agentic-security

Think before doing, prove before saving. Vestige is a local, deterministic safety kernel for AI agents. Every memory write and agent action is recorded in Strata, a signed, append-only log, so you can see exactly what your agent did and why.

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.

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.

Agentic AI EDR for developer workstations and autonomous agent swarms. Build Swarm Detection & Response platforms with Clawdstrike.

Open-source AI agent framework for cloud security across AWS, Azure, GCP and Kubernetes: AI-powered cloud detection and response, and cloud security posture management findings triage.

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

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.

Awesome LLM security tools, research, and documents

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

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

Portable security rules for the action boundary of AI agents

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

Run AI coding agents (Claude Code, opencode, Grok, ...) inside a bubblewrap sandbox; one wrapper, any agent.

A curated, structured, and continuously updated map of security risks, controls, benchmarks, architectures, and research for agentic, multi-agent, tool-using, self-improving AI systems. 🌟 Star if you like it!

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

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

The zero-trust framework for autonomous AI agents.

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

Ghostsplice repository: PoC for Cross-Channel Trust Fragmentation Attack

miniCISO is an open-source, multi-agent security staff that turns security context into evidence-driven work — with independent QA and bounded execution.