Plan docs over plan mode: Fable writes the spec, Opus executes, Fable validates
dotey · x · 2026-09-23
@dotey quotes @xicilion to share a practical multi-agent coding workflow: the key isn't plan mode, but a written spec document that fixes direction, carries context, and enables multi-agent collaboration.
@xicilion's four reasons for writing plans:
- Build consensus: one sentence can't capture a goal; details shape the architecture
- Sync context: the AI gathers info around the goal and exposes existing gaps
- Lock the path: AI drifts from the goal while solving subproblems, causing rework
- Record progress: working docs must be continuously updated
@dotey's concrete loop:
- Write a tech spec doc first; human reviews the overall direction
- Have Fable write the doc (with technical details and validation criteria), then hand it to Opus to implement
- After implementation, start a fresh session and have Fable validate against the doc
- Archive important docs and keep them updated
- For complex specs, use GPT-6 Pro in ChatGPT's web UI with repo access, then have Fable review and Opus execute
Related event: Developers Favor Spec Documents Over Plan Mode for AI Coding(4 posts)→
More from coding & agent
- Why AI-made dev tools beat one-shot game generation: freedom of process — eschadiol · 2026-09-24
- RealSense VP on AgenticROS: Letting AI Agents Directly Control Physical Robots — chrismatthieu · 2026-09-24
- AWS API Gateway's Hard 10MB Upload Limit and the Presigned URL Fix — _jaydeepkarale · 2026-09-24
- Open-source Pragma gives coding agents a terminal-first workspace with Git worktrees — tech_w0rld · 2026-09-24
- Dev builds dense task annotation system with GPT-6 Astra, ships it as an LLM skill — chris_j_paxton · 2026-09-24
- Jev-as-a-Judge: hybrid agent eval flow escalates low-confidence calls to frontier models — omarsar0 · 2026-09-24