ai-agents audit-log claude-code go
Total 83
30 days +73
pipelab-org

Open-source AI agent firewall for MCP security and agent egress. Scans mediated HTTP, MCP, A2A, and WebSocket traffic for exfiltration, SSRF, and prompt injection, and emits mediator-signed action receipts: verifiable audit evidence from outside the agent.

Created Feb 2026

agent-security ai-agent-security ai-agents ai-firewall ai-security
Total 1,008
30 days +72
beelzebub-labs

A secure low code deception runtime framework, leveraging AI for System Virtualization.

Created May 2022

acis agentic-ai-security cloudnative cloudsecurity cybersecurity
Total 2,257
30 days +57
coipond

Give the agent a machine. Just not yours. Each AI coding agent gets its own isolated machine with root, Docker, and systemd - active defense detects and stops threats automatically.

Created Jan 2026

agentic-ai ai-tools anthropic claude claude-code
Total 791
30 days +33
dataiku

Privacy proxy for your OpenAI requests

Created Oct 2025

ai-gateway data-anonymization data-masking llm-security pii-detection
Total 456
30 days +17
Fausto-404

AlterHive(幻巢)是一款面向攻防演练、红蓝对抗和安全研究场景的高交互智能欺骗蜜罐平台。平台以虚拟拓扑、会话记忆、规则引擎和多 Agent 欺骗规划为核心,将攻击者/渗透Agent引导进可控的“子网幻象”环境中,帮助防守方观察攻击意图、拖延攻击节奏、保护真实目标。

Created Jun 2026

adaptive-honeypot agent-security ai-security blue-team honeypot
Total 166
30 days +12
adithyan-ak

Offensive security framework for AI agent infrastructure - recon, credential looting, model exfiltration, poisoning, and attack-path analysis across MCP, A2A, gateways, and AI services. BloodHound for the agentic stack.

Created Apr 2026

a2a agentic-ai ai-agents ai-security attack-paths
Total 464
30 days +11
openclaw

Composable security scanning harness for agent skills

Created Jun 2026

agent-security ai-security llm-security prompt-injection security-scanner
Total 156
30 days +10
openziti

Zero trust LLM gateway. OpenAI-compatible proxy with semantic routing and load balancing across OpenAI, Anthropic, Ollama, vLLM, and any compatible backend. Identity-based access, virtual API keys, and end-to-end encryption via OpenZiti

Created Feb 2026

agentic-ai ai-gateway ai-infrastructure ai-security anthropic
Total 109
30 days +9

Top

beelzebub-labs

A secure low code deception runtime framework, leveraging AI for System Virtualization.

Created May 2022

acis agentic-ai-security cloudnative cloudsecurity cybersecurity
Total 2,257
xalgorix

Open-source AI penetration testing agent that exploit-verifies every finding. Autonomous 22-phase pentests of web apps & APIs with CVSS-scored PDF reports. Self-hosted, bring your own LLM, or run fully local with Ollama.

Created Mar 2026

ai-hacking ai-penetration-testing ai-pentesting ai-security api-security
Total 1,375
pipelab-org

Open-source AI agent firewall for MCP security and agent egress. Scans mediated HTTP, MCP, A2A, and WebSocket traffic for exfiltration, SSRF, and prompt injection, and emits mediator-signed action receipts: verifiable audit evidence from outside the agent.

Created Feb 2026

agent-security ai-agent-security ai-agents ai-firewall ai-security
Total 1,008
coipond

Give the agent a machine. Just not yours. Each AI coding agent gets its own isolated machine with root, Docker, and systemd - active defense detects and stops threats automatically.

Created Jan 2026

agentic-ai ai-tools anthropic claude claude-code
Total 791
adithyan-ak

Offensive security framework for AI agent infrastructure - recon, credential looting, model exfiltration, poisoning, and attack-path analysis across MCP, A2A, gateways, and AI services. BloodHound for the agentic stack.

Created Apr 2026

a2a agentic-ai ai-agents ai-security attack-paths
Total 464
dataiku

Privacy proxy for your OpenAI requests

Created Oct 2025

ai-gateway data-anonymization data-masking llm-security pii-detection
Total 456
packyme

LLM privacy gateway in Go — millisecond-latency PII and secret redaction. Used in production by PackyCode.

Created May 2026

ai-gateway data-redaction gitleaks golang grpc
Total 342
praetorian-inc

Simple LLM service identification - translate IP:Port to Ollama, vLLM, LiteLLM, or 60+ other AI services in seconds

Created Jan 2026

ai-security attack-surface capability golang llm-evaluation
Total 255
Fausto-404

AlterHive(幻巢)是一款面向攻防演练、红蓝对抗和安全研究场景的高交互智能欺骗蜜罐平台。平台以虚拟拓扑、会话记忆、规则引擎和多 Agent 欺骗规划为核心,将攻击者/渗透Agent引导进可控的“子网幻象”环境中,帮助防守方观察攻击意图、拖延攻击节奏、保护真实目标。

Created Jun 2026

adaptive-honeypot agent-security ai-security blue-team honeypot
Total 166
openclaw

Composable security scanning harness for agent skills

Created Jun 2026

agent-security ai-security llm-security prompt-injection security-scanner
Total 156
makalin

SecureMCP is a security auditing tool designed to detect vulnerabilities and misconfigurations in applications using the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction). It proactively identifies threats like OAuth token leakage, prompt injection vulnerabilities, rogue MCP servers, and tool poisoning attacks.

Created Jun 2025

ai-security cli-tool cli-tools golang llm-security
Total 139
SleuthCo

Security proxy for AI agents. Scans every message for prompt injection, PII, and secrets. Defense-in-depth: Go proxy + iptables firewall + eBPF kernel monitor. YAML policy engine, audit logging, 5 AI agents with RAG knowledge bases.

Created Mar 2026

ai-agents ai-security audit-logging defense-in-depth ebpf
Total 132
DmitrL-dev

Free, open tools to learn how AI apps get attacked, test a guard on your own data and check what a security test result can prove. EN/RU academy, Guard Lab, eval-bounds. Live protection: spectorn.ai

Created Dec 2025

adversarial-attacks agentic-ai-security ai-firewall ai-red-team ai-safety
Total 114
openziti

Zero trust LLM gateway. OpenAI-compatible proxy with semantic routing and load balancing across OpenAI, Anthropic, Ollama, vLLM, and any compatible backend. Identity-based access, virtual API keys, and end-to-end encryption via OpenZiti

Created Feb 2026

agentic-ai ai-gateway ai-infrastructure ai-security anthropic
Total 109
aashish254

Runtime permission firewall for AI agents - YAML policies, one-tap approvals (macOS popup/Telegram), tamper-evident signed audit log. Benchmarked: 103/103 attacks blocked, 8/8 normal dev ops allowed.

Created Sep 2026

agent-security ai-agents audit-log claude-code go
Total 83
getaxonflow

AxonFlow: Runtime control layer for production AI

Created Dec 2025

agent-runtime agentic-ai ai-control-plane ai-governance ai-observability
Total 72
Mindburn-Labs

Fail-closed execution firewall for AI agents: quarantine MCP tools, proxy OpenAI-compatible requests, emit signed receipts, and verify EvidencePacks offline.

Created Feb 2026

agent-security ai-agents ai-security cli developer-tools
Total 57
HuTa0kj

Vetix — Automated scanning, identification, and assessment of SKILL security risks, based on Golang.

Created May 2026

agent agent-skills agents claude-code eino
Total 50

Related

All topics →