dylan121322/llm-adaptive

Name: dylan121322/llm-adaptive
Author: dylan121322

Plugin插件 Native原生 ⭐ 2 MIT Models & Routing模型与路由 Sessions & Context会话与上下文

Project Overview项目介绍

Edited by Zhiwei Agent (zhiwei)由知微Agent(zhiwei)编辑

This is a native plugin built exclusively for DeepSeek Harness that adds adaptive model routing functionality to the platform. It adds a new adaptive provider to DSH’s built-in model picker, which automatically classifies every incoming LLM request based on its task complexity. Classification uses a flash classifier model and splits requests into four distinct tiers: low, medium, high, and critical. It then routes each request to the first available appropriate backend model provider based on user-defined configuration chains. It also caches classification decisions for 120 seconds to reduce redundant classifier calls.

Typical usage starts right after installation by opening DSH’s /model page and selecting the adaptive（自动路由） option from the model picker. Every subsequent LLM request will automatically be sent to the flash classifier for complexity ranking, then forwarded down the configured routing chain for the matched complexity level. If a configured provider in the chain fails to handle the request, the plugin automatically falls back to the next available provider in the chain. This fail-open design ensures that classification errors or provider outages never block user requests, and all routing decisions are logged for debugging.

To install and run this plugin, you first need a working DeepSeek Harness installation on your local machine. You can install the plugin directly via the DSH CLI using the simple command dsh plugin add llm-adaptive, or you can install it from a local source checkout by editing your package.json and running a manual npm install. After installation completes, you must restart the DSH web service to activate the new plugin. You also need a valid DeepSeek API key for the classifier to work, plus a properly formatted pool.json model configuration file. This project is released under the open source MIT license, with no associated costs for use.

这是DeepSeek Harness的原生模型路由插件，专为DSH设计，为DSH的模型选择器新增了adaptive（自动路由）选项，能够自动对每一个LLM请求按照复杂度进行分级，从低到高分为四级，再将请求根据用户配置好的路由规则转发到对应等级的后端模型提供商。

它调用DeepSeek快速模型完成请求复杂度分级，会结合当前会话的滚动目标摘要和最新对话内容做上下文感知判断，还支持分级锁定，任务中途不会随意降级，所有模型路由配置都从外部配置文件读取，不会在插件内硬编码凭证信息。

安装需要先有DeepSeek Harness环境，可通过dsh命令行直接安装，安装完成后重启DSH服务，在模型选择页选中adaptive选项即可启用。你需要提前准备好模型池配置文件，同时拥有DeepSeek API密钥供分类器调用。

View on GitHub在 GitHub 查看

Send to your DSH发给你的 DSH 安装

请帮我安装这个 DSH 插件。安装前先完成【兼容性检查 + 安全性检查】，检查通过再动手。

插件：llm-adaptive（dylan121322/llm-adaptive）
仓库：https://github.com/dylan121322/llm-adaptive
本站详情页：https://www.yhbd.top/plugins/dylan121322-llm-adaptive/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证 MIT · ⭐ 2 · 最近提交 2026-08-14 · 主语言 JavaScript

按下面顺序执行，每步先把结论告诉我，再进入下一步：

【1 兼容性检查】
① 我这边：DSH 版本、Node 版本、操作系统、当前 profile（web / desktop）。
② 读它的 README、package.json、插件 manifest，列出它要求的 DSH 版本 / Node 版本 / 操作系统 / 外部依赖 / 需要另外先装的运行时。
③ 逐条比对，结论只写「满足 / 不满足 / 未知」三种；不满足的给出可行替代方案。
④ 检查是否和我已装的插件冲突：命令名重复、skill / tool 重名、端口占用、重复注册的 MCP server。

【2 安全性检查】
① 仓库可信度：和上面「本站登记」是否一致；star / fork 数、创建时间、最近提交，是否归档或长期停更。
② 安装脚本：逐行看 package.json 的 preinstall / install / postinstall，以及 install.sh、setup.ps1 之类脚本。出现 curl|bash、下载后直接执行、混淆代码、访问与插件功能无关的域名，立刻停下来告诉我，不要继续装。
③ 依赖：列出新增依赖，标出无人维护、或与知名包拼写近似的可疑包（typosquatting）。
④ 权限与副作用：它会读写哪些目录、访问哪些域名、需要哪些 DSH 权限（filesystem / network / shell / clipboard 等），以及怎么卸载和回滚。
⑤ 如果它要求 sudo / 管理员权限，或权限明显超出功能所需，先停下来问我。

