Submissions from instavm.io

		How Claude Code and Codex approach sandboxing (instavm.io)
		2 points by thepoet 11 days ago \| past \| discuss
		Hacking 'Skill' distilled from 4200 Hackerone vulns (instavm.io)
		2 points by mkagenius 48 days ago \| past
		Show HN: Design your own post-AGI civilization with tradeoffs (instavm.io)
		1 point by thepoet 4 months ago \| past
		Anthropic Skills Can Be Programmatically Optimized (Using DSPy) (instavm.io)
		2 points by mkagenius 4 months ago \| past
		Show HN: I built a mitmproxy AI agent using 4000 paid security disclosures (instavm.io)
		2 points by mkagenius 4 months ago \| past
		I Used 4,000 Paid Bug Bounty Reports to Build a CLI-First Security Agent (instavm.io)
		2 points by mkagenius 4 months ago \| past
		InstaPoT: Using InstaVM with DSPy's Program of Thought (instavm.io)
		2 points by mkagenius 5 months ago \| past
		Anti-patterns while working with LLMs (instavm.io)
		76 points by mkagenius 5 months ago \| past \| 24 comments
		I Want Everything Local – Building My Offline AI Workspace (incl. Claude Skills) (instavm.io)
		1 point by mkagenius 6 months ago \| past
		Claude Skills but Local (and Private) (instavm.io)
		1 point by mkagenius 6 months ago \| past
		I want everything local – Building my offline AI workspace (instavm.io)
		1143 points by mkagenius 8 months ago \| past \| 286 comments