agentic-security 开源项目推荐（89 个）

samvallad33 /vestige

先思考再行动，先证明再保存。Vestige 是面向 AI 智能体的本地、确定性安全内核。每一次记忆写入和智能体操作都会记录在 Strata——一个仅追加日志中，因此你可以确切看到你的智能体做了什么以及为什么。

secureagentics /Adrian

Open-source runtime AI agent security tool - monitors and controls AI agents, catching malicious tool use, prompt injection, and policy drift in real time, before the agent acts.

RiccardoBiosas /awesome-MLSecOps

A curated list of MLSecOps tools and resources for securing machine learning and AI systems - adversarial ML defense, LLM security, AI red teaming, model scanning, supply-chain protection, and MLOps pipeline security.

backbay-labs /clawdstrike

开发者工作站和自主代理车队的AI电子数据管理。用爪击构建群聚侦测与响应平台。

cynative /cynative

Open-source AI agent framework for security engineering across AWS, GCP, Azure, Kubernetes, GitHub & GitLab: AI-powered security review, findings triage, detection and response.

thomasunise /argus

Open-source agentic security operations platform:

davidmatousek /tachi

Threat modeling and AI-reasoning vulnerability detection harness for Claude Code — STRIDE + AI + MAESTRO

beyefendi /awesome-llm-security

Agentic security, LLM security tools, research, and documents

Keyvanhardani /mythos-research

Outside-in replication of Anthropic's Mythos Preview / Project Glasswing — open-source agentic vulnerability-discovery scaffold on Claude Opus 4.7. Eight-phase sink-guided pipeline, ~$1/run, OSS self-scan and coordinated disclosure.

efij /awesome-claude-code-security

A awesome curated list of security resources, hardening tools, threat research, and governance frameworks specifically for Claude Code

ClawSecure /clawsecure-openclaw-security

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

didvc /ai-bwrap

Better agentic security inside a sandbox; one wrapper, any agent.

backbay-labs /hush

Portable security rules for the action boundary of AI agents

trnt-ai /trent-openclaw-security-assessment

Free security assessment for your OpenClaw 🦞 environment. Scans gateway config, tool permissions, MCP servers, plugins, and chained attack paths.

natnew /Awesome-Agentic-AI-Security

该项目在 GitHub 上没有填写描述

myProjectsRavi /sentinel-protocol

Open-source AI security firewall. 81 engines for PII detection, prompt injection defense, MCP security, and egress classification. Local-first. Zero cloud dependency.

pigeonlabs-ai /D2-Python

Detect and Deny - Deterministic Function-Level Guardrails for AI Agents

calebfaruki /sycophant

The zero-trust framework for autonomous AI agents.

efij /AgentDFIR

Open-source digital forensics and incident response for AI agents (Claude Code, Codex, Cursor, Gemini, Copilot…). One command: detect → collect → analyze → browse.

asset-group /ghostsplice

Ghostsplice repository: PoC for Cross-Channel Trust Fragmentation Attack