Is standalone safe to install?

SaferSkills independently audited standalone (MCP Server) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

README.md · 1 file

README.md27.9 KB · Markdown

<p> <a href="README.zh-CN.md">中文</a> · <a href="#-quick-start">Quick Start</a> · <a href="#-features">Features</a> · <a href="#-architecture">Architecture</a> · <a href="#-comparison-with-alternatives">Comparison</a> · <a href="#-usage">Usage</a> · <a href="#-mcp-tools-reference">MCP Tools</a> · <a href="#-client-configuration">Client Config</a> · <a href="#-ai-coding-agent-plugin">Plugin</a> · <a href="#-contributing">Contributing</a> </p>

</div>

A self-contained MCP server that wraps SearXNG — a free, privacy-respecting metasearch engine that aggregates results from 200+ search engines.

🚀 Quick Start

Server mode — deploy once, connect from any client:

docker run -d --name searxng-mcp --restart unless-stopped \
  -p YOUR_PORT:8888 --memory=512m --cpus=1 \
  ghcr.io/whw23/searxng-http-mcp:latest

Then connect your client to http://YOUR_HOST:YOUR_PORT/mcp/. To enable API key auth, see Authentication.

Local mode — no server needed, run directly in your client:

docker run --rm -i --memory=512m --cpus=1 ghcr.io/whw23/searxng-http-mcp:latest --stdio

Add this as a stdio MCP server in your client — see Client Configuration for details.

uvx mode — if you already have SearXNG running (install guide):

uvx searxng-http-mcp

Set SEARXNG_URL to point to your SearXNG instance (default: http://127.0.0.1:8080).

✨ Features

Search

🔍 200+ search engines — Google, Bing, DuckDuckGo, Brave, and more via SearXNG
📂 30+ categories — news, images, videos, science, IT, and more
📄 Multi-page fanout — up to 5 pages per call
💡 Autocomplete suggestions — discover relevant search terms
🗂 Engine discovery — query available engines grouped by category
🎯 Token-efficient — results trimmed to essentials

Infrastructure

📦 Self-contained — SearXNG built into Docker image
🔄 Triple transport — HTTP server, Docker stdio, and uvx standalone
🔐 Authentication — x-api-key + HTTP Basic Auth
🌐 Reverse proxy — SearXNG Web UI on the same port
⚡ Dynamic tool descriptions — live category lists injected at startup
📐 Rich JSON Schema — enum constraints, range limits, and descriptions on every parameter

🏛 Architecture

graph LR
  Client(["client:YOUR_PORT"]) --> Expose(":8888")

  subgraph Container["🐳 Docker Container"]
    direction LR
    Expose --> Auth{Auth}
    Auth -->|/mcp| MCP[FastMCP Server]
    Auth -->|/*| Proxy[Reverse Proxy]
    MCP --> SearXNG[SearXNG :8080]
    Proxy --> SearXNG
  end

  style Expose fill:none,stroke:#2496ed,stroke-dasharray:5 5,color:#2496ed

  style Client fill:#4a90d9,color:#fff,stroke:#3a7bc8
  style Container fill:#f0f4f8,stroke:#2496ed,stroke-width:2px,color:#2496ed
  style Auth fill:#f5a623,color:#fff,stroke:#d4900e
  style MCP fill:#50c878,color:#fff,stroke:#3da85e
  style Proxy fill:#9b59b6,color:#fff,stroke:#8344a5
  style SearXNG fill:#e74c3c,color:#fff,stroke:#c0392b

📊 Comparison with Alternatives

<details> <summary>Why these five?</summary>

There are 20+ SearXNG MCP servers and many more general-purpose search MCPs. Most SearXNG wrappers only expose a basic search tool, leaving SearXNG's categories, autocomplete, and engine metadata unused. We picked five alternatives that each represent a distinct category:

88plug/searxng-mcp — richest tool surface among SearXNG MCPs (7 tools: rendered fetch, research mode, parallel queries)
ihor/mcp-searxng — most GitHub stars among SearXNG MCPs
open-webSearch — top free multi-engine alternative outside the SearXNG ecosystem (Bing, Baidu, DuckDuckGo, Brave, etc.)
exa-mcp-server — most popular commercial search API MCP
Perplexity MCP — commercial AI-powered search, highest star count in the search MCP space

