Search

v0.1 policy-synthesis (descriptor-only) for MCP-server RCE defense: landlock/sandbox-exec/cgroups-v2 profile builder + CVE-replay predicates + canary tracker + append-only NDJSON audit log. v0.2 adds native enforcement + verified Acra-pattern audit-log si

Local static scanner for MCP setup, prompts, tool descriptions, and repo manifests.

Pin, diff, and lint MCP tool definitions — npm audit / SRI for MCP tools.

MCP Server Tool Inspector - connect, list tools, estimate token usage

Trust policy layer for AI agents, MCP servers, skills, and plugins.

security layer between your data and agent!

Secure Local LLM Swarm over MCP

Deny-by-default local stdio proxy and policy inspection CLI for MCP servers.

Sandbox runtime for secure JavaScript code execution

MCP Server Security Standard (MSSS): an open, testable security control standard for certifying MCP servers, with levels, evidence requirements, and reporting schemas.

In-A-Lign Agent Provenance & Security MCP Server

Conversational cloud security compliance auditing. MCP server enabling natural language queries about AWS security posture via Claude and other AI assistants.

Secure execution layer for LLM/MCP agents

Security scanner for Python MCP server code

https://github.com/eltociear/skill-audit-mcp

Local (stdio) MCP server for the Sublime Security platform API — analyze email messages, validate detection rules, run threat hunts, and perform security analysis.

The governance-native Agent OS. Cryptographically sealed 10-Law security, tri-model cognitive orchestration, and secure full-stack computer use with immutable CAC accountability.

Security scanner for MCP (Model Context Protocol) servers

Model Context Protocol (MCP) endpoint security auditor

Scan MCP servers for potential threats & security findings.