Performing Searches (Augmented Search) for Openclaw

Equips AI agents with concurrent web search, hybrid retrieval, and deep programming documentation access.

sebrinass
v1.1.6
Mar 5, 2026
0
1.2k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install skill-7

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 skill-7 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 Performing Searches (Augmented Search)?

Performing Searches is a high-performance search augmentation skill designed for AI agents. It leverages SearXNG and hybrid retrieval techniques to deliver exceptionally relevant results, boosting search precision to approximately 80% through embedding-based re-ranking. By integrating this capability from the Openclaw Skills ecosystem, developers can enable their agents to perform multi-keyword concurrent searches and extract clean content from complex web pages.

Beyond standard web queries, it provides specialized hooks for code search and library documentation via Context7 integration. This makes it an essential tool for agents that need to stay updated with real-time web data or assist in complex software development tasks while maintaining a high degree of accuracy and performance.

Performing Searches (Augmented Search) Use Cases

  • Conducting concurrent multi-keyword web research to gather diverse perspectives quickly.
  • Searching for specific programming library documentation and code examples during development.
  • Extracting structured text content from URLs with Javascript rendering support.
  • Improving the reliability of agent responses using hybrid retrieval and embedding re-ranking.
  • Implementing local or self-hosted search capabilities to maintain data privacy within Openclaw Skills workflows.

How Performing Searches (Augmented Search) Works

  1. The agent identifies a need for external information and formulates a thought process including up to three concurrent search keywords.
  2. The skill queries a SearXNG instance to retrieve raw search results from multiple engines simultaneously.
  3. If an embedding API is configured, the system performs a hybrid retrieval process to re-rank results for maximum relevance.
  4. The agent can then use the read tool to fetch and parse specific URL content, handling complex JS-rendered pages.
  5. For technical queries, the skill optionally routes requests to Context7 to find specific programming library IDs and documentation.

Performing Searches (Augmented Search) Setup

Prerequisites

A running SearXNG instance is required.

docker run -d --name searxng -p 8080:8080 searxng/searxng:latest
docker run -d --name augmented-search -p 3000:3000 \
  -e SEARXNG_URL=http://host.docker.internal:8080 \
  ghcr.io/sebrinass/mcp-augmented-search:latest

npm Setup

npm install -g augmented-search
SEARXNG_URL=http://localhost:8080 augmented-search

Performing Searches (Augmented Search) Data Schema & Taxonomy

Component Data Type Description
Search Results Array List of re-ranked web links, snippets, and source engine metadata.
Page Content String Cleaned Markdown or text extracted from target URLs with JS rendering fallback.
Library Metadata Object Context7 library identifiers, versions, and documentation paths.
Embedding Config JSON API endpoints and model specifications for nomic-embed-text or OpenAI compatible models.

Performing Searches (Augmented Search) Advanced Features

  • Concurrent keyword processing (up to 3 keywords per request) to significantly reduce search latency.
  • Hybrid retrieval mode increasing search relevance from 50% to 80% using local or cloud-based embeddings.
  • Intelligent content extraction with pagination, section-specific reading, and paragraph range selection.
  • Built-in support for Basic Auth and HTTP/HTTPS proxies for secure and versatile enterprise environments.
  • Full integration with the Model Context Protocol (MCP) for seamless use within the wider Openclaw Skills framework.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*