Tinysearch

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/tinysuitehq-tinysearch

About

Use cases: web-search, web-scraping, browser-automation

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools

3 read · 1 write · 0 delete · 1 not yet classified

Secrets

1 (API keys, tokens or passwords)

Install scripts

Not scanned yet

Runs as

Python package

Tools (5)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• browser_actBrowser Act

  One action on the page browser_navigate opened. look() reads it without touching it. click(target) / type(target, text, submit) act on a ref you saw in a returned view; never invent a ref or selector, never type credentials. wait_for(time | text | text_gone). tabs(tab_action, index). close() when done. All but close return a page view, so find narrows what comes back to the matching nodes and their context, not the whole tree. Pass it unless you truly need the tree: a full page can be 50x larger. Read-only steps revealing already-public content (pagination, expanding a section, a cookie banner…

• browser_navigateBrowser Navigate

  Open an exact URL in a live browser and return a view of the page. Call scrape_urls on this exact URL first -- most pages, including ordinary articles, are static, and scrape_urls reads them for a fraction of the cost of a browser. Reach for this tool only when that call actually came back thin, empty, or clearly incomplete, or the page needs a read-only interaction first (a cookie banner, a "load more" control, content that renders after load, a client-side search box) to reveal what was asked for. Not a discovery tool. find narrows what comes back to the matching nodes and their context, not…

• get_current_datetimeGet Current Datetime

  Return the current date and time in UTC. Call this first for time-sensitive questions, relative dates such as latest, this year, or last month, or before adding year/month/day context to a search query.

• scrape_urlsScrape URLs

  Inspect one to five URL/query pairs. Each item has url and optional query; omit query or use '*' for page-order content, or provide a focused query describing needed evidence. Related links are navigation candidates; batch URLs only after selecting them. Returns each item's success or error independently.

• searchSearch

  Fast discovery for one to five independent items. Use one item for a simple lookup, multiple only for independent subquestions, and domains for hard source restrictions. Does not crawl or rerank.

Signals

Stars

233

Forks

18

Open issues

9

Contributors

4

Weekly downloads

105

Last commit

Oct 8, 2026 (20 days ago)

Last release

Oct 8, 2026 (20 days ago)

Last package publish

Oct 8, 2026 (20 days ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name

io.github.TinySuiteHQ/tinysearch

Version

0.6.6

Auth

API key

Packages

oci docker.io/marcellm01/tinysearch:v0.6.6pypi tinysuite-search

License

MIT

Repository

github.com/TinySuiteHQ/TinySearch

Website

github.com/TinySuiteHQ/TinySearch

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

Official MCP Registrygithub-graphql