I route Claude Code steps by decision type, not difficulty

说明与用途

按决策类型而非难度路由 Claude Code 步骤，将 57% 输出转移到更便宜的模型，提升会话输出 60%，同时保持顶级模型使用不变。

问题

Choosing the right Claude model for each code step was inefficient; hard‑task → strong‑model mapping did not improve output distribution.

结果

Routing by decision type moved majority of output to cheaper models, increased session output by ~60%, and kept top‑model output constant, improving efficiency without higher cost.

工作流

• 1. Identify decision type of each code step (already decided, local choice, dependent on discovery, verdict before merge).
• 2. Route step to appropriate model: Haiku for already decided, Sonnet for local choice, Opus for discovery‑dependent, Opus reviewer for verdict.
• 3. Execute step in subagent with dedicated Claude instance and context.
• 4. Return short result to main session for validation.
• 5. Main session accepts result or requests changes; merges/pushes require user approval.
• 6. Use plugin "maddog" in advisor‑mode to automate routing and validation.

证据

69% of the output came from the top models (Opus, Fable)

来源

Hard task → strong model didn’t fix it. Difficulty is the wrong signal

来源

57% of output ran on Haiku or Sonnet, against 31% without it.

来源

Output per session rose about 60%, while top-model output stayed flat (~124K tokens both).

来源