Skip to main content
Glama
ngc-shj

SearXNG MCP Server

by ngc-shj
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

SearXNG MCP Server

A bridge to use SearXNG as an MCP server from Claude Code.

日本語版 README はこちら

Features

  • searxng_search: Web search with multiple engines

  • searxng_images: Image search

  • searxng_news: News search

Related MCP server: SearxNG MCP Server

Prerequisites

  1. SearXNG instance running with JSON format enabled

    Important: SearXNG's default configuration only allows HTML format. To use this MCP server, you must enable JSON format by mounting a custom configuration file.

    First, create your configuration file from the example:

    cp searxng-config/settings.yml.example searxng-config/settings.yml
# Edit settings.yml to set your own secret_key

    Then start SearXNG with the configuration:

    # Using Docker (recommended)
docker run -d \
  -p 8080:8080 \
  -v $(pwd)/searxng-config/settings.yml:/etc/searxng/settings.yml:ro \
  -e SEARXNG_BASE_URL=http://localhost:8080/ \
  searxng/searxng

# Or install locally
# See: https://docs.searxng.org/admin/installation.html

    The searxng-config/settings.yml.example file includes the necessary configuration to enable JSON format:

    use_default_settings: true

search:
  formats:
    - html
    - json

server:
  secret_key: "your-secret-key-here"
  limiter: false
  image_proxy: true

  2. Verify SearXNG is running with JSON format enabled

    curl "http://localhost:8080/search?q=test&format=json"

    If you get a 403 Forbidden error, the JSON format is not enabled. Make sure you mounted the configuration file correctly.

Installation

cd searxng-mcp-server
npm install
npm run build

Claude Code Configuration

Method 1: Using CLI (Recommended)

# Add to local scope (current project)
claude mcp add --transport stdio searxng -- node /path/to/searxng-mcp-server/dist/index.js

# Add to user scope (all projects)
claude mcp add --transport stdio searxng --scope user -- node /path/to/searxng-mcp-server/dist/index.js

To add environment variables:

claude mcp add --transport stdio searxng \
  --env SEARXNG_BASE_URL=http://localhost:8080 \
  -- node /path/to/searxng-mcp-server/dist/index.js

Method 2: Manual Configuration

Project scope (.mcp.json in project root):

{
  "mcpServers": {
    "searxng": {
      "command": "node",
      "args": ["/path/to/searxng-mcp-server/dist/index.js"],
      "env": {
        "SEARXNG_BASE_URL": "http://localhost:8080"
      }
    }
  }
}

User scope (~/.claude.json):

{
  "mcpServers": {
    "searxng": {
      "command": "node",
      "args": ["/path/to/searxng-mcp-server/dist/index.js"],
      "env": {
        "SEARXNG_BASE_URL": "http://localhost:8080"
      }
    }
  }
}

Verify Installation

# List configured MCP servers
claude mcp list

# Inside Claude Code
/mcp

Auto-approve Tool Calls (Optional)

By default, Claude Code asks for confirmation each time a SearXNG tool is called. To skip confirmations, add the following to ~/.claude/settings.json:

{
  "permissions": {
    "allow": [
      "mcp__searxng__searxng_search",
      "mcp__searxng__searxng_images",
      "mcp__searxng__searxng_news"
    ]
  }
}

Environment Variables

Variable

Default

Description

SEARXNG_BASE_URL

http://localhost:8080

SearXNG server URL

Usage Examples

From Claude Code:

Web Search

Search for "Rust programming language features" using SearXNG

Image Search

Find images of "Mount Fuji" using SearXNG

News Search

Get latest news about "AI developments" using SearXNG

Troubleshooting

403 Forbidden error with JSON format

If you get a 403 Forbidden error when using format=json, the JSON format is not enabled in your SearXNG instance. Make sure you:

  1. Mount the configuration file when starting Docker:

    docker run -d \
  -p 8080:8080 \
  -v $(pwd)/searxng-config/settings.yml:/etc/searxng/settings.yml:ro \
  -e SEARXNG_BASE_URL=http://localhost:8080/ \
  searxng/searxng

  2. Verify the configuration includes JSON format:

    docker exec <container_id> grep -A 3 "formats" /etc/searxng/settings.yml

Cannot connect to SearXNG

# Check if SearXNG is running
curl "http://localhost:8080/search?q=test&format=json"

# If not running (Docker) - make sure to mount the config file
docker run -d \
  -p 8080:8080 \
  -v $(pwd)/searxng-config/settings.yml:/etc/searxng/settings.yml:ro \
  -e SEARXNG_BASE_URL=http://localhost:8080/ \
  searxng/searxng

No results returned

SearXNG aggregates results from multiple search engines. If no results are returned:

  1. Check if your SearXNG instance has engines enabled

  2. Try specifying engines: engines: "google,bing,duckduckgo"

  3. Check SearXNG logs for errors

MCP server not showing up

# Verify server is registered
claude mcp list

# Check server health
claude mcp get searxng

License

MIT

Available Tools

3 tools
searxng_imagesB

Search for images using SearXNG. Returns image results with URLs and thumbnails.

