Local-first, no-API-key MCP server for web search, fetch, and document reading. Multi-engine (DuckDuckGo, Mojeek, Startpage), smart Playwright fallback, FTS5 cache, LLM-tuned output.
MCPpedia last refreshed this data
Free Search MCP is an MCP server that local-first, no-API-key MCP server for web search, fetch, and document reading. Multi-engine (DuckDuckGo, Mojeek, Startpage), smart Playwright fallback, FTS5 cache, LLM-tuned output. Its tool list has not been published yet over stdio, sse and http, requires no API key, and scores 59/100 on MCPpedia's security, maintenance and efficiency rubric.
Config is the same across clients — only the file and path differ.
{
"mcpServers": {
"search": {
"args": [
"search-mcp"
],
"command": "uvx"
}
}
}Are you the author?
Add this badge to your README to show your security score and help users find safe servers.
A local-first, no-API-key Model Context Protocol server that gives any LLM (Claude, GPT, local Ollama, …) the ability to search the web, fetch and clean up pages, and read documents — without you signing up for a single
This server supports HTTP transport. Be the first to test it — help the community know if it works.
Five weighted categories — click any category to see the underlying evidence.
No known CVEs.
Checked free-search-mcp against OSV.dev.
Be the first to review
Have you used this server?
Share your experience — it helps other developers decide.
Sign in to write a review.
Others in search / browser
Context7 Platform -- Up-to-date code documentation for LLMs and AI code editors
Monitor browser logs directly from Cursor and other MCP compatible IDEs.
The official MCP server implementation for the Perplexity API Platform
Production ready MCP server with real-time search, extract, map & crawl.
MCP Security Weekly
Get CVE alerts and security updates for Free Search MCP and similar servers.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.
A local-first, no-API-key Model Context Protocol server that gives any LLM (Claude, GPT, local Ollama, …) the ability to search the web, fetch and clean up pages, and read documents — without you signing up for a single search API.
It bundles together the best ideas from a handful of open-source MCPs into one Python package, and adds the LLM-ergonomics and reliability work they were each missing.
research("how does reciprocal rank fusion work", depth=3)
↓
# Research brief: how does reciprocal rank fusion work
_engines: duckduckgo, mojeek, googlenews · sources: 3 · ~3,400 tokens_
## Sources
- [1] Reciprocal rank fusion | Elasticsearch Reference — <https://…>
- [2] Hybrid Search Scoring (RRF) | Microsoft Learn — <https://…>
- [3] RRF explained in 4 mins — Medium — <https://…>
## Documents
…full Markdown bodies of each page, ready for the LLM to read…
One tool call. Three sources. No API key. No OPENAI_API_KEY-but-for-search
shakedown.
One command — the keyless engines work immediately, no signup, no key, no checkout (needs uv):
claude mcp add search -- uvx free-search-mcp # Claude Code
codex mcp add search -- uvx free-search-mcp # Codex
Any other MCP client: point it at the command uvx free-search-mcp (stdio). The
first run downloads the package from PyPI; every HTTP engine works with no
further setup.
Optional — browser-rendered engines (startpage, zhihu, …) and JS-heavy
page fetches need Chromium once:
uvx --from free-search-mcp playwright install chromium
Without it, HTTP search/fetch keep working, and any call that needs the browser returns that exact install command instead of a cryptic failure.
Configuration (all optional) lives in ~/.config/search-mcp/.env — see
Configuration.
curl -LsSf https://raw.githubusercontent.com/sweetcornna/free-search-mcp/main/scripts/install.sh | bash -s -- --client claude-code
It clones or updates the project under ~/.local/share/free-search-mcp,
installs uv, syncs dependencies, installs Chromium for rendered engines
(with OS deps on Linux), smoke-tests the server, and registers the search
MCP server in Claude Code user scope.
Other client targets:
curl -LsSf https://raw.githubusercontent.com/sweetcornna/free-search-mcp/main/scripts/install.sh | bash -s -- --client claude-desktop
curl -LsSf https://raw.githubusercontent.com/sweetcornna/free-search-mcp/main/scripts/install.sh | bash -s -- --client codex
curl -LsSf https://raw.githubusercontent.com/sweetcornna/free-search-mcp/main/scripts/install.sh | bash -s -- --client generic
curl -LsSf https://raw.githubusercontent.com/sweetcornna/free-search-mcp/main/scripts/install.sh | bash -s -- --client add-mcp
curl -LsSf https://raw.githubusercontent.com/sweetcornna/free-search-mcp/main/scripts/install.sh | bash -s -- --client all
Codex, Cursor, Cline, Continue, Zed, and generic agent guidance: docs/AGENT_USAGE.md.
Optional extras, any time (the defaults already work without them):
uv run search-mcp-admin # bilingual browser UI / 中英双语配置页: http://127.0.0.1:8765
uv run search-mcp-login zhihu # one-time Zhihu login (persists cookies)
Prefer containers? docker compose run --rm search-mcp. Claude Desktop and
other clients: see Install below.