Querit Search

Web search skill for OpenClaw powered by Querit.ai.

Search the web, get structured results, and extract page content as markdown — all from your OpenClaw agent.

Quick Start

1. Install the skill (pick one method below)
2. Set your API key: export QUERIT_API_KEY="your-key-here"
3. Use it: Ask OpenClaw to "search the web for..." and it handles the rest

Installation

One-line installer

curl -fsSL https://raw.githubusercontent.com/interskh/querit-search/main/install.sh | bash

Manual (git clone)

git clone https://github.com/interskh/querit-search.git ~/.openclaw/skills/querit-search
cd ~/.openclaw/skills/querit-search && npm ci

Manual (copy)

Copy the skill files to ~/.openclaw/skills/querit-search/ and run npm ci in that directory.

Configuration

Get a free API key at querit.ai (1,000 queries/month, no credit card required).

Option A: Environment variable

export QUERIT_API_KEY="your-key-here"

Add to your ~/.bashrc, ~/.zshrc, or ~/.profile to persist.

Option B: OpenClaw config

In ~/.openclaw/openclaw.json:

{
  "skills": {
    "entries": {
      "querit-search": {
        "apiKey": "your-key-here"
      }
    }
  }
}

Option C: .env file

Create ~/.openclaw/.env:

QUERIT_API_KEY=your-key-here

Usage Examples

Just ask OpenClaw naturally:

• "Search for React server components best practices"
• "Find the latest Node.js release notes"
• "Look up how to use CSS container queries"
• "Search for TypeScript 5.4 new features and read the first result"

OpenClaw will use the skill automatically when it needs to search the web.

CLI Reference

search.js

search.js <query> [options]

Options:
  -n <count>                Number of results (default: 5, max: 100)
  --lang <language>         Language filter
  --country <country>       Country filter
  --date <range>            Date filter: d1 (day), w1 (week), m1 (month), y1 (year)
  --site-include <domain>   Only include results from this domain (repeatable)
  --site-exclude <domain>   Exclude results from this domain (repeatable)
  --content                 Also fetch and extract page content as markdown
  --json                    Output raw JSON

Supported languages: english, japanese, korean, german, french, spanish, portuguese

Supported countries: argentina, australia, brazil, canada, colombia, france, germany, india, indonesia, japan, mexico, nigeria, philippines, south korea, spain, united kingdom, united states

content.js

content.js <url>

Fetches a URL and extracts the main readable content as markdown using Readability + Turndown.

API Limits

Plan	Price	Queries	Rate limit
Free	$0	1,000/month	1 QPS
Pro	$4/1,000 requests	Unlimited	10 QPS
Enterprise	$6/1,000 requests	Unlimited	Unlimited

Additional constraints:

• Query: max 72 characters (auto-truncated with warning)
• Results: max 100 per query
• Site filters: max 20 domains each

Troubleshooting

"QUERIT_API_KEY environment variable is not set"

Make sure your API key is exported in the shell where OpenClaw runs. If using OpenClaw config, check the path is skills.entries.querit-search.apiKey.

"npm ci failed" during install

Try npm install instead. Make sure you have Node.js 18+ (node -v).

"API returned error 401"

Your API key is invalid or expired. Generate a new one at querit.ai.

"API returned error 429"

Rate limit exceeded. Free tier allows 1 query per second and 1,000 queries per month.

"Could not extract readable content from page"

Some pages (SPAs, paywalled content, PDFs) can't be extracted by Readability. This is expected — the search results themselves are still useful.

Contributing

1. Clone the repo
2. Make changes
3. Test locally:

   ./search.js "test query"
   ./search.js "test" --json
   ./content.js https://example.com

4. Submit a PR

License

MIT