Tavily Proxy Search for Openclaw

A powerful search bridge that enables AI agents to perform authenticated web research through a self-hosted TavilyProxyManager instance.

ugvfpdcuwfnh
v1.0.0
Apr 9, 2026
1
784
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install tavily-proxy-search

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install tavily-proxy-search using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Tavily Proxy Search?

Tavily Proxy Search is a specialized tool designed for developers who need the high-quality research capabilities of Tavily-style searching within a self-hosted or proxied environment. As part of the Openclaw Skills ecosystem, this skill provides a seamless interface to query web results, gather sources, and retrieve AI-ready summaries without connecting directly to the official Tavily API endpoints.

By utilizing a Bearer master-key authentication system, it ensures secure and private communication between your AI coding agent and your search infrastructure. It is optimized for speed and data efficiency, allowing for granular control over search depth and result formatting to perfectly suit various automated research workflows.

Tavily Proxy Search Use Cases

  • Performing real-time web research through a private proxy to maintain data sovereignty.
  • Gathering high-quality sources and snippets for AI context windows during complex coding tasks.
  • Generating quick summaries or short answers for user queries using the Openclaw Skills search framework.
  • Integrating a stable web search schema into automated workflows that require consistent JSON or Markdown outputs.

How Tavily Proxy Search Works

  1. The user or agent triggers a search request using the provided Python script within the Openclaw Skills environment.
  2. The script validates the required TAVILY_PROXY_MASTER_KEY and checks for the configured proxy URL.
  3. A secure HTTP request is dispatched to the TavilyProxyManager /search endpoint with Bearer token authentication.
  4. The proxy processes the request, communicating with upstream search providers as configured.
  5. The skill receives the raw data and transforms it into the requested format (JSON, Brave-compatible, or Markdown).
  6. The final cleaned search results are returned to the agent for further processing or display.

Tavily Proxy Search Setup

To get started with this component of Openclaw Skills, ensure your TavilyProxyManager is running and then configure your credentials:

# Set your proxy credentials in your environment
export TAVILY_PROXY_URL="http://127.0.0.1:8080/search"
export TAVILY_PROXY_MASTER_KEY="your-master-key"

Alternatively, you can add these values to your ~/.openclaw/.env file to ensure persistence across sessions.

Tavily Proxy Search Data Schema & Taxonomy

The skill supports multiple output schemas to ensure compatibility with different agent requirements within the Openclaw Skills ecosystem:

Format Description Key Fields
raw The default Tavily-style JSON output. query, results, answer
brave A stabilized schema optimized for web search tools. title, url, snippet
md A compact, human-readable Markdown list. Title, URL, and Snippet list

All JSON responses maintain strict structure for reliable automated parsing.

Tavily Proxy Search Advanced Features

  • Support for variable search depth (basic or advanced) depending on your proxy configuration.
  • Flexible output formatting including a Brave-compatible mode for easy integration with existing search-dependent tools.
  • Master-key Bearer authentication for enterprise-grade security and access control.
  • Direct integration with Openclaw Skills workspace for seamless multi-agent research operations.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*