Install
$ agentstack add mcp-unitedideas-nothumansearch-mcp ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v1.3.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v1.3.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Not Human Search — MCP server
A remote MCP server for searching the agentic web. Discover APIs, services, and tools AI agents can actually call — ranked by agentic readiness score (0-100).
Live endpoint: https://nothumansearch.ai/mcp (streamable-http transport, no install)
Registry listing: ai.nothumansearch/search
Install in one command
claude mcp add --transport http nothumansearch https://nothumansearch.ai/mcp
Any MCP-compatible client (Claude Code, Cline, Cursor, Zed, Goose, etc.) can connect to the same endpoint.
Tools
| Tool | Purpose | |------|---------| | search_agents | Find agent-ready websites, APIs, and services by keyword, category, or minimum readiness score. | | get_site_details | Get the full 7-signal readiness report for a specific domain. | | get_stats | Index statistics: total sites, average score, top category. |
What "agentic readiness" means
Every site in the index is scored 0-100 based on 7 signals that determine how cleanly an AI agent can interact with it:
llms.txt(25 pts)ai-plugin.json(20 pts)- OpenAPI spec (20 pts)
- Structured API (15 pts)
- MCP server (10 pts)
robots.txtAI bot rules (5 pts)- Schema.org markup (5 pts)
Other interfaces
- Website: https://nothumansearch.ai
- REST search API:
GET https://nothumansearch.ai/api/v1/search?q= - On-demand check (CI-friendly):
POST https://nothumansearch.ai/api/v1/check - Embeddable badge:
https://nothumansearch.ai/badge/{domain}.svg - Submit a site:
POST https://nothumansearch.ai/api/v1/submit
Manifest
{
"$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
"name": "ai.nothumansearch/search",
"title": "Not Human Search",
"description": "Search engine for AI agents. Find websites and APIs ranked by agentic readiness.",
"version": "1.0.0",
"websiteUrl": "https://nothumansearch.ai",
"remotes": [
{ "type": "streamable-http", "url": "https://nothumansearch.ai/mcp" }
]
}
Deeper docs
See [USAGE.md](./USAGE.md) for a practical guide: tool arguments, recipes ("find me an email API that agents can actually call"), scoring details, and design principles.
License
The MCP server source lives in a private repo. This README is MIT-licensed — use the claude mcp add command freely.
Source & license
This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: unitedideas
- Source: unitedideas/nothumansearch-mcp
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v1.3.0 Imported from the upstream source.