A high-performance search skill for AI agents featuring concurrent web retrieval and specialized code documentation search.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install performing-searches
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install performing-searches 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 Augmented Search?
Augmented Search is a sophisticated toolset designed to provide AI agents with high-efficiency local and web search capabilities. By integrating SearXNG and specialized code search providers, this skill allows agents to perform hybrid retrieval that significantly improves search relevance, reaching up to 80% accuracy through embedding-based re-ranking. It is a critical component for developers building Openclaw Skills that require reliable, real-time access to the web or technical documentation.
The skill excels at handling complex search workflows, allowing for simultaneous multi-keyword queries and deep content extraction. It handles the nuances of modern web pages, including JavaScript rendering and paginated reading, ensuring that agents receive the most comprehensive and clean data possible for their reasoning tasks.
Augmented Search Use Cases
- Performing concurrent searches across multiple keywords to gather diverse perspectives on a single topic.
- Retrieving specific programming library documentation and code samples for React, Vue, or other frameworks.
- Extracting clean text and headings from complex, JavaScript-heavy websites.
- Implementing a private, local search infrastructure for agents using SearXNG instances.
How Augmented Search Works
- The AI agent initiates a search request containing specific keywords and a thought process log.
- The skill executes concurrent queries (up to 3 keywords) via a SearXNG instance or code library index.
- Search results are processed through hybrid retrieval and optional embedding re-ranking to prioritize the most relevant information.
- The agent can then use the read tool to fetch full content, extract specific sections, or paginate through long documents.
- Specialized library tools allow for targeted identification of programming packages and direct retrieval of their technical manuals.
Augmented Search Setup
To deploy this search capability within Openclaw Skills, you first need a SearXNG instance. The recommended way to install is via Docker:
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
Alternatively, you can install via npm:
npm install -g augmented-search
SEARXNG_URL=http://localhost:8080 augmented-search
Augmented Search Data Schema & Taxonomy
The skill organizes data into four primary functional areas to support agent operations:
| Tool | Primary Data Output |
|---|---|
| search | Concurrent search results including snippets, URLs, and relevance rankings. |
| read | Markdown-formatted web content with support for pagination and section targeting. |
| library_search | Validated library IDs and metadata for technical packages. |
| library_docs | Structured documentation segments and code examples. |
Configuration is managed via environment variables such as SEARXNG_URL and optional embedding API keys for re-ranking logic.
Augmented Search Advanced Features
- Hybrid retrieval re-ranking that boosts search accuracy to approximately 80% using embedding models.
- Multi-keyword concurrency allowing up to three simultaneous searches per request.
- Intelligent content extraction with JavaScript rendering fallback and heading-only modes.
- Context7 integration for high-fidelity programming library and documentation lookup.
- Support for both MCP protocol and standard REST API endpoints for maximum flexibility.
SKILL.md
Loading
Related Openclaw Skills
Blank Files Gateway
A high-performance gateway to blankfiles.com providing direct download URLs for binary test files across numerous formats and categories.
Blank Files Gateway
A binary test-file gateway that provides direct download URLs for blank files across hundreds of formats via the blankfiles.com API.
Skill Reviews
A decentralized feedback registry that allows AI agents to publish versioned reviews and read community performance data for various tools.
Skill Review Registry
A public, programmatic registry that enables AI agents to publish versioned reviews and access community feedback for Openclaw Skills.
Performing Searches (Augmented Search)
Equips AI agents with concurrent web search, hybrid retrieval, and deep programming documentation access.
Openclaw Project Management Governance
A structured governance framework that enables Openclaw agents to manage complex projects through subagent orchestration and rigorous logging.
Featured*
