Astra-6.2 'lazy' backlash blamed on user education: default manual mode vs goal mode
alexisgallagher · x · 2026-11-22
- A quoted post defends Astra-6.2 against accusations of laziness: the model defaults to a precise "scalpel" manual mode with clear action boundaries, and users complaining it won't act simply haven't switched to plan/goal auto mode.
- The author argues the real issue is insufficient user education and urges lead @thsottiaux to leverage his reach to teach users, or risk losing them.
- Reposter alexisgallagher agrees with the appreciation and suggests people RTFM, admitting he hasn't done so himself.
More from coding & agent
- Indie dev ships SteerCamp, an 'OS for the post-AI era' workspace — pramodk73 · 2026-11-22
- FrankenRobotics: an AI-agent-run robotics lab that batch-designs 256 robots per GPU run — doodlestein · 2026-11-22
- AI short film 'The Wow Lunch': Midjourney images animated by Seedance 2.5 via Claude MCP — azed_ai · 2026-11-22
- Muse Code CLI vs AFK Pilot vs T3 Code: a feature-by-feature comparison chart — PawelHuryn · 2026-11-22
- Open-source llm-mock server fakes OpenAI, Anthropic, Gemini and Azure APIs so CI tests never need a real key — OkSkirt1039 · 2026-11-22
- David Hoang predicts local models will be the developer obsession this holiday — davidhoang · 2026-11-22