</details>

<table> <thead> <tr> <th>Feature</th> <th>✨ This project</th> <th><a href="https://github.com/88plug/searxng-mcp">88plug/searxng-mcp</a></th> <th><a href="https://github.com/ihor-sokoliuk/mcp-searxng">ihor/mcp-searxng</a></th> <th><a href="https://github.com/Aas-ee/open-webSearch">open-webSearch</a></th> <th><a href="https://github.com/exa-labs/exa-mcp-server">exa-mcp-server</a></th> <th><a href="https://github.com/perplexityai/modelcontextprotocol">Perplexity MCP</a></th> </tr> </thead> <tbody> <tr><td colspan="7"><strong>Search</strong></td></tr> <tr><td>200+ engines via SearXNG</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>30+ search categories</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>Multi-page fanout</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>Autocomplete suggestions</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>Engine discovery tool</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>Dynamic tool descriptions</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td colspan="7"><strong>Infrastructure</strong></td></tr> <tr><td>Self-contained (built-in search)</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">✅</td><td align="center">N/A</td><td align="center">N/A</td></tr> <tr><td>Zero-install Docker deploy</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>HTTP + stdio transport</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td></tr> <tr><td>Authentication</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">✅</td><td align="center">✅</td></tr> <tr><td>Web UI reverse proxy</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>AI Coding Agent Plugin</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">❌</td><td align="center">✅</td></tr> <tr><td colspan="7"><strong>General</strong></td></tr> <tr><td>Free & open source</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">❌ (paid API)</td><td align="center">❌ (paid API)</td></tr> <tr><td>Privacy (self-hosted)</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">✅</td><td align="center">❌</td><td align="center">❌</td></tr> <tr><td>Language</td><td align="center">Python</td><td align="center">Python</td><td align="center">Node.js</td><td align="center">TypeScript</td><td align="center">TypeScript</td><td align="center">TypeScript</td></tr> <tr><td>GitHub Stars</td><td align="center"><a href="https://github.com/whw23/searxng_http_mcp"><img src="https://img.shields.io/github/stars/whw23/searxng_http_mcp?style=social" alt="stars"></a></td><td align="center"><a href="https://github.com/88plug/searxng-mcp"><img src="https://img.shields.io/github/stars/88plug/searxng-mcp?style=social" alt="stars"></a></td><td align="center"><a href="https://github.com/ihor-sokoliuk/mcp-searxng"><img src="https://img.shields.io/github/stars/ihor-sokoliuk/mcp-searxng?style=social" alt="stars"></a></td><td align="center"><a href="https://github.com/Aas-ee/open-webSearch"><img src="https://img.shields.io/github/stars/Aas-ee/open-webSearch?style=social" alt="stars"></a></td><td align="center"><a href="https://github.com/exa-labs/exa-mcp-server"><img src="https://img.shields.io/github/stars/exa-labs/exa-mcp-server?style=social" alt="stars"></a></td><td align="center"><a href="https://github.com/perplexityai/modelcontextprotocol"><img src="https://img.shields.io/github/stars/perplexityai/modelcontextprotocol?style=social" alt="stars"></a></td></tr> </tbody> </table>

<details> <summary>Why fewer tools?</summary>

MCP is designed for composition — clients connect multiple specialized servers, each doing one thing well. Some alternatives bundle URL fetching, rendered page extraction, multi-query fan-out, or research modes into the search server. We keep the tool surface to three (search, autocomplete, engine discovery) by design:

URL fetching is a separate concern. MCP clients already ship dedicated tools (WebFetch, Playwright MCP, Jina Reader). Bundling fetch into a search server mixes responsibilities and duplicates the client ecosystem.
Multi-query parallel search is client-side orchestration. LLM clients can fire multiple search calls in parallel — a search_many tool only adds token overhead for tool selection with no real benefit.
Research / synthesis belongs in the LLM layer. The model is the best synthesizer. Pushing multi-step research logic into the MCP server couples application concerns to infrastructure.

