tool-security

Activation-probe security scanner for AI agent tooling. Reads a model's internal activations to detect poisoned MCP servers, skills, and packages before install.

Static pre-production security scanner for AI systems, covering models, datasets, RAG, prompts, agent tools, MCP, and AI supply-chain risk.

Risk-aware approval layer for high-risk AI Agent tool calls

Agent Runtime Security Platform, Open-source zero-trust security and governance layer for autonomous AI agents featuring runtime policy enforcement, DLP scanning, MCP tool verification, and distributed tracing.

Native Elixir security runtime for MCP and agent-tool boundaries: policy gating, credential scanning, attestations, trust bundles, and audit trails.

Practical MCP readiness, security and stateless-migration resources for teams moving MCP servers toward enterprise deployment.

Adaptive threat intelligence for AI agent security — semantic memory, multi-turn escalation, output scanning, rate limiting, and prompt hardening.

Local deterministic intake CLI for statically inspecting MCP server package surfaces before agent admission.

Defense-in-depth security middleware for LLM and agent pipelines—prompt-injection defense, provenance, tool mediation, PII protection, and observe-before-enforce policies.

Free MCP security checks, examples and developer tools for understanding MCP configuration, tool exposure, change and runtime trust posture.

OT-aware authorization gateway between an LLM agent and an industrial OPC UA server — a process-aware policy layer (value ranges, rate limits, interlocks) with human approval and audit.

Local firewall for MCP tool calls, descriptors, and results.

Official ECZ-ID MCP Trust product and documentation hub. Local-first MCP configuration and change inspection for VS Code.