astra thrives on context: minimal prompting massively underperforms, dev finds
brandon_galang · x · 2026-09-12
Developer brandongalang shares hands-on findings about astra: its in-context learning is exceptional, adapting behavior wildly to what's in its context window. Two key implications: it responds very differently to agents.md and skill files, and — counterintuitively — deliberately constraining context actually hurts, as it hyperfocuses on whatever gets in. Unlike models that degrade with fuller contexts, astra benefits dramatically and was even trained to leave itself notes across compacting windows. Minimal prompters will underperform; assemble and curate context before execution instead.
More from coding & agent
- Sentry CEO Dropped Claude Months Ago, Slams Overcomplicated Coding Harnesses After Hitting Codex Limits — zeeg · 2026-09-12
- LangChain explains agent harnesses in under 90 seconds — LangChain · 2026-09-12
- Sentry CEO: Smarter Models Now Overshoot Simple Tasks — I Want Fewer Mistakes, Not More Initiative — zeeg · 2026-09-12
- Fable 5.1 agent demo shows self-onboarding and lifelong learning at work — ysu_nlp · 2026-09-12
- Vyact: Open-Source Desktop App Builds Reviewable Workflows Around Local LLMs — vyact · 2026-09-12
- Runway launches MCP to generate images and videos inside ChatGPT, Claude and Cursor — runwayml · 2026-09-12