Agenten stärken
Installation vom Herausgeber. Kein PlugMCP-Endpunkt.
Host a copy with our keys and logs.
Noch kein paketiertes Release in der Registry. Folgen Sie der README des Repositorys.
Remote-URL oder ein Befehl auf Ihrem Rechner? Worin sich die beiden Installationen unterscheiden
Auf PlugMCP ausführen
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
Starten
unlimited-search is a Python CLI and MCP server for reading public web content when a normal direct fetch is not enough. It combines platform public routes, browser-like HTTP identities, content fallbacks, public archive fallbacks, and media metadata extraction behind one local tool.
It is built for agents and automation that need usable text from public URLs. It is not intended to bypass logins, paywalls, CAPTCHA, private networks, account restrictions, IP bans, or access controls.
Prerequisite: Python 3.12 or newer.
python -m pip install unlimited-search
unlimited-search read https://en.wikipedia.org/wiki/OpenAI --max-content-chars 800
The command returns JSON with page content, a verdict, request metadata, and an attempt trace.
For MCP clients, install the package and register the stdio server:
{
"mcpServers": {
"unlimited-search": {
"command": "unlimited-search",
"args": ["serve"]
}
}
}
See Platform coverage for the full support matrix and known gaps.
Install from PyPI:
python -m pip install unlimited-search
Update:
python -m pip install --upgrade unlimited-search
Remove:
python -m pip uninstall unlimited-search
unlimited-search help
unlimited-search <command> [arguments]
Commands:
serve Start the MCP stdio server
read <url> Read a public URL
diagnose <url> Diagnose access without full content
media <url> Extract public media metadata
help Show this help
Common examples:
unlimited-search read https://example.com --max-content-chars 1000
unlimited-search diagnose https://example.com
unlimited-search media https://www.youtube.com/watch?v=dQw4w9WgXcQ
Fallback-focused smoke tests:
unlimited-search read https://example.com --no-public-routes --max-attempts 0 --max-content-chars 500
unlimited-search read http://www.whitehouse.gov/1600/presidents/barackobama --no-public-routes --max-attempts 1 --max-content-chars 500
Useful flags:
The MCP server runs over stdio:
unlimited-search serve
Available tools:
More configuration examples are in MCP configuration.
read_public_url tries the least invasive public routes first, then progressively broader recovery paths:
external.alternate/feed, /rss, and /atom.xmlyt-dlp metadata extraction for known public media hosts.Fallback successes are reported as weak_ok or suspect_ok, not strong_ok, because recovered content can be incomplete, stale, or metadata-only.
unlimited-search is a public-content reader. It should stop or report failure when a target requires authentication, payment, CAPTCHA, private network access, or hard anti-abuse bypassing.
The reader rejects private, loopback, link-local, multicast, reserved, and metadata-service targets by default. Redirects are checked before fallback routes are allowed to continue.
Returned HTML, JSON, RSS, archive text, and metadata should be treated as untrusted content.
uv sync --extra dev
uv run unlimited-search read https://example.com --max-content-chars 300
uv run unlimited-search serve
uv run pytest
uv build
Run the live eval set when changing routes or fallback behavior:
uv run python scripts/run_eval.py --list
uv run python scripts/run_eval.py
uv run python scripts/run_eval.py --markdown eval-results/report.md --csv eval-results/report.csv
uv run python scripts/run_eval.py --baseline eval-results/eval-20260707T000000Z.jsonl --fail-on-regression
The default eval cases live in scripts/eval_urls.yaml. Difficult sites such as NamuWiki, TikTok, Naver Search, Amazon, and Google Scholar are optional so remote blocking or rate limits do not fail the whole run.
MCP server and CLI for reading public web pages through resilient public routes.
Tools
Noch keine Tool-Liste. Remote-Server erhalten eine nach dem ersten Health-Check.
FAQ
Kopieren Sie auf dieser Seite die stdio-Konfiguration für Cursor, Claude Desktop, Claude Code oder VS Code. Sie nutzt die Installation des Herausgebers, keinen PlugMCP-Proxy.
In diesem Eintrag ist keine erforderliche Umgebungsvariable aufgeführt.
Unlimited Search läuft lokal über stdio.
Zuletzt geprüft: 2026-10-28. MCP server and CLI for reading public web pages through resilient public routes.