Instead we invest in what the alternatives above lack: complete SearXNG API coverage (categories, autocomplete, engine metadata — capabilities most wrappers leave on the table), self-contained deployment, authentication, Web UI reverse proxy, and AI coding agent plugin integration (Claude Code / Copilot CLI / Codex CLI).

</details>

📖 Usage

🌐 HTTP Mode (default)

# Without authentication
docker run -d --name searxng-mcp --restart unless-stopped \
  -p YOUR_PORT:8888 --memory=512m --cpus=1 \
  ghcr.io/whw23/searxng-http-mcp:latest

# With authentication
docker run -d --name searxng-mcp --restart unless-stopped \
  -p YOUR_PORT:8888 --memory=512m --cpus=1 \
  -e API_KEY=your-secret-key \
  ghcr.io/whw23/searxng-http-mcp:latest

<table> <tr><td>🔗 <strong>MCP Endpoint</strong></td><td><code>http://YOUR_HOST:YOUR_PORT/mcp/</code></td></tr> <tr><td>🖥 <strong>SearXNG Web UI</strong></td><td><code>http://YOUR_HOST:YOUR_PORT/</code></td></tr> </table>

📡 stdio Mode

docker run --rm -i --memory=512m --cpus=1 \
  ghcr.io/whw23/searxng-http-mcp:latest --stdio

No ports exposed. Communication via stdin/stdout. SearXNG runs internally for the MCP tools.

🐍 uvx Mode

# Connect to a local SearXNG instance (default: http://127.0.0.1:8080)
uvx searxng-http-mcp

# Connect to a remote SearXNG instance
SEARXNG_URL=http://YOUR_SEARXNG_HOST:YOUR_SEARXNG_PORT uvx searxng-http-mcp

Requires Python 3.14+ and an existing SearXNG instance. No Docker needed.

⚙️ Environment Variables

<table> <thead> <tr><th>Variable</th><th>Default</th><th>Description</th></tr> </thead> <tbody> <tr><td><code>API_KEY</code></td><td><em>(empty, no auth)</em></td><td>API key for authentication</td></tr> <tr><td><code>SEARXNG_URL</code></td><td><code>http://127.0.0.1:8080</code></td><td>SearXNG instance URL (for uvx/standalone mode)</td></tr> </tbody> </table>

🔐 Authentication

When API_KEY is set, all requests require one of:

`x-api-key` header — for MCP clients: x-api-key: your-key
HTTP Basic Auth — for browsers

[!TIP] Browser Login: When accessing the Web UI with API_KEY enabled, the browser will show a login dialog. Leave the username empty and enter your API key as the password.

<img src="assets/browser_login.svg" alt="Browser Login Dialog" width="420">

When API_KEY is not set, all requests are open.

🔧 MCP Tools Reference

<details> <summary>🔍 <code>search</code> — Search the web using SearXNG</summary>

<br>

Aggregates results from 200+ search engines with privacy.

<table> <thead> <tr><th>Parameter</th><th>Type</th><th>Required</th><th>Default</th><th>Description</th></tr> </thead> <tbody> <tr><td><code>query</code></td><td>string</td><td>yes</td><td>—</td><td>The search query to use</td></tr> <tr><td><code>categories</code></td><td>string</td><td>no</td><td>""</td><td>Comma-separated category names (e.g., <code>general,news,science</code>)</td></tr> <tr><td><code>engines</code></td><td>string</td><td>no</td><td>""</td><td>Comma-separated engine names (e.g., <code>google,arxiv,wikipedia</code>)</td></tr> <tr><td><code>language</code></td><td>string</td><td>no</td><td>""</td><td>Search language code (e.g., <code>en</code>, <code>zh</code>, <code>ja</code>)</td></tr> <tr><td><code>time_range</code></td><td>enum</td><td>no</td><td>null</td><td><code>day</code>, <code>week</code>, <code>month</code>, <code>year</code></td></tr> <tr><td><code>safesearch</code></td><td>enum</td><td>no</td><td>0</td><td><code>0</code>=off, <code>1</code>=moderate, <code>2</code>=strict</td></tr> <tr><td><code>pageno</code></td><td>int ≥1</td><td>no</td><td>1</td><td>Starting page number</td></tr> <tr><td><code>pages</code></td><td>int 1–5</td><td>no</td><td>1</td><td>Number of pages to fetch in parallel</td></tr> <tr><td><code>max_results</code></td><td>int 1–100</td><td>no</td><td>10</td><td>Maximum number of results to return</td></tr> <tr><td><code>format</code></td><td>enum</td><td>no</td><td>compact</td><td><code>compact</code> (title/url/content) or <code>full</code> (+ engines/score/category/date)</td></tr> </tbody> </table>

Returns: results, answers, suggestions, corrections, infoboxes.

