dsh-plugin-rag

by yytbit

1 工具与能力github收录于 08-23

DeepSeek Harness 的本地知识库 RAG

Local knowledge base RAG for DeepSeek Harness

安装

dsh plugin --profile web add github:yytbit/dsh-plugin-rag

GitHub 源码安装:首次需按提示配置 allowBuilds 构建授权后重试

安装与环境配置指引、插件开发教程见 DSH 中文社区文档 ↗

安装即在你的机器上以你的权限运行第三方代码——它可读写文件、使用凭据、访问网络,DSH 的工具审批不会为插件代码加沙箱。「检测到 manifest」仅代表发现 dsh.bundle / dsh.plugin 清单,不构成兼容性或安全审查;安装前请审阅源码,不熟悉的插件先在不含密钥的环境试用。

README

目录

Local knowledge base RAG for DeepSeek Harness.

What it does

Indexes project files (markdown, code, configs) and provides keyword search to find relevant context. No external APIs or embedding models needed -- pure local TF-IDF search.

Install

dsh plugin --profile your-profile add dsh-plugin-rag

CLI

# Search for relevant files
rag "authentication" ./src

# Show index stats
rag --index ./src ./docs

What gets indexed

  • Markdown (.md, .txt)
  • Code (.ts, .js, .py, .rs, .go, .java, .c, .cpp)
  • Config (.json, .yaml, .toml, .env)
  • Docs (.html, .xml, .sql, .graphql)

Skips: node_modules, .git, dist, build, pycache

Configuration

- id: rag
  name: dsh-plugin-rag
  config:
    indexPaths: ['.']
    maxResults: 3
    maxBytes: 4096
    enableInjection: true

How it works

  1. Scans directories for text files
  2. Builds a keyword index (TF-IDF scoring)
  3. Injects index stats into system prompt
  4. Agent uses CLI to search when needed

No embedding models, no external APIs, no heavy dependencies.

License

MIT -- YYTbit

原始 README: https://github.com/YYTbit/dsh-plugin-rag/blob/master/README.md ↗