Tuned Opus-style AGENTS.md makes DeepSeek v4.1 actually finish coding tasks
teortaxesTex · x · 2026-09-12
- Developer @g023dev tested DeepSeek v4.1 on building a landing page plus a small game, with and without his Opus-tuned AGENTSV2.md prompt file.
- With the tuned AGENTS.md, the model became a "completionist" — the game ran and looked good. Without it, the model failed to finish the game.
- @teortaxesTex calls it clever: rather than asking the model to role-play a stronger one, you embed distilled Opus-like behavior rules into agents.md — effectively resurrecting distillation via prompt engineering.
- Useful reference for anyone tuning agent instruction files.
More from coding & agent
- Agents need explicit explain/propose/execute modes: 'how would you' is a proposal, not permission — OriginalHospital · 2026-09-12
- Scaling parallel Claude Code agents: the bottleneck is now human oversight, not the model — tylerjdunn · 2026-09-12
- Dev burns entire Codex quota in under 56 minutes after reset — chriscarm · 2026-09-12
- Pipecat v1.9 adds Meta's Muse Voice Transcribe, the lowest semantic-WER STT model tested — solyarisoftware · 2026-09-12
- Jensen Huang: Even AGI Won't Auto-Boost Company Productivity — You Need an Agent Harness — 大模型之路 · 2026-09-12
- How to Build a 'Company Brain' That Gets Smarter Every Week — Roger_M_Taylor · 2026-09-12