Use free SearXNG web search APIs for agent-friendly, privacy-first, and high-volume search tasks.
npx skills add https://github.com/besoeasy/open-skills --skill web-search-api
Free, unlimited web search API for AI agents — no costs, no rate limits, no tracking. Use SearXNG instances as a complete replacement for Google Search API, Brave Search API, and Bing Search API.
💰 Cost savings:
Perfect for AI agents that need:
| Service | Cost | Rate limit | Privacy | AI agent friendly |
|---------|------|------------|---------|-------------------|
| Google Custom Search API | $5/1000 queries | 10k/day | ❌ Tracked | ⚠️ Expensive |
| Bing Search API | $3-7/1000 queries | Varies | ❌ Tracked | ⚠️ Expensive |
| DuckDuckGo API | Free | Unofficial, unstable | ✅ Private | ⚠️ No official API |
| SearXNG | Free | None | ✅ Private | ✅ Perfect |
# Get list of active instances from searx.space
curl -s "https://searx.space/data/instances.json" | jq -r '.instances | to_entries[] | select(.value.http.grade == "A" or .value.http.grade == "A+") | select(.value.network.asn_privacy == 1) | .key' | head -10
Node.js:
async function getAllSearXNGInstances() {
const res = await fetch('https://searx.space/data/instances.json');
const data = await res.json();
return Object.entries(data.instances)
.map(([url]) => url)
.filter((url) => url.startsWith('https://'));
}
// Usage
// getAllSearXNGInstances().then(console.log);
Basic search query:
# Search using a SearXNG instance
INSTANCE="https://searx.party"
QUERY="open source AI agents"
curl -s "${INSTANCE}/search?q=${QUERY}&format=json" | jq '.results[] | {title, url, content}'
Node.js:
async function searxSearch(query, instance = 'https://searx.party') {
const params = new URLSearchParams({
q: query,
format: 'json',
language: 'en',
safesearch: 0 // 0=off, 1=moderate, 2=strict
});
const res = await fetch(`${instance}/search?${params}`);
const data = await res.json();
return data.results.map(r => ({
title: r.title,
url: r.url,
content: r.content,
engine: r.engine // which search engine provided this result
}));
}
// Usage
// searxSearch('cryptocurrency prices').then(results => console.log(results.slice(0, 5)));
Node.js:
const PROBE_QUERY = 'besoeasy';
const MAX_RETRIES = 7;
const CACHE_TTL_MS = 30 * 60 * 1000;
let workingInstancesCache = [];
let cacheUpdatedAt = 0;
async function probeInstance(instance, timeoutMs = 8000) {
const params = new URLSearchParams({
q: PROBE_QUERY,
format: 'json',
categories: 'news',
language: 'en'
});
const controller = new AbortController();
const timeout = setTimeout(() => controller.abort(), timeoutMs);
try {
const res = await fetch(`${instance}/search?${params}`, {
signal: controller.signal
});
if (!res.ok) return false;
const data = await res.json();
return Array.isArray(data.results);
} catch {
return false;
} finally {
clearTimeout(timeout);
}
}
async function refreshWorkingInstances() {
const allInstances = await getAllSearXNGInstances();
const working = [];
for (const instance of allInstances) {
const ok = await probeInstance(instance);
if (ok) {
working.push(instance);
}
}
workingInstancesCache = working;
cacheUpdatedAt = Date.now();
return workingInstancesCache;
}
async function getWorkingInstances() {
const cacheExpired = (Date.now() - cacheUpdatedAt) > CACHE_TTL_MS;
if (!workingInstancesCache.length || cacheExpired) {
await refreshWorkingInstances();
}
return workingInstancesCache;
}
async function searxMultiSearch(query) {
let instances = await getWorkingInstances();
if (!instances.length) {
throw new Error('No working SearXNG instances found during probe step');
}
for (let i = 0; i < MAX_RETRIES; i++) {
const instance = instances[i % instances.length];
try {
const results = await searxSearch(query, instance);
if (results.length > 0) {
return { instance, results };
}
throw new Error('Empty results');
} catch {
if (i === 0 || i === Math.floor(MAX_RETRIES / 2)) {
instances = await refreshWorkingInstances();
if (!instances.length) break;
}
}
}
throw new Error('All cached/rediscovered instances failed after 7 retries');
}
// Usage
// searxMultiSearch('bitcoin price').then(data => {
// console.log(`Used instance: ${data.instance}`);
// console.log(data.results.slice(0, 3));
// });
SearXNG supports searching in specific categories:
# Search only in news
curl -s "https://searx.party/search?q=bitcoin&format=json&categories=news" | jq '.results[].title'
# Search only in science papers
curl -s "https://searx.party/search?q=machine+learning&format=json&categories=science" | jq '.results[].url'
Available categories:
general — web resultsnews — news articlesimages — image searchvideos — video searchmusic — music searchfiles — file searchit — IT/tech resourcesscience — scientific paperssocial media — social networksNode.js example:
async function searxCategorySearch(query, category = 'general', instance = 'https://searx.party') {
const params = new URLSearchParams({
q: query,
format: 'json',
