dsh-thinking-levels

DSH host plugin: thinking-level (reasoning_effort) control for dsh — auto-adjust per tool round, with optional manual lock and per-tool timing telemetry.

AI 分析

自动或手动控制模型思考级别，并提供工具调用耗时统计。适合使用支持推理强度的模型（如 DeepSeek-R1），希望在效果与响应速度/Token成本间取得平衡的用户。

パッケージ

dsh-thinking-levels

バージョン

0.1.0

最終更新

2026/08/18

Configuration

Two surfaces share one schema:

Assembly — the plugin row's config: in the profile composition (e.g. cordis.yml):

config:
  level: high            # off | low | high | max | auto
  allowDowngrade: false  # forbid the scheduler dropping below `high`
  allowUpgrade: false    # forbid the scheduler lifting to `max`

Runtime — the dsh-settings namespace thinking-levels (level, allowDowngrade, allowUpgrade, enabled): changes apply to the next model request, no restart needed.

Defaults: { enabled: true, level: 'auto', allowDowngrade: true, allowUpgrade: false }.

dsh-thinking-levels

AI 分析

インストール

ドキュメント

Configuration