searxng-search
Search the web privately using the SearXNG meta-search engine at {{SEARXNG_HOST}}:{{SEARXNG_PORT}} with aggregated results from multiple search providers.
Best use case
searxng-search is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Search the web privately using the SearXNG meta-search engine at {{SEARXNG_HOST}}:{{SEARXNG_PORT}} with aggregated results from multiple search providers.
Teams using searxng-search should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/searxng-search/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How searxng-search Compares
| Feature / Agent | searxng-search | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Search the web privately using the SearXNG meta-search engine at {{SEARXNG_HOST}}:{{SEARXNG_PORT}} with aggregated results from multiple search providers.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# SearXNG Search Skill
SearXNG meta-search engine is available at `http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}` within the Docker network.
## Basic Search
Perform a web search and get JSON results:
```bash
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=your+search+query&format=json"
```
## Search with Parameters
Fine-tune results using categories, engines, and language filters:
```bash
# Search specific categories
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=machine+learning&format=json&categories=science,it"
# Search with specific engines
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=python+tutorial&format=json&engines=google,duckduckgo,brave"
# Search in a specific language
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=bonjour&format=json&language=fr"
# Restrict to a time range
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=latest+news&format=json&time_range=week"
```
## Image Search
```bash
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=landscape+photography&format=json&categories=images"
```
## News Search
```bash
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=tech+news&format=json&categories=news&time_range=day"
```
## Paginated Results
```bash
# Get page 2 of results (pageno is 1-indexed)
curl "http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/search?q=docker+tutorial&format=json&pageno=2"
```
## Response Structure
Results are returned as JSON with this structure:
```json
{
"query": "your search query",
"number_of_results": 12345,
"results": [
{
"url": "https://example.com/article",
"title": "Article Title",
"content": "Snippet of the search result content...",
"engine": "google",
"category": "general",
"score": 8.5
}
],
"suggestions": ["related query 1", "related query 2"],
"infoboxes": []
}
```
## Available Categories
- `general` — standard web search
- `images` — image search
- `videos` — video search
- `news` — news articles
- `music` — music and audio
- `files` — file downloads
- `it` — IT and tech-specific results
- `science` — academic and scientific content
- `social_media` — social media posts
## Available Time Ranges
- `day` — last 24 hours
- `week` — last 7 days
- `month` — last 30 days
- `year` — last 365 days
## Tips for AI Agents
- Always append `&format=json` to get machine-readable results instead of HTML.
- Combine multiple categories with commas for broader search: `categories=general,it,science`.
- Use `time_range=week` or `time_range=month` when you need recent information.
- Parse the `results` array and use the `score` field to rank relevance.
- For factual queries, cross-reference results from multiple entries in the `results` array.
- Use `suggestions` from the response to refine follow-up searches.
- Check SearXNG health at `http://{{SEARXNG_HOST}}:{{SEARXNG_PORT}}/healthz`.
- Respect rate limits — add a short delay between rapid consecutive searches.Related Skills
weaviate-search
Perform hybrid vector and keyword search using Weaviate at {{WEAVIATE_HOST}}:{{WEAVIATE_PORT}}.
meilisearch-index
Index and search documents with typo-tolerant full-text search using Meilisearch at {{MEILISEARCH_HOST}}:{{MEILISEARCH_PORT}}.
elasticsearch-search
Full-text search and analytics with Elasticsearch
youtube-growth
Act as an expert YouTube Strategy Consultant. Apply the Creator Unlock N.I.C.E.R. Framework for conducting channel audits, niche validation, and data-backed video ideation/thumbnail generation.
xyops-automate
Build and manage automation pipelines using xyOps at {{XYOPS_HOST}}:{{XYOPS_PORT}}.
xml-parse
Parse and transform XML/HTML documents using command-line tools in the shared volume at {{SHARED_VOLUME}}.
woodpecker-ci
Lightweight container-native CI/CD with Woodpecker
whisper-transcribe
Transcribe audio and video files to text using the Whisper speech-to-text API at {{WHISPER_HOST}}:{{WHISPER_PORT}}.
web-interface-guidelines
Checklist for reviewing UI code for compliance with comprehensive web interface, accessibility, performance, and content guidelines — based on Vercel's Web Interface Guidelines.
web-design-reviewer
Inspect web interfaces for layout, responsive, accessibility, and visual issues, then apply targeted source code fixes and re-verify results.
watchtower-update
Auto-update Docker containers using Watchtower.
vaultwarden-manage
Self-hosted password management with Vaultwarden