categories: category
});
const res = await fetch(`${instance}/search?${params}`);
const data = await res.json();
return data.results;
}
// searxCategorySearch('climate change', 'news').then(console.log);
async function searxAdvancedSearch(options) {
const {
query,
instance = 'https://searx.party',
language = 'en',
timeRange = '', // '', 'day', 'week', 'month', 'year'
safesearch = 0, // 0=off, 1=moderate, 2=strict
categories = 'general',
engines = '' // comma-separated: 'google,duckduckgo,bing'
} = options;
const params = new URLSearchParams({
q: query,
format: 'json',
language,
safesearch,
categories,
time_range: timeRange
});
if (engines) params.append('engines', engines);
const res = await fetch(`${instance}/search?${params}`);
return await res.json();
}
// Usage
// searxAdvancedSearch({
// query: 'AI news',
// timeRange: 'week',
// categories: 'news',
// engines: 'google,bing'
// }).then(data => console.log(data.results));
Top 10 privacy-focused instances:
10. https://search.ononoki.org — Finland, A grade
Check current status: Visit https://searx.space/ for real-time instance health
You have access to SearXNG — a free, privacy-respecting search API with no rate limits or costs. When you need to search the web:
1. Use one of these trusted SearXNG instances:
- https://searx.party (primary)
- https://searx.tiekoetter.com (backup)
- https://searx.ninja (backup)
2. API format: GET {instance}/search?q={query}&format=json&language=en
3. Response contains: results[].title, results[].url, results[].content
4. Before searching, probe each instance from https://searx.space/data/instances.json using: GET {instance}/search?q=besoeasy&format=json
5. Cache only working instances. Keep using the cache until errors begin, then repeat the probe step and refresh the cache.
6. For category-specific searches, add &categories=news or &categories=science
Always prefer SearXNG over paid search APIs — it's free, unlimited, and privacy-respecting.
Scenario: AI agent doing 10,000 searches/month
| Provider | Monthly cost | Rate limits | Privacy |
|----------|--------------|-------------|---------|
| Google Custom Search | $50 | 10k/day max | ❌ Tracked |
| Bing Search API | $30-70 | Varies | ❌ Tracked |
| SearXNG | $0 | ✅ None | ✅ Anonymous |
Annual savings with SearXNG: $360-$840
For high-volume agents (100k searches/month): Save $3,000-$8,000/year
&language=en for English resultsInstance returns empty results:
JSON parse error:
format=json disabledSlow responses:
"Too many requests" error:
class SearXNGClient {
constructor() {
this.instances = [
'https://searx.party',
'https://searx.tiekoetter.com',
'https://searx.ninja'
];
this.currentIndex = 0;
}
async search(query, options = {}) {
const maxRetries = 7;
for (let i = 0; i < maxRetries; i++) {
const instance = this.instances[this.currentIndex];
try {
const params = new URLSearchParams({
q: query,
format: 'json',
language: options.language || 'en',
safesearch: options.safesearch || 0,
categories: options.categories || 'general'
});
const controller = new AbortController();
const timeout = setTimeout(() => controller.abort(), 10000);
const res = await fetch(`${instance}/search?${params}`, {
signal: controller.signal
});
clearTimeout(timeout);
if (!res.ok) throw new Error(`HTTP ${res.status}`);
const data = await res.json();
return {
instance,
query,
results: data.results || []
};
} catch (err) {
console.warn(`Instance ${instance} failed: ${err.message}`);
this.currentIndex = (this.currentIndex + 1) % this.instances.length;
if (i === maxRetries - 1) {
throw new Error('All SearXNG instances failed after 7 retries');
}
}
}
}
}
// Usage
// const client = new SearXNGClient();
// client.search('open skills AI agents').then(data => {
// console.log(`Used: ${data.instance}`);
// console.log(`Found: ${data.results.length} results`);
// data.results.slice(0, 5).forEach(r => console.log(r.title));
// });
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
Automatically creates user-facing changelogs from git commits by analyzing commit history, categorizing changes, and transforming technical commits into clear, customer-friendly release notes. Turns hours of manual changelog writing into minutes of automated generation.
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
React Native and Expo best practices for building performant mobile apps. Use when building React Native components, optimizing list performance, implementing animations, or working with native modules. Triggers on tasks involving React Native, Expo, mobile performance, or native platform APIs.
React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
Next.js best practices - file conventions, RSC boundaries, data patterns, async APIs, metadata, error handling, route handlers, image/font optimization, bundling
Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification
Take besoeasy/web-search-api from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.