npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@anonyjcy/dsh-plugin-j-space

v1.0.1

Published

J-Space Cognition Suite V3.7 native Agent Preset and standalone plugin for DeepSeek Harness

Readme

🚀 dsh-plugin-j-space

简体中文 | English Version

J-Space Cognition Suite V3.7 原生 DeepSeek Harness (DSH) Agent Preset 预设与独立 Cordis 插件包。
注入内部表征认知路由、工作区状态外化账本(.jspace/)与自适应检验,全面释放大语言模型推理潜能。


🌟 项目简介

dsh-plugin-j-spaceJ-Space Cognition Suite V3.7 完整集成为 DeepSeek Harness 的一等公民 Agent 预设 (Preset)

它不是粗暴的 Prompt 拼接,而是真正参与 Cordis Agent Scope 生命周期隔离、多层认知路由、工作区状态外化账本(.jspace/)与自适应验证 的完整体系,完全解耦并兼容任意大语言模型(DeepSeek-Chat、DeepSeek-Reasoner、Claude、GPT 等)。

✨ 核心特性

  1. 即插即用(Zero-Config Preset):部署预设后,DeepSeek Harness 会话创建菜单自动出现 J-Space Cognition Suite 预设。
  2. 全生命周期 Scope 隔离:遵循 Cordis 作用域规范,工具和认知技能严格限定在 J-Space Agent 会话中,不污染其他预设。
  3. 工作区状态外化账本(Active Ledger):通过 jspace.py 在任务工作区(cwd)自动建立 .jspace/ 认知账本,实现目标跟踪、接缝审计(Seam)、自检断言与断点恢复。
  4. 模型解耦与动态路由:无缝适配各类基底模型,动态解析模型路由与工作区上下文。

📊 实验与评测数据报告(实测对比)

完整测试报告引自原作者实测:DeepSeek-V4-J-Space-Capability-Realization-Report

🔬 评测方法与实验设置

  • 评测基底DeepSeek-V4-Flash-Vision-Exp
  • 运行环境:DeepSeek Harness (标准模式)
  • 评测方式:对权威基准子集与同类型小集(Terminal-Bench 2.1 中 medium 20 / hard 10,DeepSWE 中 TypeScript 10 / Python 10 / Go 10 / JavaScript 2 / Rust 2,GAIA 中 level1 / level3 等)进行严格的 有/无 J-Space 臂对照(A/B Testing),同模型、同环境、同采样参数,仅切换 J-Space 接入。
  • 测算维度:① 准确率(Accuracy);② 墙钟与 token 效率(Wall-Clock & Token Efficiency)。

1. 主基准测试准确率对比(Main Benchmark Table)

| Benchmark 基准测试 | DeepSeek V4-Flash (基线) | DeepSeek V4-Flash + J-Space V3.7 | GLM-5.3 | Kimi-K3 | Opus-4.8 | Fable 5 (w/ fallback) | | :--- | :---: | :---: | :---: | :---: | :---: | :---: | | HLE (w/o tools) | *37.8 | 37.8 | — | 43.5 | 49.8 | 53.3 | | HLE (w/ tools) | *51.5 | 51.9 | 62.5 | 56.0 | 57.9 | 63.0 | | Terminal Bench 2.1 | 83.9 | 85.5 | 88.2 | 88.3 | 85.0 | 88.0 | | NL2Repo | 57.7 | 60.4 | 58.0 | 58.0 | 69.7 | — | | CyberGym | 75.3 | 77.8 | 84.5 | 80.0 | 78.3 | 83.1 | | DeepSWE | 59.3 | 61.8 | 66.9 | 67.5 | 58.0 | 70.0 | | Toolathlon-Verified | 75.9 | 77.4 | 73.0 | 76.5 | 76.2 | 77.9 | | Agents' Last Exam | 27.3 | 28.3 | 28.5 | 27.6 | 25.7 | 23.8 | | AutomationBench (Public) | 25.7 | 27.6 | 48.2 | 30.8 | 27.2 | 29.1 | | ⭐ 综合均分 (Average) | 56.99 | 58.61 | 64.54 | 60.96 | 58.33 | 62.13 |