【3 安装】
上面两步没有「不满足」和「高危项」时才执行；用官方推荐方式安装，不要自行提权。

【4 汇报】
用表格输出：检查项 / 结论 / 依据 / 是否需要我决策。拿不准的一律写「未知」并说明要我怎么确认——不要猜，也不要替我决定。

Send this message to DSH in your current session: it verifies compatibility and security first (answering met / not met / unknown item by item) and only installs once everything checks out — it will stop and ask you if it finds a high-risk item. The box scrolls; the copy is the full prompt. CLI install commands may not be accurate across systems, so DSH is the safer route.把上面这条消息直接发给当前会话里的 DSH：它会先核对兼容性与安全性（逐条给「满足 / 不满足 / 未知」），确认没问题再安装，有高危项会停下来问你。框内可滚动，复制到的是完整提示词；安装命令不一定准确，发给 DSH 更稳。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 warning1 项注意

Only 2 stars - very few users, little community feedback星标只有 2，几乎没人在用，遇到问题缺少社区反馈

DSH walks through these 9 checksDSH 会逐条核对这 9 项

Compatibility兼容性

DSH, Node, OS and profile requirementsDSH 版本 / Node 版本 / 操作系统 / profile 是否满足要求
External dependencies and runtimes (Electron / Python / Docker, ...)外部依赖与运行时（Electron / Python / Docker 等）是否齐备
Conflicts with installed plugins: command names, skill / tool names, ports, duplicate MCP registration与已装插件是否冲突：命令名、skill / tool 重名、端口占用、重复 MCP 注册

Security安全性

Repo matches the facts registered here; archived or abandoned?仓库是否与页面登记一致，是否归档或长期停更
Safety of preinstall / install / postinstall and install.sh / setup.ps1preinstall / install / postinstall 与 install.sh、setup.ps1 是否安全
curl|bash, download-then-execute, obfuscation, unrelated domains → stop immediatelycurl|bash、下载即执行、混淆代码、无关域名 → 立刻停止
Typosquatting or unmaintained packages among the new dependencies新增依赖里有没有 typosquatting 或无人维护的包
Requested permissions vs. what the feature actually needs申请了哪些权限、是否超出功能所需（filesystem / network / shell / clipboard）
Any sudo / admin requirement, plus uninstall and rollback是否要求 sudo / 管理员权限，以及卸载与回滚方式

Anything uncertain must be marked unknown with a note on how to confirm it. This site's signal screen is a static snapshot, not a security audit.拿不准的必须标「未知」并说明要我怎么确认。本站的信号筛查是静态快照，不能替代安全审计。

Or use CLI install (for developers)或使用命令行安装（适合开发者）

CLI Install命令行安装

dsh plugin add llm-adaptive

把 dylan121322/llm-adaptive 加入你的 DSH 配置（web profile）即可启用。

READMEREADME

3.1 KB · 缓存于 2026-08-25 04:53:41 · 3.1 KB

llm-adaptive

Adaptive model routing plugin for DeepSeek Harness. Adds an adaptive provider to the model picker: every LLM request is classified by a flash classifier (low / medium / high / critical) and routed to the matching backend provider through config-driven chains.

Features

Per-request complexity classification — deepseek-v4-flash called directly (never through a proxy, no recursion).
Context-aware judging — injects a rolling session-goal summary plus the recent turns into the classifier prompt (continuation / wrap-up / error-loop rules).
Sticky level protection — a mid-task downgrade is held at the previous level unless the message carries explicit downgrade or wrap-up signals.
Config-driven routing chains — chains come from pool.json → routing.levels ($active expands to the active provider, missing entries fall back to defaults); transport failures walk down the chain.
Classifier config from the pool — URL / model / key reference read from the classifier section of pool.json (no hardcoded credentials).
Fail-open — any classification failure degrades to medium; never blocks a request.
Observable — every decision (level, cause: llm/sticky/cache) is written to the plugin log.
120s decision cache — keyed by user-text head plus goal fingerprint.

Requirements

DeepSeek Harness (dsh)
A model pool file at ~/.dsh/tools/cc-switch-sync/pool.json with:
- classifier section: url, model, key_ref (resolved against ~/.dsh/.credentials.yaml, pool api_key as fallback)
- routing.levels: low / medium / high / critical chains
A DeepSeek API key for the classifier

The pool file is produced by the cc-switch-sync import tool (or can be authored by hand). The plugin reads it on every request, so pool edits take effect immediately.

Install

dsh plugin add llm-adaptive

or, from a local checkout:

cd ~/.dsh/profiles/web && npx pnpm@10 install   # with "llm-adaptive": "file:plugins/llm-adaptive"

Showing the opening section of the README — the full document lives in the repository以上为 README 开头摘要，完整文档在仓库内 · View the full README on GitHub →在 GitHub 查看完整 README →

渲染基于仓库 dylan121322/llm-adaptive 的 README.md。在 GitHub 查看原文 →