kolawong/dsh-plugin-j-space

Name: kolawong/dsh-plugin-j-space
Author: kolawong

Plugin插件 Native原生 ⭐ 5 Apache-2.0 Productivity & Tasks效率与任务 Migration & Compatibility迁移与兼容

J-Space 认知套件 V3.6 - DeepSeek Harness（DSH）的推理时认知控制与深度推理插件

Project Overview项目介绍

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

This is a native DeepSeek Harness (DSH) inference-time cognitive control plugin that packages the open-source J-Space Cognition Suite Version 3.6, built exclusively to enhance the reasoning fidelity, context persistence, and verification discipline of LLMs like DeepSeek-V4-Pro/Flash and Kimi during complex, long-horizon tasks. You can install it via the DSH CLI with a single command pointing to this GitHub repository, or clone it manually to your local DSH plugins directory. It works across all major desktop operating systems including macOS, Linux, and Windows, and includes built-in automatic language detection for both English and Chinese.

J-Space addresses four common pain points that LLMs experience during long multi-step reasoning: working set overload, representation drift, uncontrolled retries, and premature completion. It uses six core cognitive mechanisms to create a structured, actively managed working memory for the model without changing model weights or requiring fine-tuning. Users can switch between four runtime modes via the DSH Web UI settings or a local configuration file, and the default on-demand mode automatically activates for complex tasks. It is designed for developers and researchers working on complex multi-step tasks like architecture refactoring, bug tracing, and theorem proving.

This plugin is released under the open-source Apache License 2.0, and it packages upstream work from the original J-Space Cognition Suite V3.6, with full attribution retained for the original author. No fine-tuning or weight modification of the underlying LLM is required to use this plugin, as all cognitive control is applied during inference time. After installation, you just need to restart your DSH process to activate the plugin, and you can trigger it either implicitly or via an explicit user prompt. There are no extra dependencies or complex configuration steps required to get started with the plugin.

这是一款面向DeepSeek Harness（DSH）的原生推理时认知控制插件，封装了开源的J-Space认知套件V3.6版本，专门用于提升DeepSeek-V4-Pro/Flash、Kimi等大模型在处理复杂长周期任务时的推理保真度、上下文持续性和验证规范性。它支持通过DSH CLI一键安装或手动克隆安装，兼容macOS、Linux、Windows全平台，还自带中英文多语言适配功能。

J-Space针对大模型长多步推理中常见的工作集过载、表示漂移、失控重试、提前完成四大问题，通过六大核心认知机制结构化优化大模型的推理工作空间。用户可在DSH的Web设置界面或配置文件中切换四种运行模式，默认按需激活，也可设置为始终开启、自动触发或完全关闭。它适合处理架构重构、漏洞追踪、定理证明等复杂多步骤任务的开发者和研究人员使用。

本插件采用Apache 2.0许可证开源，基于J-Space认知套件上游开源成果封装，完整保留了原作者的署名和授权信息。它不需要对大模型进行微调或修改权重，仅在推理阶段注入认知控制逻辑，首次使用只需安装完成后重启DSH即可生效。用户可通过隐式触发或明确提示激活插件，无需额外复杂配置。

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：dsh-plugin-j-space（kolawong/dsh-plugin-j-space）
仓库：https://github.com/kolawong/dsh-plugin-j-space
本站详情页：https://www.yhbd.top/plugins/kolawong-dsh-plugin-j-space/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证 Apache-2.0 · ⭐ 5 · 最近提交 2026-08-20 · 主语言 Python

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 5 stars - very few users, little community feedback星标只有 5，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 github:kolawong/dsh-plugin-j-space

把 kolawong/dsh-plugin-j-space 加入你的 DSH 配置（web profile）即可启用。

READMEREADME

6.4 KB · 缓存于 2026-08-23 03:41:47 · 6.4 KB

@custom/dsh-plugin-j-space

English | 中文说明

J-Space Cognition Suite (V3.6) is an inference-time cognitive control plugin designed for DeepSeek Harness (DSH). It enhances the reasoning fidelity, context persistence, and verification discipline of LLMs (especially DeepSeek-V4-Pro / Flash and Kimi models) during complex, long-horizon tasks.

🌟 Why J-Space?

During long multi-step reasoning, coding, and autonomous workflows, large language models frequently suffer from four major inference-time losses:

Working-Set Overload: Too many active constraints dilute attention.
Representation Drift: Global invariants, architectural definitions, or goals gradually mutate across steps.
Uncontrolled Retry: Repeating failed routes without carrying diagnostic hypotheses.
Premature Completion: Mistaking fluent conversational output for verified execution.

J-Space turns the model's accessible working memory into a structured, actively managed internal workspace without changing model weights or requiring fine-tuning.

⚙️ 6 Core Cognitive Mechanisms

Module	Mechanism	Impact
Broadcast Hub	Shared constraints derived once and broadcast	Prevents cross-file & cross-step representation drift
Dense Track	`✓ / ? / ✗` symbol registers with lossless plain-text expansion	Enforces stepwise falsification & rigorous self-verification
Directed Focus	Workspace limited to 1-2 active concepts	Eliminates working-set cognitive overload
Bridge Reasoning	Mandates intermediate bridging before conclusion	Eliminates conclusion-first rationalization
Self-Monitoring	Autonomously detects reasoning degeneration	Triggers rollback with explicit diagnosis
Workspace Ledger	Persistent state externalization (`jspace.py`)	Maintains durable memory across task seams & subagents

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 kolawong/dsh-plugin-j-space 的 README.md。在 GitHub 查看原文 →