* 注:HLE 数据未披露,沿用 DeepSeek V4-Flash-0731。综合均分覆盖六列均有值的 7 个项目行。


2. 速度与 Token 消耗效率对比(Speed & Token Efficiency)

| Benchmark 基准测试 | 墙钟时间比 τ | 提速幅度 | 输出 Token 变化 | 总 Token 变化 | 单位时间得分 (产出比) | 每成功任务成本 | | :--- | :---: | :---: | :---: | :---: | :---: | :---: | | HLE (w/o tools) | *1.02 | −2% | −10% | +5% | 0.98× | +5% | | HLE (w/ tools) | 0.88 | +14% | −22% | +3% | 1.15× | +2% | | Terminal Bench 2.1 | 0.79 | +27% | −28% | −3% | 1.29× | −5% | | DeepSWE | 0.78 | +28% | −28% | −3% | 1.34× | −7% | | Toolathlon-Verified | 0.86 | +16% | −25% | +2% | 1.19× | +0% | | AutomationBench (Public) | 0.76 | +32% | −31% | −5% | 1.41× | −12% |

* 注:HLE (w/o tools) 的 τ=1.02 是有意为正(即单轮无工具任务下略微变慢),因为在单轮短任务中注入完整的技能条目是净开销;而在长链路、多轮工具交互任务中(如 Terminal Bench、DeepSWE、AutomationBench),J-Space 认知套件带来 +14% ~ +32% 的大幅提速降低 28%~31% 的输出 Token 冗余,单位时间产出比提升高达 1.15× ~ 1.41×


🚀 安装与一键部署

本插件内置了开箱即用的原生 Node.js CLI 工具,无需额外安装其他依赖即可直接执行安装:

方式一:通过 npm / pnpm 安装(官方源)

# npm 安装
npm install -D @anonyjcy/dsh-plugin-j-space

# pnpm 安装
pnpm add -D @anonyjcy/dsh-plugin-j-space

# 运行 CLI 一键部署预设
npx @anonyjcy/dsh-plugin-j-space install

方式二:克隆仓库直接安装(本地使用)

git clone https://github.com/AnonyJcy/dsh-plugin-j-space.git
cd dsh-plugin-j-space

# 一键部署预设到 ~/.dsh/.agent-presets/j-space
node bin/cli.js install

# 检查安装状态与完整性
node bin/cli.js status

💡 使用方法

1. Web UI 界面

  1. 打开 DeepSeek Harness Web 界面,点击 New Session(新建会话)。
  2. Agent Preset 下拉选单中,直接选择 J-Space Cognition Suite
  3. 选择任意兼容的模型(deepseek-chat / deepseek-reasoner 等)开始任务。

2. CLI 命令行

dsh --preset j-space "全面重构此模块并补充单元测试"

3. Cordis 配置文件组装 (cordis.yml)

- id: j-space-plugin
  name: '@anonyjcy/dsh-plugin-j-space'
  config:
    autoDeploy: true

🧩 核心架构与数据流

flowchart TD
    A[新建 Session] --> B[选择 j-space 预设]
    B --> C[Preset Discovery: AgentPresets.list]
    C --> D[Preset Mount: AgentPresets.mount]
    D --> E[Agent Scope]
    E --> F1[Persona: J-Space V3.7 认知系统]
    E --> F2[Tools: 完整编码与思考工具]
    E --> F3[Skill Filesystem: 挂载 skills/j-space/]
    E --> F4[J-Space Suite: SKILL.md, 9大模块, 4大参考, 3大脚本]
    E --> G[Session Model Route: 任意兼容模型]
    G --> H[Agent 运行 J-Space 认知闭环]
    H --> I[Task Workspace: 生成并维护 .jspace/ 账本]

🛠️ CLI 命令一览

node bin/cli.js install    # 安装 J-Space Preset 到 DSH 用户预设目录 (~/.dsh/.agent-presets/j-space)
node bin/cli.js uninstall  # 干净卸载 J-Space Preset
node bin/cli.js verify     # 校验已安装预设的套件完整性
node bin/cli.js status     # 查看当前安装状态与配置路径

📄 开源许可证

本项目基于 MIT License 开源。套件第三方声明见 THIRD_PARTY_NOTICES.md