Capability-based agent runtime with fine-grained policies . Brokering access directly within the agent's operating context, with zero setup and zero latency

Open-source AI agent firewall for MCP security: agent egress control, DLP, SSRF, and prompt injection defense.

Universal adapter between AI agents and model providers

The security agent that fights back. Watches your Linux server from inside, detects threats with kernel-level eBPF, and stops them with on-device AI. Open-source, self-hosted, dry-run by default. Apache-2.0.

Offline security scanner for AI-agent repos, skills, plugins, and MCP servers.

Free OpenClaw security scanner. 3,000+ agents audited. 3-Layer Audit Protocol. OWASP ASI 10/10 coverage. AI agent integrity layer.

A curated timeline of real AI agent security incidents, breaches, and vulnerabilities (2024-2026). Every entry sourced and dated.

AI got hands. This is the leash. Policy, audit, kill switch for any AI agent with access to your accounts.

LLM guardrails & prompt injection detection for Python. Auto-instruments LangChain, CrewAI, OpenAI, LiteLLM + 8 more frameworks. PII masking, toxicity detection, policy CI/CD. One line, zero code changes.

25 production-tested defensive security skills for Claude Code - WordPress, VPS, Cloudflare, Next.js hardening, AI agent guardrails, MCP security, prompt injection defense, OWASP LLM Top 10, LLM coding failure modes (slopsquatting, hallucinated APIs, sycophancy), incident response, GDPR/DACH compliance. MIT, battle-tested.

Security scanner for MCP-connected AI agent pipelines — 206 rules, 66 detectors, OWASP Agentic Top 10 + MCP Top 10, EU AI Act / SOC 2 / ISO 27001 / HIPAA compliance mapping. v0.3.24.

Security scanner for AI agent plugins, skills, MCPs, and configurations. Detects prompt injection, malware, credential theft, and obfuscated payloads.

GitHub Action + CLI to detect malicious patterns in MCP servers, AI agent skills, and plugins. 68+ attack signatures.

Official SupraWall MCP (Model Context Protocol) Security Plugin. Enforces deterministic guardrails, least-privilege tool access, and PII interception for AI agents.

Security proxy for AI agents (OpenClaw, LangChain, CrewAI). Protects API keys, enforces spend limits, blocks dangerous endpoints. Local-first Rust binary — your credentials never leave your machine.

AI agent runtime governance control plane: intercept tool calls with PII protection, approvals, and formal verification.

Stop AI agents before they do damage — behavioral sequence detection blocks credential exfiltration, config harvesting, and secret leakage before the network call fires

OpenPAKT: Open Protocol for Agent Knowledge Trust — a language-agnostic specification for AI agent security findings, scenarios, and CI policy gating.

Portable runtime policy and audit layer for AI agents - HTTP/HTTPS proxy enforcing egress policies, inspecting content, materializing secrets, and recording every decision.

🧠 C.A.S.E. Hybrid Cloud+Local AI Framework ── 雲端 AI 出計畫，本地 AI 執行苦勞。Llama.cpp 部署實戰指南，資料物理隔離，讓 Token 花在刀口上、不受 Quota 所困。