"Helix" matching MCP servers:

Provides three AI security tools for Claude: scanning MCP servers for security risks, validating HDP delegation chains against IETF standards, and triaging security alerts with kill-chain analysis.

Sits between your AI agents and their MCP servers and governs every tool call: policy rules, cross-server spend caps, human approval for risky actions, and a full audit trail. No changes to agent code or MCP servers. The tools listed here come from the bundled demo upstream in your own deployment Helio exposes whatever MCP servers you put behind it.

A reference MCP server that enforces identity-aware tool access using role-based permissions (Reader, Analyst, Auditor) and supports multiple identity routes like Entra RBAC, Copilot Studio, and Okta XAA. It provides read-only access to test evidence with explicit denial paths for unauthorized tools.