mcpdirectory.dev
9,129 servers · updated 23 Sept

Helixar MCP server

community Active last commit 50 d
{
  "mcpServers": {
    "helixar": {
      "type": "http",
      "url": "https://mcp.helixar.ai/mcp"
    }
  }
}
file .mcp.json (in your project root)version 0.0.1

What it does

A remote MCP server (and optional local stdio server) that provides security tooling for AI agents: scanning MCP server manifests against detection rules, validating HDP delegation chains, and auditing software release artifacts. It is aimed at developers and security practitioners working with MCP-based agentic systems.

  • Scan an MCP server manifest by URL or raw JSON to detect issues such as missing authentication, destructive tools without confirmation guards, or prompt-injection phrasing in tool descriptions.
  • Validate an HDP (Helixar Delegation Protocol) delegation chain for scope escalations, depth violations, expired hops, and missing signatures against the IETF draft specification.
  • Audit software release artifacts in a dist directory for secrets, metadata leaks, and license gaps, or run a full hardening pipeline including obfuscation, signing, and attestation.

Tools this server exposes

ToolDoes
helixar_inspect_mcpScans an MCP server URL or manifest JSON against Sentinel detection rules, returning a risk score, findings, and a security brief; quick mode checks 8 rules wit
helixar_hdp_validateValidates an HDP delegation chain against the IETF draft spec, surfacing scope escalations, depth violations, expired hops, and missing signatures with citation
helixar_releaseguardWraps the ReleaseGuard CLI binary to scan release artifacts for secrets, metadata leaks, and license gaps, or to run a full harden pipeline; available only when

Before you enable it

An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.