</details>

<details> <summary>💡 <code>autocomplete</code> — Get search query suggestions</summary>

<br>

<table> <thead> <tr><th>Parameter</th><th>Type</th><th>Required</th><th>Description</th></tr> </thead> <tbody> <tr><td><code>query</code></td><td>string</td><td>yes</td><td>Partial query string to get suggestions for</td></tr> </tbody> </table>

</details>

<details> <summary>🗂 <code>engine_info</code> — Discover available engines and categories</summary>

<br>

No parameters. Returns the list of enabled engines grouped by category.

Returns:

{
  "categories": ["general", "images", "videos", "news", ...],
  "engines": ["google", "bing", "duckduckgo", ...],
  "category_engines": {
    "general": ["google", "bing", "duckduckgo", "brave", ...],
    "science": ["arxiv", "google scholar", "pubmed", ...],
    ...
  }
}

Use this to discover what engines are available before calling search with specific engines or categories filters.

</details>

🔌 Client Configuration

<details> <summary><img src="assets/icons/claude.svg" height="16" alt="Claude"> <b>Claude Desktop</b></summary>

Server mode — edit ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "searxng": {
      "url": "http://YOUR_HOST:YOUR_PORT/mcp/",
      "headers": {
        "x-api-key": "your-secret-key"
      }
    }
  }
}

Local mode:

{
  "mcpServers": {
    "searxng": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]
    }
  }
}

uvx mode:

{
  "mcpServers": {
    "searxng": {
      "command": "uvx",
      "args": ["searxng-http-mcp"],
      "env": { "SEARXNG_URL": "http://YOUR_SEARXNG_HOST:YOUR_SEARXNG_PORT" }
    }
  }
}

</details>

<details> <summary><img src="assets/icons/claude.svg" height="16" alt="Claude"> <b>Claude Code</b></summary>

Server mode:

claude mcp add --transport http --header "x-api-key: your-secret-key" searxng http://YOUR_HOST:YOUR_PORT/mcp/

Local mode:

claude mcp add --transport stdio searxng -- docker run --rm -i --memory=512m --cpus=1 ghcr.io/whw23/searxng-http-mcp:latest --stdio

uvx mode:

claude mcp add --transport stdio searxng -- uvx searxng-http-mcp

</details>

<details> <summary><img src="assets/icons/codex.svg" height="16" alt="Codex"> <b>Codex</b></summary>

Server mode — add to ~/.codex/config.toml:

[mcp_servers.searxng]
url = "http://YOUR_HOST:YOUR_PORT/mcp/"
http_headers = { "x-api-key" = "your-secret-key" }

Local mode:

[mcp_servers.searxng]
command = "docker"
args = ["run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]

uvx mode:

[mcp_servers.searxng]
command = "uvx"
args = ["searxng-http-mcp"]

</details>

<details> <summary><img src="assets/icons/cursor.svg" height="16" alt="Cursor"> <b>Cursor</b></summary>

Server mode — edit .cursor/mcp.json:

{
  "mcpServers": {
    "searxng": {
      "url": "http://YOUR_HOST:YOUR_PORT/mcp/",
      "headers": {
        "x-api-key": "your-secret-key"
      }
    }
  }
}

Local mode:

{
  "mcpServers": {
    "searxng": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]
    }
  }
}

uvx mode:

{
  "mcpServers": {
    "searxng": {
      "command": "uvx",
      "args": ["searxng-http-mcp"],
      "env": { "SEARXNG_URL": "http://YOUR_SEARXNG_HOST:YOUR_SEARXNG_PORT" }
    }
  }
}

