When agent-built CLIs are too opaque, humans just let the agent run them
craigbalding · x · 2026-11-20
Security researcher craigbalding observes a design tension: when an agent-generated CLI demands three long opaque identifiers, human laziness kicks in — users skip comprehension and tell the agent to "get this working", letting agents execute their own creations. His question: who are you designing for, humans or agents?
More from coding & agent
- Kaho AI learning tool update: notes that auto-draw diagrams, page-side Q&A — abhi9u · 2026-11-20
- Microsoft's free 18-lesson AI agents course hits 80.5k GitHub stars — victor_explore · 2026-11-20
- Self-hacking with abliterated models: biggest risks were infra side channels, not app code — cyb3rops · 2026-11-20
- Pocket City: a pixel-art Amsterdam canal simulation built entirely on real geodata — anselm · 2026-11-20
- Mnemos vs Claude Code memory test details: 65%/41%/42% recall across three blind tests — RileyRalmuto · 2026-11-20
- Solo-built Mnemos memory system recalls 2x more facts than Claude Code's native memory — RileyRalmuto · 2026-11-20