DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Luxueliu Reasoning Efforts — DSH Plugin for DeepSeek Harness
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in
← Plugins
L

luxueliu-reasoning-efforts

Luxueliu Reasoning Efforts

Reasoning-effort ladders for every non-ds model on a local LLM gateway route in DeepSeek Harness: declares selectable thinking levels (off/low/medium/high/xhigh/max) for hand-declared gateway models that would otherwise show no reasoning control at all, and re-applies them to the live profile settin

The plugin will be installed here. Keep web if you are unsure.

npx -y @deepseek-ai/dsh plugin --profile web add github:luxueliu/luxueliu-reasoning-efforts#5525d3aae60999164203ff15cddf12d2cc117cdc
READMECompatibilityVersions

Description

Reasoning-effort ladders for every non-ds model on a local LLM gateway route in DeepSeek Harness: declares selectable thinking levels (off/low/medium/high/xhigh/max) for hand-declared gateway models that would otherwise show no reasoning control at all, and re-applies them to the live profile settings after DSH upgrades overwrite node_modules. Ships the route-level reasoningEfforts source patch (llm-pi-ai), the 20-alias gold ladder table (efforts.json), and a check/re-apply command (/efforts).

Compatibility and provenance

Luxueliu Reasoning Efforts is published as luxueliu-reasoning-efforts and currently resolves to version 0.1.0. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
any
Release source
github
Registry updated
8/21/2026

Versions

0.1.0stable
8/21/2026

Related plugins

Loading related plugins…

Latest
0.1.0
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
Unavailable
Files
Unavailable
Surface
any
License
MIT
Source
github
GitHub
★ 1
Weekly downloads
0
Last push
8/21/2026
View source ↗
README badge

Click the badge to copy Markdown for your README.

Do you maintain this Plugin?Claim benefit · Priority security scan

Verify the GitHub repository declared in package.json to manage this listing. After you claim it, Hub will prioritize a security scan of the current version and publish the result when it passes.

Claim this Plugin →
Report an issue

Related plugins

More verified plugins in models-usage.

Usage@linxin666/dsh-usageUsage statistics plugin for the dsh web GUI: per-provider balance and coding-plan quota detection plus a live token usage ledger, with the current session provider's today usage on the sidebar entryWhale Widgetdsh-whale-widgetDeepSeek balance whale widget in the bottom-right corner of the DSH Web interface: balance/today’s usage/peak-off-peak pricing, customizable bubble click sequence (text/balance/today/peak-off-peak/image/random phrases and parallel weighted selection), per-line styles and fonts, floating quick editinUsage Stats@ychris12138/dsh-usage-statsToken usage heatmap, provider balances, and subscription quotas for the dsh web GUICodex Connectdsh-codex-connectChatGPT OAuth and Codex models for DeepSeek Harness.

README

luxueliu-reasoning-efforts

💡 Gemini / Grok / Kimi 在 DSH 里「不会思考」?一条命令解锁 20 个模型的推理强度

DeepSeek Harness 插件:让本地 LLM 网关路由上除 ds 系以外的任何模型,也能在 DSH 里显示推理强度(reasoning effort)选择器,并在 DSH 升级后一键重打档位。

由 luxueliu 制作并开源(MIT)。

这是什么

DSH 的模型选择器只在模型被声明为「可推理」时才会显示推理强度档位。走本地 LiteLLM 网关的手工声明模型(fable5 / opus5 / gemini3.x / grok / kimi-* / glm5.x 等)默认没有这个能力——于是它们既不显示档位选择器,也无法从对话里切换思考强度。这个问题官方长期未修,本插件就是它的补丁 + 档位真理源。

本插件提供三层能力:

  1. 档位真理源 efforts.json:20 个网关别名的 reasoningEfforts 档位表(off/low/medium/high/xhigh/max,按各模型真实支持的档位声明)。
  2. 检查 / 重打命令 /efforts:对比 live ~/.dsh/settings.yaml 与金样档位表,检查漂移、一键重打(自动备份),DSH 升级或界面回写 settings 后档位漂了随时拉回来。
  3. 路由级源码补丁 upstream.patch:给 llm-pi-ai 增加「路由级 reasoningEfforts 默认值」——整条网关声明一次档位表,所有未单独声明的手工模型继承(PR 分支已推送:luxueliu/deepseek-harness 的 feat/pi-ai-route-reasoning-efforts;上游 org 的 API 限制需在网页手动建 PR,合并前可先手动应用本补丁)。

安装

要求:Node.js 22.18+ 与一个 DeepSeek Harness Web profile。

dsh plugin --profile web add github:luxueliu/luxueliu-reasoning-efforts

重启 DSH 后:

/efforts        # 检查档位是否漂移
/efforts apply  # 把金样档位写回 live settings(先留 settings.yaml.bak-reapply-*)

源码补丁(可选,二选一)

路由级默认值是「网关档位表写一次」的便利特性;若只想让现有 20 个别名显示档位,装完插件 /efforts apply 即可(模型级声明)。

要用路由级能力(或等官方合入前先体验):

# 方式一:应用随包补丁(D:\DSH 已内置 reapply 脚本时走官方脚本)
powershell -NoProfile -ExecutionPolicy Bypass -File D:\DSH\reapply-after-dsh-upgrade.ps1

# 方式二:直接对 DSH 源码应用 upstream.patch(在仓库根执行)
git apply upstream.patch

命令

/efforts          # 检查 live settings.yaml 与 efforts.json 是否一致
/efforts apply    # 有漂移则重打(自动备份 settings.yaml.bak-reapply-*)

配置

插件的 cordis.patch.yml 已带默认配置,可在 profile 的 cordis.patch.yml 覆盖:

- insert:
    - id: luxueliu-reasoning-efforts
      name: luxueliu-reasoning-efforts
      config:
        settingsPath: ""        # live settings.yaml;留空自动探测 ~/.dsh/settings.yaml
        checkOnStart: true      # 启动时检查档位漂移并提示
        applyOnStart: false     # 启动时自动重打(默认关,先 check)
        effortsFile: "efforts.json"

文件说明

文件用途
efforts.json20 个别名的 reasoningEfforts 档位真理源
settings.yaml.gold完整 settings 快照(含默认模型,整份覆盖用,慎用)
litellm-config.yaml.gold网关配置快照(含 allowed_openai_params: reasoning_effort)
reapply.py独立重打脚本(只写档位,不动默认模型;--check 只查)
restore-settings-gold.ps1整份 settings 回滚到金样(会覆盖默认模型,慎用)
upstream.patchllm-pi-ai 路由级 reasoningEfforts 源码补丁
lib/index.js/efforts 命令实现(检查 + 重打,文本级替换保 off: 拼写)

License

MIT © luxueliu