dsh-browser
omdsh-dev
Chrome 侧边栏插件,让 DeepSeek Harness 直接操控你的浏览器,无需视觉能力也能自动完成网页操作。
falling-ts102更新于 18 天前
本地优先的激进上下文压缩插件,用低配模型精简历史记录,让实时提示词保持小巧快速且私密。
实用分
82 / 100
配置门槛
开箱即用
协议
MIT
语言
JavaScript
复制命令,或者把整个安装任务交给你的 Harness 智能体。
git clone https://github.com/falling-ts/dsh-force-compact.git请帮我安装 DeepSeek Harness 插件「dsh-force-compact」: - GitHub 仓库:falling-ts/dsh-force-compact(https://github.com/falling-ts/dsh-force-compact) - 类型:cordis 插件(装到 DSH profile) - 注意:该插件声明 DSH 版本要求 >=0.2.1-rc.1,安装前请先确认我的版本是否满足。 - 该插件开箱即用,无需额外配置。 请先查看仓库 README 确认安装步骤,然后按…
来自项目自身文档的清洗摘录。
dsh-force-compact Aggressive, local-first context compaction for DeepSeek Harness agents. A DSH Cordis function plugin that keeps the agent's working context lean by design: serve Qwen3.8‑27B on a self-hosted llama.cpp with a modest context, and the plugin shrinks the conversation itself — a large-window feel with no API cost and no data egress. 中文 --- Why - Self-hosted inference — the agent talks to a local OpenAI-compatible llama.cpp server through the standard DeepSeek adapter; no separate adapter needed. - Low context, high signal — instead of fighting a small cap, the plugin shrinks the conversation, so the agent reasons over a tight prompt while keeping deep memory in the compressed head. - Think off for compactions, passthrough everywhere else — disableThinking: true (default) turns thinking off on this plugin's own compaction summarization call only ; every other model request rides the machine's configuration unchanged. - Private & free — no per-token billing, no egress. --- What it does Two compaction engines coexist behind one facade ( resolveCompaction ), transparent to callers:…
在 GitHub 打开来自同一生态方向的相似工具。
omdsh-dev
Chrome 侧边栏插件,让 DeepSeek Harness 直接操控你的浏览器,无需视觉能力也能自动完成网页操作。
mistnest
在 DSH 里开个小说工作台,和 AI 一起写正文、理设定、推情节,续写改稿都顺手。
Finderchangchang
写份产品简报,AI 自动挑镜头写文案,一条命令生成竖版宣传片,便宜模型也能出好片。