</details>

<details> <summary><img src="assets/icons/githubcopilot.svg" height="16" alt="Copilot"> <b>VS Code Copilot</b></summary>

Server mode — add to .vscode/mcp.json:

{
  "servers": {
    "searxng": {
      "type": "http",
      "url": "http://YOUR_HOST:YOUR_PORT/mcp/",
      "headers": {
        "x-api-key": "your-secret-key"
      }
    }
  }
}

Local mode:

{
  "servers": {
    "searxng": {
      "type": "stdio",
      "command": "docker",
      "args": ["run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]
    }
  }
}

uvx mode:

{
  "servers": {
    "searxng": {
      "type": "stdio",
      "command": "uvx",
      "args": ["searxng-http-mcp"],
      "env": { "SEARXNG_URL": "http://YOUR_SEARXNG_HOST:YOUR_SEARXNG_PORT" }
    }
  }
}

</details>

<details> <summary><img src="assets/icons/windsurf.svg" height="16" alt="Windsurf"> <b>Windsurf</b></summary>

Server mode — add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "searxng": {
      "serverUrl": "http://YOUR_HOST:YOUR_PORT/mcp/",
      "headers": {
        "x-api-key": "your-secret-key"
      }
    }
  }
}

Local mode:

{
  "mcpServers": {
    "searxng": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]
    }
  }
}

uvx mode:

{
  "mcpServers": {
    "searxng": {
      "command": "uvx",
      "args": ["searxng-http-mcp"],
      "env": { "SEARXNG_URL": "http://YOUR_SEARXNG_HOST:YOUR_SEARXNG_PORT" }
    }
  }
}

</details>

<details> <summary><img src="assets/icons/cline.svg" height="16" alt="Cline"> <b>Cline</b></summary>

Configure via Cline's MCP settings panel in VS Code (Cline > MCP Servers > Add).

Server mode:

{
  "mcpServers": {
    "searxng": {
      "url": "http://YOUR_HOST:YOUR_PORT/mcp/",
      "headers": {
        "x-api-key": "your-secret-key"
      }
    }
  }
}

Local mode:

{
  "mcpServers": {
    "searxng": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]
    }
  }
}

uvx mode:

{
  "mcpServers": {
    "searxng": {
      "command": "uvx",
      "args": ["searxng-http-mcp"],
      "env": { "SEARXNG_URL": "http://YOUR_SEARXNG_HOST:YOUR_SEARXNG_PORT" }
    }
  }
}

</details>

<details> <summary><img src="assets/icons/opencode.svg" height="16" alt="OpenCode"> <b>OpenCode</b></summary>

Server mode — edit opencode.json:

{
  "mcp": {
    "searxng": {
      "type": "remote",
      "url": "http://YOUR_HOST:YOUR_PORT/mcp/",
      "headers": {
        "x-api-key": "your-secret-key"
      }
    }
  }
}

Local mode:

{
  "mcp": {
    "searxng": {
      "type": "local",
      "command": ["docker", "run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]
    }
  }
}

uvx mode:

{
  "mcp": {
    "searxng": {
      "type": "local",
      "command": ["uvx", "searxng-http-mcp"]
    }
  }
}

</details>

<details> <summary><img src="assets/icons/hermes.svg" height="16" alt="Hermes"> <b>Hermes Agent</b></summary>

Server mode — edit ~/.hermes/config.yaml:

mcp_servers:
  searxng:
    url: "http://YOUR_HOST:YOUR_PORT/mcp/"
    headers:
      x-api-key: "your-secret-key"

Local mode:

mcp_servers:
  searxng:
    command: "docker"
    args: ["run", "--rm", "-i", "--memory=512m", "--cpus=1", "ghcr.io/whw23/searxng-http-mcp:latest", "--stdio"]

uvx mode:

mcp_servers:
  searxng:
    command: "uvx"
    args: ["searxng-http-mcp"]

</details>

🧩 AI Coding Agent Plugin

Compatible with Claude Code, GitHub Copilot CLI, and OpenAI Codex CLI. Add the marketplace, then install the plugin that fits your setup:

/plugin marketplace add whw23/searxng_http_mcp

All plugins include:

