hol-guard

Open-source antivirus and firewall for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and skills at ru

What is hol-guard?

Open-source antivirus and firewall for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and skills at runtime.

Overview of hol-guard

What it is

HOL Guard is an open-source antivirus for AI agents. It reviews agent actions before they run, including shell commands, file access, package installs, and MCP tool calls. It detects secret exposure, destructive operations, prompt injection, and supply-chain risks, then allows, blocks, or requests approval according to policy. It runs locally without an account and provides a CLI and local dashboard. Optional Guard Cloud adds shared history, team policy, and fleet management.

Key features

  • Reviews shell commands and file access for destructive operations, sensitive file access, credential exposure, and suspicious outbound commands.
  • Evaluates package-manager operations against supply-chain intelligence before installation.
  • Inventories local artifacts and reviews new or changed plugins, skills, and agent configuration before launch.
  • Inspects MCP server configuration and reviews tool calls through supported hooks and managed proxies.
  • Screens supported events for prompt injection and sensitive content.
  • Routes decisions to native prompts or the approval center and records local receipts for review.
  • Includes a plugin scanner CLI for security and quality checks before publishing agent plugins, skills, and MCP integrations.

Use cases

  • Protecting AI agents (Codex, Claude Code, GitHub Copilot CLI, Cursor, Cline, Gemini CLI, Grok, Hermes, Kimi Code, Pi, oh-my-pi, OpenClaw, OpenCode, Antigravity, ZCode, Devin) from risky actions at runtime.
  • Blocking secret exposure, destructive file operations, prompt injection, and malicious package installs.
  • Scanning workspace dependencies for supply-chain risks.
  • Inspecting and verifying agent plugins, skills, and MCP servers before publishing or launching.
  • Managing approvals and reviewing decision history locally or with optional cloud sync.

Getting started

Requires Python 3.10 or newer and pipx.bashpipx install hol-guardhol-guard initCheck installation:bashhol-guard --versionhol-guard statusTo update:bashhol-guard updateFor the plugin scanner:bashpipx install plugin-scannerplugin-scanner scan .

License and pricing

License: Apache-2.0. Pricing: open source (no stated cost).

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/hashgraph-onlinehol-guard.svg)](https://yunodb.com/tools/hashgraph-onlinehol-guard?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/hashgraph-onlinehol-guard?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/hashgraph-onlinehol-guard.svg" alt="Featured on YunoDB" height="20"></a>