Agents inside apps vs. operating human tools: a short debate
andpoul · x · 2026-09-25
Developer andpoul debates bartonsmith on what "designed for agents" means: an app exposing a plugin/scripting API for external agents, or embedding agents in-app. He's bullish on the former, arguing in-app agents have never beaten an agent operating the human's own tool, while conceding simultaneous use remains unsolved; the other side notes prompt-box-only tasks will still thrive for non-professional needs.
Related event: Debate: should agents be built into apps or operate them from outside(2 posts)→
More from coding & agent
- ChatGPT bug: final prompt at context limit silently overwrites previous state, corrupting handoffs — xtraa · 2026-09-25
- Link launches financial insights letting AI agents personalize via your transaction history — jeff_weinstein · 2026-09-25
- Google Cloud Unveils AlloyDB's Agentic Database Architecture for AI Workloads — rseroter · 2026-09-25
- Opus 5.5 recreates an AI-generated video as a playable 90s fantasy walking sim — Successful-Earth678 · 2026-09-25
- Anthropic's CI job volume grew 25x in six months — here's how they scaled test selection — JeremyCMorgan · 2026-09-25
- Perplexity's Fast Search powers Hermes Agent with 160ms p50 latency, free for all tiers — denisyarats · 2026-09-25