A powerful CLI tool that provides AI agents with live web access and deep reasoning capabilities via OpenRouter.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-search-tool
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 openclaw-search-tool 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 OpenClaw Research Tool?
The OpenClaw Research Tool is a specialized utility designed to bridge the gap between static LLM training data and the live web. By leveraging OpenRouter and the :online model suffix, it enables agents to browse the internet, retrieve real-time data, and synthesize complex answers with accurate citations. This tool is a fundamental addition to any suite of Openclaw Skills, ensuring that agents can handle documentation lookups, market research, and fact-checking with ease.
It is built to handle the high-latency nature of deep reasoning tasks, providing detailed status updates and progress tracking. Whether used as a standalone CLI or integrated into an autonomous agent workflow, it provides the necessary context and reasoning depth required for professional-grade technical research.
OpenClaw Research Tool Use Cases
- Retrieving live API documentation and rate limits for software development.
- Conducting deep market research and competitive analysis with cited sources.
- Fact-checking news or current events that occurred after a model's training cutoff.
- Performing multi-source synthesis for complex technical architecture comparisons.
- Offloading long-running research tasks to sub-agents to keep primary sessions responsive.
How OpenClaw Research Tool Works
- The tool receives a natural language query via CLI arguments or stdin.
- It establishes a connection to OpenRouter, utilizing models like GPT-5.2 with the :online suffix to enable web browsing.
- The model performs multiple web searches, reads relevant pages, and extracts key information.
- Depending on the selected effort level, the model conducts internal reasoning to resolve contradictions and synthesize a final answer.
- The final output is delivered as Markdown-formatted text with citations on stdout, while telemetry and logs are sent to stderr.
OpenClaw Research Tool Setup
To integrate this into your Openclaw Skills, follow these steps:
# Install the research tool via Cargo
cargo install openclaw-search-tool
# Set your OpenRouter API key
export OPENROUTER_API_KEY='your_api_key_here'
# Run a test query
research-tool "Latest features in the newest Rust release"
OpenClaw Research Tool Data Schema & Taxonomy
The tool outputs data in a stream-friendly format to allow for easy integration with other Openclaw Skills.
| Output Stream | Data Type | Description |
|---|---|---|
| stdout | Markdown | The final synthesized answer including URL citations. |
| stderr | Text/Metrics | Progress indicators, reasoning traces, and token usage statistics. |
| Environment | Configuration | Uses OPENROUTER_API_KEY, RESEARCH_MODEL, and RESEARCH_EFFORT for behavior control. |
OpenClaw Research Tool Advanced Features
- Support for four reasoning effort levels (low, medium, high, xhigh) to balance speed and depth.
- Model flexibility allowing any OpenRouter model to be used with live web access via the :online suffix.
- Sub-agent optimization guidelines to prevent session timeouts during long-running (10+ minute) research tasks.
- Custom system prompts to force the research agent into specific professional personas or domain expertise.
- Pipe-friendly architecture with --stdin support for automated workflows and script integration.
SKILL.md
Loading
Related Openclaw Skills
Huckleberry Baby Tracker
A comprehensive AI agent skill that enables natural language tracking of baby sleep, feeding, diapers, and growth via the Huckleberry app API.
Apple Media Control
A comprehensive command-line integration for managing Apple TV, HomePod, and AirPlay devices using the pyatv library.
OKX DEX Swap & Broadcast
A comprehensive tool for executing end-to-end on-chain token swaps with MEV protection and automated slippage via the OKX DEX Aggregator.
OKX DEX Aggregator Quote
A specialized skill for fetching real-time, optimized cryptocurrency swap quotes across multiple blockchain networks using the OKX DEX Aggregator API.
Browser Auto Download
An intelligent browser-based automation tool that handles dynamic web content and multi-step download navigation where standard CLI tools fail.
Agent Orchestrator
A meta-agent skill that orchestrates complex macro tasks by decomposing them into subtasks performed by autonomous, specialized sub-agents.
METADATA
Featured*
