Dsh Data Agent

InstallableBundlenpmfresh

Data Agent for DeepSeek Harness: session-scoped database connections with a dedicated agent preset that lets AI write SQL and iterate against live execution feedback.

Install Dsh Data Agent

Clean install: prebuilt on npm and declares a `dsh.bundle` layer.

The one-line version

dsh plugin --profile my-profile add @yejiming/dsh-data-agent

The easy path: this bundle is published to npm, so the install pulls a prebuilt artifact — no build step runs on your machine and no build approval is needed.

From npm — installs a prebuilt artifact, no build step on your machine

# 1. add the bundle to a profile — --profile my-profile points at $DSH_HOME/profiles/my-profile
dsh plugin --profile my-profile add @yejiming/dsh-data-agent

# 2. verify: the bundle should show up as a layer in the resolved config
dsh --profile my-profile --dump-config

# 3. boot the profile
dsh --profile my-profile

Package: @yejiming/dsh-data-agent

Prefer to install from source instead?

From GitHub — pulls source, and may ask for build approval on first run

# 1. add the bundle to a profile — --profile my-profile points at $DSH_HOME/profiles/my-profile
dsh plugin --profile my-profile add github:omdsh-dev/dsh-data-agent#<commit-sha>

# 2. verify: the bundle should show up as a layer in the resolved config
dsh --profile my-profile --dump-config

# 3. boot the profile
dsh --profile my-profile

Prerequisites

API key: Not required.
Network access: Not required at runtime.
Language: TypeScript

Install check

Every field the verdict was derived from, so you can re-derive it yourself

Field checked	Result	Weight	What the spec says about it
lifecycleScripts	pass	info	no install-time scripts
dsh.bundle	pass	fatal	package.json declares a `dsh.bundle` layer
dsh.bundle.patch file	pass	fatal	patch file present at `cordis.patch.yml`
npm registry	pass	fatal	published as `@yejiming/dsh-data-agent` — prebuilt artifacts

Verdict reason: on npm with a valid bundle declaration
package.json read: package.json
Rules applied from: docs/user/develop/basic/publish.md
Checked at: 2026-08-15

From the README

Excerpt as published by the author, plain text, unedited

用AI写过SQL的同学都有这种体验，AI现在写代码能力已经很强了，但SQL逻辑老写不对。**原因是AI并没有与数据库操作形成Agent Loop**，它只能根据静态指令生成SQL，却无法感知执行结果、无法根据报错或返回数据动态调优。这个插件就是来填这个坑的。它复用DeepSeek Harness强大的Agent主循环能力，让AI连上数据库并获得实时反馈，同时删掉所有跟数据无关的上下文和工具，让AI专注于SQL生成和业务数据分析。我利用DeepSeek Harness的Agent预设功能，定义了专用的Data Agent预设。仅保留read、edit、write三个DSH自带的 tools，并自定义sqlcmd tool替代bash tool。懂行的朋友一眼就能看出，这是借鉴了Pi Agent的设计，只使用最基本的工具。用起来也很简单：在对话界面配好数据库连接，授权AI访问权限，然后就可以向AI提问，让AI帮你查询、更新、分析。 - **数据库连接管理**：按会话连接 MySQL / PostgreSQL / SQLite / Oracle / Hive / Impala（SQLite 走文件路径，Oracle 填服务名/SID，Hive/Impala 填默认库），连接状态驻留服务端内存，布局切换不丢；密码仅内存、经环境变量或 stdin 连接前缀传给客户端，绝不落盘。 - **数据库工作台**（内嵌于会话输入框上方）：连接配置卡（连接成功后折叠为摘要行，可展开查看）；库表浏览（点击「库表」按钮弹出 Modal：单击库展开表列表（可滚动），点击表查看结构）；SQL 命令框（编辑并运行 SQL，非 agent 通道，结果等宽展示）。连接配置持久化到浏览器 localStorage，切换页面/重启自动回填并重连。开始对话后工作台自动变为左侧栏，对话记录与输入框在右侧。 -

Topics

GitHub topics on this repository

#data-agent #deepseek-harness #dsh #dsh-plugin