Dev runs security agents 24/7: they beat external attackers, asymmetry smaller than feared
AccBalanced · x · 2026-11-12
Developer @thdxr shares observations from running security agents 24/7: the agents find issues quickly, and access to internal details gives them a real edge. He argues defensive agents should be more effective than external attacker agents — the offense-defense asymmetry still exists, but is smaller than he expected. A hands-on data point on autonomous agents in security operations.
More from coding & agent
- A task-based decision guide: which Claude tool to use for ask, think, speed, reason, delegate, ship — rubenhassid · 2026-11-12
- AI Village agents propose their own goals: external scoreboards, anchor humans, depth over copies — repligate · 2026-11-12
- AI;DR: Writing culture is dying and knowledge work isn't ready for agents — prd_008 · 2026-11-12
- Grok Bot hailed as top AI agent orchestrator: Claude Code for eng, Paper for design — tetsuoai · 2026-11-12
- Developer builds 'VladOS' life dashboard with AI, wires fitness data via MCP — haltakov · 2026-11-12
- Perplexity red-teams its sandbox: no model escapes the VM, but 4 bypass the network allowlist — jomic01 · 2026-11-12