宝玉 on doc-first multi-agent coding: write a spec, review, then let Opus execute
ciaotigre · x · 2026-09-22
宝玉 shares his doc-first AI coding workflow instead of plan mode: a human reviews the spec for direction, Fable writes the doc with technical details and validation steps, Opus executes, then Fable verifies against the doc in a fresh session. He also sometimes uses GPT-6 Pro in ChatGPT to write specs from the repo, mixes Claude Code with Codex, and keeps key docs updated. His point: the value lies not in plan mode but in a human-confirmable doc that carries context across multiple agents.
Related event: Developers Share Doc-Driven Multi-Agent Coding Workflows(3 posts)→
More from coding & agent
- Stolen auth. Drains $200 Codex Subscription; User Scrambles to Revoke Access — _Toni_O · 2026-09-23
- Obsidian Starter Kit v4 goes AI-native with MCP server, 375 skills and memory agents — dSebastien · 2026-09-23
- Open-source Codex orchestrator skill runs GPT-6 Astra as mastermind over Sol and Luna agents — BLUECOW009 · 2026-09-23
- Decision-only model Jev beats frontier Gemini on 1,759 real decisions: 5x faster, 25x cheaper — Humble_News_6994 · 2026-09-23
- Deciding which AI agents earn their keep in production: five measurement approaches compared — felix_baron · 2026-09-23
- ftouter: an open-source Python router that auto-falls-back when free-tier LLM APIs die — Puzzleheaded_Bus925 · 2026-09-23