← Back to Skills Marketplace
426
Downloads
1
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install liang-tavily-extract
Description
Extract content from specific URLs using Tavily's extraction API. Returns clean markdown/text from web pages.
README (SKILL.md)
Tavily Extract
Extract clean content from specific URLs. Ideal when you know which pages you want content from.
Authentication
Get your API key at https://tavily.com and add to your OpenClaw config:
{
"skills": {
"entries": {
"tavily-extract": {
"enabled": true,
"apiKey": "tvly-YOUR_API_KEY_HERE"
}
}
}
}
Or set in environment variable:
export TAVILY_API_KEY="tvly-YOUR_API_KEY_HERE"
Quick Start
Using the Script
node {baseDir}/scripts/extract.mjs "https://example.com/article"
node {baseDir}/scripts/extract.mjs "url1,url2,url3"
node {baseDir}/scripts/extract.mjs "url" --query "authentication API"
Examples
# Single URL
node {baseDir}/scripts/extract.mjs "https://docs.python.org/3/tutorial/classes.html"
# Multiple URLs
node {baseDir}/scripts/extract.mjs "https://example.com/page1,https://example.com/page2"
# With query focus
node {baseDir}/scripts/extract.mjs "https://example.com/docs" --query "authentication API"
# Advanced extraction for JS pages
node {baseDir}/scripts/extract.mjs "https://app.example.com" --depth advanced --timeout 60
Options
| Option | Description | Default |
|---|---|---|
--query \x3Ctext> |
Rerank chunks by relevance | - |
--chunks \x3Cn> |
Chunks per URL (1-5, requires query) | 3 |
--depth \x3Cmode> |
Extract depth: basic or advanced |
basic |
--format \x3Cfmt> |
Output format: markdown or text |
markdown |
--timeout \x3Csec> |
Max wait time (1-60 seconds) | varies |
--json |
Output raw JSON | false |
Extract Depth
| Depth | When to Use |
|---|---|
basic |
Simple text extraction, faster |
advanced |
Dynamic/JS-rendered pages, tables, structured data |
Tips
- Max 20 URLs per request - batch larger lists
- Use
--query+--chunksto get only relevant content - Try
basicfirst, fall back toadvancedif content is missing - Set longer
--timeoutfor slow pages (up to 60s) - Check
failed_resultsin JSON output for URLs that couldn't be extracted
Usage Guidance
This skill appears coherent, but consider the following before installing: (1) The script sends URLs (and the service will fetch page content) to Tavily's API — do not send private/internal URLs you don't want a third party to access. (2) Protect the TAVILY_API_KEY like any secret and avoid reusing it across unrelated services. (3) Ensure your Node runtime supports global fetch (Node 18+ or supply a fetch polyfill). (4) Check Tavily's privacy, retention, and billing policies if you will submit many or sensitive URLs.
Capability Analysis
Type: OpenClaw Skill
Name: liang-tavily-extract
Version: 1.0.0
The OpenClaw skill bundle for 'tavily-extract' is benign. The `SKILL.md` file provides clear, functional instructions without any prompt injection attempts or hidden agendas. The `scripts/extract.mjs` script correctly implements the described functionality, making a `fetch` request to the legitimate Tavily API endpoint (https://api.tavily.com/extract) using an API key from the environment variable `TAVILY_API_KEY`. There is no evidence of data exfiltration, malicious command execution, persistence mechanisms, or obfuscation. All actions are directly aligned with the stated purpose of extracting web content.
Capability Assessment
Purpose & Capability
Name/description, required binary (node), required env var (TAVILY_API_KEY), and the shipped script all consistently implement a client for Tavily's extraction API. There are no unrelated credentials, binaries, or config paths requested.
Instruction Scope
SKILL.md only instructs how to supply the API key and run the provided script. The script accepts URLs and options and sends them to https://api.tavily.com/extract. It does not read other local files, environment variables, or system paths beyond TAVILY_API_KEY.
Install Mechanism
This is an instruction-only skill with a single node script (no install spec). That's low-risk. One note: the script uses the global fetch API, so it requires a Node runtime that provides fetch (Node 18+ or equivalent polyfill).
Credentials
Only TAVILY_API_KEY is required and is the primary credential used for Authorization in requests to the Tavily API. No additional secrets or unrelated env vars are requested.
Persistence & Privilege
always is false and the skill does not request any persistent system-level privileges or modify other skills. It can be invoked autonomously (platform default), which is expected for a skill of this type.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install liang-tavily-extract - After installation, invoke the skill by name or use
/liang-tavily-extract - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of tavily-extract skill.
- Extracts clean markdown or text from URLs using Tavily's extraction API.
- Supports single or multiple URLs (up to 20 per request).
- Optional query-based content reranking and chunk limiting.
- Basic and advanced extraction modes for static or JS-rendered pages.
- Multiple output formats: markdown, text, or raw JSON.
- Requires TAVILY_API_KEY for authentication.
Metadata
Frequently Asked Questions
What is Tavily Extract?
Extract content from specific URLs using Tavily's extraction API. Returns clean markdown/text from web pages. It is an AI Agent Skill for Claude Code / OpenClaw, with 426 downloads so far.
How do I install Tavily Extract?
Run "/install liang-tavily-extract" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Tavily Extract free?
Yes, Tavily Extract is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Tavily Extract support?
Tavily Extract is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Tavily Extract?
It is built and maintained by Liang (@matthew77); the current version is v1.0.0.
More Skills