DSH Plugin HubDSH 插件市场

staff-os/dsh-ragflow

Plugin插件 Native原生 ⭐ 2 Memory & Knowledge记忆与知识库

DeepSeek Harness 的 RAGFlow 知识库检索插件。

Project Overview项目介绍

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

This is a plugin built exclusively for DeepSeek Harness (DSH) that adds RAGFlow knowledge base retrieval capability to DSH agents. It ships with a native DSH bundle manifest and can be installed directly using the DSH CLI with the command dsh plugin --profile web add "github:staff-os/dsh-ragflow#main". Once installed, it exposes a ragflow_retrieve tool that queries configured RAGFlow datasets and returns relevant document chunks with similarity scores and source attribution. Since the compiled lib/ directory is committed to the repository, no build step is required at install time. After installation, users just need to restart the DSH web process to activate the plugin.

The plugin follows DSH's three-role capability pattern, splitting into four plugin entry points: service definition, service provider, model-facing tool, and user-facing configuration page. Each entry point can be overridden, replaced, or removed individually without modifying other parts, making it easy to customize for specific use cases. Before the tool can be used, users need to have a running RAGFlow instance (either self-hosted or cloud), a valid RAGFlow API key, and at least one dataset with parsed documents configured. Users can choose from three configuration methods: web configuration page, environment variables, or YAML config.

Configuration can be done via three methods, and the API key is stored securely through DSH's credential service so it never appears as plaintext in configuration files. Currently, the plugin only supports retrieval functionality, and does not include dataset or document management features like creating, uploading, or parsing documents. Several advanced RAGFlow retrieval options are also not yet exposed, including cross-language search, metadata filtering, and knowledge graph integration. The plugin does not support streaming responses, and results are rendered as a generic search card rather than a custom citation card.

这是一个专为 DeepSeek Harness 开发的原生插件，用于对接 RAGFlow 知识库的检索能力。它会为 DSH 代理提供一个 ragflow_retrieve 工具，能够查询用户配置的 RAGFlow 数据集，返回匹配的文档片段，同时附带相似度评分和来源信息。插件遵循 DSH 原生插件规范打包，自带 bundle 清单，可以直接通过 DSH 命令行工具一键安装。

插件遵循 DSH 三角色能力模式设计，拆分为服务定义、服务提供者、工具消费者和配置页面四个入口，支持单独覆盖、替换或移除任意模块，方便自定义改造。用户完成安装后，需要先配置 RAGFlow 的实例地址、API 密钥以及至少一个已完成文档解析的数据集，才能正常使用检索功能。

配置可以通过 Web 配置页面、环境变量或 YAML 三种方式完成，API 密钥会通过 DSH 凭据服务存储，不会明文保存在配置文件中。目前插件仅支持检索功能，不提供数据集和文档管理能力，部分 RAGFlow 高级检索选项暂未开放，源码已提交编译产物，安装时无需重新构建。

View on GitHub在 GitHub 查看

Recommended强烈推荐 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-ragflow（staff-os/dsh-ragflow）
仓库：https://github.com/staff-os/dsh-ragflow
本站详情页：https://www.yhbd.top/plugins/staff-os-dsh-ragflow/
本站登记：类型 plugin · 归类 原生 DSH 插件 · 许可证未声明 · ⭐ 2 · 最近提交 2026-08-18 · 主语言 TypeScript

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 2 warnings2 项注意
  • No license declared - all rights reserved by default; ask the author before commercial use or redistribution未声明开源许可证 —— 默认「保留所有权利」，商用或再分发前先问作者
  • 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 --profile web add "github:staff-os/dsh-ragflow#main"

把 staff-os/dsh-ragflow 加入你的 DSH 配置（web profile）即可启用。

READMEREADME

10.2 KB · 缓存于 2026-08-27 05:42:29 · 10.2 KB

dsh-ragflow

English | 中文

RAGFlow knowledge-base retrieval for the DeepSeek Harness. It gives the agent a ragflow_retrieve tool that queries your RAGFlow datasets and returns document chunks with similarity scores and source names.

Design

The package follows the Harness three-role capability pattern: one seam, one provider, one consumer — shipped as one package with four plugin entry points (the fourth is the configuration page over the same values), so a profile can override, replace, or drop any single role without touching the others.

Role Module Plugin name Responsibility
Service Definition src/index.ts @deepseek-ai/dsh-ragflow Owns ctx.ragflow: provider registry, order-independent selection, maxChunks enforcement
Service Provider src/http.ts @deepseek-ai/dsh-ragflow/http Calls POST /api/v1/retrieval, resolves credentials, normalizes chunks
Consumer src/tool.ts @deepseek-ai/dsh-ragflow/tool The model-facing tool: schema, prompt guidance, chunk bound, presentation
Consumer src/config.ts @deepseek-ai/dsh-ragflow/config The person-facing configuration page over the same settings and credential seams

The provider and the consumer depend only on the Service Definition, never on each other. Replacing the backend means replacing one row:

- id: ragflow-http
  name: 'your-own-ragflow-provider'

Prerequisites

  1. A running RAGFlow instance — self-hosted or cloud. Defaults to http://localhost:9380.
  2. An API key — create one in RAGFlow.
  3. A dataset with parsed documents — RAGFlow rejects a retrieval that names no dataset and no document, so at least one dataset id must be configured.

Install

dsh plugin --profile web add "github:staff-os/dsh-ragflow#main"

dsh plugin add forwards the source to pnpm as-is, so any pnpm-recognized source works:

dsh plugin --profile web add link:/path/to/dsh-ragflow   # local development

lib/ is committed, so no build script runs at install time and pnpm needs no build allowance. Restart dsh web afterwards, then verify the four rows landed:

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 staff-os/dsh-ragflow 的 README.md。 在 GitHub 查看原文 →

Related相关推荐

← 上一个 Prev dsh-plugin-thinking-api 下一个 Next dsh-sketchup →