MCP Directory
Back

helm-ai-kernel

by Mindburn-Labs · Go · ★ 75

Fail-closed execution firewall for AI agents: quarantine MCP tools, proxy OpenAI-compatible requests, emit signed receipts, and verify EvidencePacks offline.

#agent-security#ai-agents#ai-security#cli#developer-tools#devsecops#evidencepack#execution-boundary#go#llm-security#mcp#model-context-protocol#offline-verification#openai-compatible#policy-engine#proofgraph#sandbox#self-hosted#signed-receipts#zero-trust

Install

go install github.com/Mindburn-Labs/helm-ai-kernel@latest

Auto-generated from the repo's primary language — check the repo's own README for its exact install instructions.

Claude Desktop config

A starting point for your claude_desktop_config.json (auto-generated — the repo's README may specify a published package name or required env vars):

{
  "mcpServers": {
    "helm-ai-kernel": {
      "command": "npx",
      "args": [
        "-y",
        "github:Mindburn-Labs/helm-ai-kernel"
      ]
    }
  }
}

From the README

**A local firewall for AI-agent actions.** HELM sits between Claude Code, Codex, MCP tools, shell commands, and other agent actions. It decides , , or , then writes a signed receipt you can verify later. Ask your agent to do something risky. HELM blocks or escalates the action before it runs, then records the decision. No cloud account. No model key. No Docker. No production credentials. | Agent tries to... | HELM does this | Proof | | --- | --- | --- | | Run a destructive shell command | | signed receipt | | Use an unknown MCP tool | | quarantine record | | Read protected secrets | | f…
Read full README on GitHub →

Related ai & ml servers