Skills
Distill anyone — colleagues, partners, or even fictional characters — into an AI Skill that thinks and speaks like them, trained from your chat history and descriptions.
View details → 
titanwings
hol-guard
Skills 498 stars Resource
Open-source antivirus for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and skills at runtime.
Project details
GitHub stars 498
LicenseApache-2.0
Primary languagePython
Last push2026-08-29T19:23:18Z
Maintainerhashgraph-online
Added to this directory2026-08-21
Screenshots
About this project
Open-source antivirus for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and skills at runtime.
More about this project
Most security tools see only one part of an AI agent's attack surface. Code scanners run after files change. Sandboxes constrain a process but do not understand agent intent. MCP gateways see MCP traffic but not local shell commands, package installs, skills, hooks, or agent configuration.
Project FAQ: hol-guard
What is hol-guard?
hol-guard is a DeepSeek Harness ecosystem resource maintained by hashgraph-online. Open-source antivirus for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and skills at runtime.
Where can I get hol-guard?
Source code and usage instructions are available at https://github.com/hashgraph-online/hol-guard. Follow the repository README for the correct setup steps.
Is hol-guard free?
hol-guard is a community open-source project released under the Apache-2.0 license. Review the repository license and documentation before use.
Related apps and resources
Skills
An ADHD behavioral coaching skill for DeepSeek Harness — guides you through tasks with gentle, actionable structure.
View details → 
zimai233
Skills
Portable agent skills for building, publishing, and remixing games — usable from dsh, Claude Code, and Codex.
View details → 
paean-ai