Keep the built-in web_search tool without an official DeepSeek API key: disable web-search-deepseek and use Tavily keyless by default.
Install
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:shinjiyu/dsh-plugin-search
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time. Only install sources you trust, and pin a commit (github:owner/repo#sha).
README
不用官方 DeepSeek API，web_search 不再 401。
默认零配置。装上就把内置 web-search-deepseek 关掉，工具名还是 web_search。
对应：Discussions #779 · #408 · 介绍帖：Show and tell #1715
安装
dsh plugin --profile web add github:shinjiyu/dsh-plugin-search
然后重启 dsh web。旧会话的工具表不会更新。
配置
可以不配。没 key 走 Tavily keyless。
有账号再设：
TAVILY_API_KEY=tvly-...
不要把 key 写进仓库。
30 秒验收
- 主模型用 GLM / 网关 / 任何非官方 DeepSeek
- 不要设
DEEPSEEK_API_KEY - 问「今天 DeepSeek Harness 最新 Discussions 说了什么」
- 过关：
web_search能出链接，不再红字 401
不要做
- 不要改
runtime/或官方web-search-deepseek - 不要把它宣传成搜索引擎全家桶
社区
微信群：deepseek harness 交流群
微信群码大约 7 天过期（本张到 2026-08-22）。
宣发稿在 docs/PROMO.md。
License
MIT
Links
More in this category
liustack/modlens★ 1463
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
Anionex/dsh-vision-toolkit★ 372
Vision tasks for text-only models: intent-aware image Q&A, long-screenshot OCR, UI reproduction, grounding, and pixel diff.
zhaoolee/notes★ 141
Export DSH conversations as Smartisan Notes-style PNGs, or create and update Markdown notes in a configured account-scoped workspace.
Lum1104/dsh-browser★ 107
Chrome sidebar extension that lets DSH operate your browser directly, no vision capabilities required.
liustack/modsearch★ 98
Web search bridge for text-only agents: ask the web or X, get structured JSON evidence (search, fetch, citations).
ysr666/dsh-vision-router★ 83
Free vision for text-only agents: built-in keyless vision chain plus pixel tools (Q&A, grounding, crop, pixel diff, colors, OCR, SVG trace, cutout, screenshots); paste an image to use it.