Your AI agent can run commands. It should not get a free pass.
This is a dev post classified by Jev as Security (a tool drop), kept by the Dev Radar because it carries real work, not commentary.
Your AI agent can run commands. It should not get a free pass. HOL Guard is an open-source, local-first security layer for builders running AI agents, tools, plugins, skills, MCP servers, and package installs. It helps you put a review step in front of risky agent activity by checking supported shell commands, file access, package operations, and MCP tool calls before they run, then applying your policy. Key features: • Pre-execution reviews – checks supported agent actions before they run • Policy-led decisions – allows, blocks, or requests approval based on your policy • Broad protection
Posted by Dan Kornas (99.1k followers) 1 h ago · 7 likes · 655 views · view the original post on X. Kept by the Dev Radar as Security.
More dev work like this
- Digital forensics investigations can stall in the handoffs. This repo keeps the evidence… — @DanKornas
- Pentest is not a workflow, It is a search problem. — @0x0SojalSec
- I just built the open-source version of this — @imarikchakma
- Security teams shouldn't have to wait for a critical vulnerability to trigger an… — @thenewstack
- Cloudflare just turned AI agents into SECURITY AUDITORS. — @tonysimons_
- Wow, these slides are fantastic to just read through and contemplate. 😍… — @zooko
- How hackers use SQL Injection to compromise a website — @Hamzaonchain
- Dostlar selamlar, globaldeki en büyük Cybersecurity SFT finetuning datasetlerinden olan… — @AlicanKiraz0
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 15.3k posts from 4.8k X accounts over the last 21 days, 1.7k tools, 12 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-20 10:05 UTC. Full method.