ParametersJSON Schema
NameRequiredDescriptionDefault
queryYesThe image search query
enginesNoComma-separated engine names to use
languageNoSearch language (e.g., 'en', 'ja')
pagenoNoPage number (default: 1)

TDQS

B3.1/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden of behavioral disclosure. It mentions that the tool 'Returns image results with URLs and thumbnails,' which adds some context about output format. However, it lacks details on rate limits, authentication needs, error handling, or other behavioral traits. The description doesn't contradict annotations (none provided).

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is appropriately sized and front-loaded with the core purpose in the first sentence. It consists of two concise sentences that earn their place by stating the action and output. There is zero waste or redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's moderate complexity (4 parameters, no output schema, no annotations), the description is minimally adequate. It covers the purpose and output format but lacks details on usage guidelines, behavioral traits, and parameter nuances. Without an output schema, it should ideally explain return values more thoroughly, but it provides a basic overview.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all parameters (query, engines, language, pageno). The description adds no additional meaning beyond what the schema provides, such as examples or usage tips. Baseline 3 is appropriate when the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Search for images using SearXNG' specifies the action and resource. It distinguishes from sibling tools (searxng_news, searxng_search) by focusing on images, though it doesn't explicitly mention the distinction. The description is specific but lacks explicit sibling differentiation.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives. It doesn't mention when to choose this over searxng_search or searxng_news, nor does it specify any prerequisites or exclusions. Usage is implied by the purpose but not explicitly stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

searxng_newsB

Search for news articles using SearXNG. Returns recent news with titles, URLs, and summaries.

ParametersJSON Schema
NameRequiredDescriptionDefault
queryYesThe news search query
enginesNoComma-separated engine names to use
languageNoSearch language (e.g., 'en', 'ja')
pagenoNoPage number (default: 1)

TDQS

B3.1/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It mentions 'Returns recent news' but doesn't disclose behavioral traits such as rate limits, authentication needs, pagination behavior, or error handling. For a search tool with no annotation coverage, this leaves significant gaps in understanding how it operates.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise and front-loaded: two sentences that directly state the tool's function and output. Every sentence earns its place by providing essential information without redundancy or unnecessary details.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's moderate complexity (4 parameters, no output schema, no annotations), the description is minimally adequate. It covers the basic purpose and output but lacks details on behavioral traits, usage guidelines, and parameter nuances. With no output schema, it should ideally explain return values more thoroughly, but it does mention titles, URLs, and summaries.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all parameters (query, engines, language, pageno). The description adds no additional meaning beyond what the schema provides, such as examples or constraints. Baseline 3 is appropriate when the schema handles parameter documentation effectively.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Search for news articles using SearXNG' specifies the verb ('search') and resource ('news articles'), and 'Returns recent news with titles, URLs, and summaries' adds output details. It distinguishes from sibling tools like 'searxng_images' and 'searxng_search' by focusing on news, though it doesn't explicitly compare to them.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives. It mentions 'news articles' but doesn't specify scenarios (e.g., breaking news, topic monitoring) or contrast with siblings like 'searxng_search' for general web searches. Usage is implied by the context but not explicitly stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 3 tool updatesv1.0.0
    • First observedsearxng_images
    • First observedsearxng_news
    • First observedsearxng_search
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.5/5.0

Scored across 3 tools

Disambiguation5/5

Each tool has a clearly distinct purpose targeting different content types: images, news articles, and general web search. The descriptions specify different result formats (thumbnails vs. summaries vs. snippets), making them easily distinguishable with no overlap.

Naming Consistency5/5

All tools follow a perfect verb_noun pattern with 'searxng_' prefix and consistent snake_case: searxng_images, searxng_news, searxng_search. The naming is completely predictable and uniform throughout the set.

Tool Count3/5

With only 3 tools, the set feels somewhat thin for a search server, potentially limiting functionality. While it covers basic search types, more specialized operations (e.g., video search, advanced filters) might be missing, making it borderline minimal.

Completeness4/5

The tools provide good coverage for core search functionality across three major content types. However, there are minor gaps such as lack of video search, advanced query parameters, or result filtering tools that could enhance the search experience.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • SearXNG MCP Server

    The-AI-Workshops
    F
    license
    Not graded
    quality
    D
    maintenance
    Enables AI agents to perform privacy-respecting web searches through SearXNG, with support for multiple search engines, categories, and advanced filtering options.
    26
    -

  • SearxNG MCP Server

    IceWreck
    F
    license
    A
    quality
    D
    maintenance
    Enables privacy-respecting web, image, video, and news searches through SearxNG metasearch engine, with options for URL fetching and time filtering.
    5
    96 PyPI
    -

  • SearXNG MCP Server

    benreichman
    F
    license
    Not graded
    quality
    D
    maintenance
    Enables web search and content scraping from multiple engines via a local SearXNG instance, allowing AI assistants to retrieve and extract web content.
    1
    -

  • web-search-mcp

    LMilazzo
    F
    license
    A
    quality
    B
    maintenance
    Enables private, free web searches through a self-hosted SearXNG instance, replacing default search tools without API keys, rate limits, or third-party tracking.
    1
    -

Appeared in Searches