hol-guard is a security scanner and agent guardrail harness developed by Hashgraph Online. Built to safeguard agentic runtime environments and developer workflows, it detects vulnerabilities, validates plugins and tools, and enforces runtime safety boundaries for AI coding agents and automated execution harnesses.
Combined Scanner & Guard CLI — Integrates dynamic guard monitoring and static scanner checks within a single command-line binary interface. - Agent Skill & Harness Integration — Features out-of-the-box integration definitions (such as Droid skills and headless runners) for auto-invoking security scanning inside agent orchestrators. - Flexible Output Formats — Supports standard terminal outputs alongside structured formats like JSON (--format json) for automated CI/CD and agent consumption. - Subcommand Auto-Prefixing — Intelligently resolves command execution across security scan suites and runtime detection directives. - ClusterFuzzLite Support — Includes integrated fuzzing targets for continuous security auditing and vulnerability mitigation.
Use cases
Use Case: Securing Agent Tool & Plugin Execution Developers running local agent orchestrators can invoke hol-guard to validate MCP plugins, third-party skills, and agent tools before execution, preventing unverified or dangerous payloads from executing.
Use Case: Automated CI/CD Agent Inspection CI/CD pipelines can execute hol-guard scan --format json to automatically inspect repository-level agent configurations, ensuring security standards and boundary policies are respected prior to deployment.
Use Case: Headless Agent Monitoring Teams orchestrating autonomous agent swarms can delegate runner security tasks to hol-guard in headless mode to detect runtime rule breaches and unsafe filesystem operations.
Who it is for
hol-guard is designed for security engineers, platform teams, and AI tool developers looking to establish safety guardrails and scan surfaces around AI coding agents and plugins. It is not intended for end users looking for a standalone chat or conversational code-generation assistant.
Install guide
Installation Install hol-guard via the project repository or standard package management: bash git clone https://github.com/hashgraph-online/hol-guard.git cd hol-guard
First Run Verify the CLI binary setup and list available combined mode subcommands: bash hol-guard --help
Example Session Run a vulnerability scan against your agent workspace and output the results in JSON format: bash hol-guard scan --format json hol-guard detect
ActPlane — ActPlane is an open-source agent execution and policy enforcement harness developed by the Eunomia-bpf organization. It functions as a security…
Google AX — Google AX is an open-source agent harness and execution orchestrator developed by Google. It provides runtime infrastructure and execution sandboxing…
Skillreaper — Skillreaper is an open-source skill lifecycle management and auditing tool for AI agent ecosystems, developed by thousandflowers. It analyzes agent…
Agent Island — Agent Island is an open-source agent orchestration harness created by developer tristan666666 on GitHub. Designed as a framework for running…
Stratless — Stratless is an open-source agent harness and execution orchestrator developed by the Stratless AI team. It provides a modular infrastructure…
NemoClaw — NemoClaw is an open-source agent orchestration harness and execution framework developed by NVIDIA. It provides structured runtimes, modular skill dispatching,…