🔍 `/web-search-via-searxng` skill — search the web with category selection guide and usage rules
🤖 `web-searcher` agent — research subagent with:
🌐 Multi-language parallel search — auto-detects query language, searches in multiple languages for broader coverage
🛡 Source credibility tiers — Tier 1/2/3 classification with AI poisoning defense
✅ Cross-validation — key facts require 2+ independent sources
🔍 WebFetch deep reading — reads full pages from credible sources, not just snippets
🔄 Multi-round iterative search — up to 3 rounds with automatic deepening on gaps
⏱ Timeliness evaluation — flags outdated content, states source date ranges
📝 Footnote-style references — structured output with inline source citations

<details> <summary>🐍 <b>Standalone mode</b> — uvx, no Docker, bring your own SearXNG</summary>

<br>

/plugin install searxng-http-mcp@searxng-http-mcp-standalone

Runs via uvx. Requires Python 3.14+ and an existing SearXNG instance. Set SEARXNG_URL in ~/.claude/settings.json:

{
  "env": {
    "SEARXNG_URL": "http://YOUR_SEARXNG_HOST:YOUR_SEARXNG_PORT"
  }
}

</details>

<details> <summary>🐳 <b>Local mode</b> — Docker stdio, zero config</summary>

<br>

/plugin install searxng-http-mcp@searxng-http-mcp

Runs SearXNG in a local Docker container via stdio. Requires Docker installed.

</details>

<details> <summary>🌐 <b>Remote mode</b> — connect to a deployed server via HTTP</summary>

<br>

/plugin install searxng-http-mcp@searxng-http-mcp-remote

Connects to a deployed SearXNG MCP server. Requires env vars SEARXNG_MCP_URL and SEARXNG_API_KEY.

Add to ~/.claude/settings.json under the env field:

{
  "env": {
    "SEARXNG_MCP_URL": "http://YOUR_HOST:YOUR_PORT/mcp/",
    "SEARXNG_API_KEY": "your-api-key"
  }
}

Then restart Claude Code.

</details>

🛠 SearXNG Configuration

<br>

Access the SearXNG Web UI at http://YOUR_HOST:YOUR_PORT/ to configure search engines, languages, and other settings. Changes persist during the container's lifetime.

</details>

<details> <summary>💾 <b>Via Volume Mount</b> — persistent configuration</summary>

<br>

Mount the SearXNG config directory for persistent configuration:

docker run -d --name searxng-mcp --restart unless-stopped \
  -p YOUR_PORT:8888 --memory=512m --cpus=1 \
  -v /path/to/searxng-config:/etc/searxng \
  ghcr.io/whw23/searxng-http-mcp:latest

SearXNG generates settings.yml on first startup. The container automatically enables JSON format output required by MCP tools.

</details>

🏗 Build from Source

git clone https://github.com/whw23/searxng_http_mcp.git
cd searxng_http_mcp
docker build -t searxng-http-mcp:local .
docker run -d --name searxng-mcp --restart unless-stopped \
  -p YOUR_PORT:8888 --memory=512m --cpus=1 \
  searxng-http-mcp:local

🤝 Contributing

See [CONTRIBUTING.md](CONTRIBUTING.md) for the full workflow, CI requirements, and development setup.

🍴 Fork the repository and enable GitHub Actions in your fork
🌿 Create a feature branch from dev
✍️ Make your changes
✅ Run tests: pytest tests/ -v — CI must pass in your fork before opening a PR
📬 Submit a PR to dev

Development happens on the dev branch. Merges to main trigger image builds.

📄 License

MIT — MCP server code.

SearXNG itself is licensed under AGPL-3.0-or-later.

standalone— mcp server

Is standalone safe to install?

🚀 Quick Start

✨ Features

Search

Infrastructure

🏛 Architecture

📊 Comparison with Alternatives

📖 Usage

🌐 HTTP Mode (default)

📡 stdio Mode

🐍 uvx Mode

⚙️ Environment Variables

🔐 Authentication

🔧 MCP Tools Reference

🔌 Client Configuration

🧩 AI Coding Agent Plugin

🛠 SearXNG Configuration

🏗 Build from Source

🤝 Contributing

📄 License

Audit the pieces. Scan the whole. Decide.