A deterministic research orchestration skill that chains DeepSeek, Gemini, and Kimi agents to transform raw queries into structured, high-fidelity briefs.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install search-analyst-pair
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 search-analyst-pair 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 Search Analyst Pair?
Search Analyst Pair is a specialized extension for Openclaw Skills designed to eliminate the unpredictability of ad-hoc AI routing. By implementing a strict 3-hop chain, it ensures that research tasks are handled with a clear separation of concerns. This skill is ideal for developers and researchers who require a verifiable trail of information, moving from raw data collection to deep analytical processing before reaching a final recommendation.
By leveraging the unique strengths of different LLMs—DeepSeek for exhaustive searching, Gemini for nuanced analysis, and Kimi for cohesive synthesis—the Search Analyst Pair provides a professional-grade intelligence pipeline. It ensures that every response under the Openclaw Skills ecosystem is backed by a structured methodology rather than a single-pass generation.
Search Analyst Pair Use Cases
- Critical Decision Support: Use the skill when source traceability and risk assessment are mandatory for business or technical decisions.
- Market Research: Automate the collection of industry updates followed by a SWOT analysis and strategic summary.
- Technical Migration Planning: Process complex framework updates by gathering facts, identifying breaking changes, and generating a migration roadmap.
- Legal & Compliance Monitoring: Track weekly regulatory changes and assess implementation risks for existing workflows.
How Search Analyst Pair Works
- Search Phase (DeepSeek): Triggered by the
/huntcommand, the first agent gathers verified facts and primary sources from the web. - Analysis Phase (Gemini): The search results are passed to the Analyst agent, which identifies key points, potential risks, and data gaps.
- Synthesis Phase (Kimi): The Main agent combines the search findings and the analytical assessment into a polished, actionable final report.
- Structured Delivery: The workflow outputs three distinct sections to the user: Search findings, Analyst assessment, and Main conclusion.
Search Analyst Pair Setup
To deploy this skill within your Openclaw environment, ensure your gateway is reachable and the following dependencies are met:
# Ensure environment variables are set
export OPENCLAW_BASE_URL="your_gateway_url"
export OPENCLAW_TOKEN="your_api_token"
# Verify required binaries are installed
python3 --version
curl --version
jq --version
Agent Configuration:
You must define three specific agent IDs in your config: main, search, and analyst. Ensure tools.agentToAgent.enabled=true is set to allow the 3-hop handoff.
Search Analyst Pair Data Schema & Taxonomy
The Search Analyst Pair skill organizes data into a deterministic hierarchy to ensure reviewability. The output follows this schema:
| Section | Source Agent | Content Type |
|---|---|---|
| Search findings | DeepSeek | Raw facts, URLs, and verified data points. |
| Analyst assessment | Gemini | Identification of trends, risks, and critical insights. |
| Main conclusion | Kimi | Executive summary and actionable recommendations. |
All workflows are triggered by the /hunt prefix, ensuring that these Openclaw Skills requests are routed through the multi-agent pipeline rather than the standard chat interface.
Search Analyst Pair Advanced Features
- Deterministic Orchestration: Prevents model drifting by forcing a fixed
Search -> Analyst -> Mainexecution order. - Fallback Execution Policy: Includes a FastAPI fallback runner that takes over if the primary agent-to-agent spawn fails, ensuring no research task is dropped.
- Multi-Model Synergy: Optimizes costs and performance by assigning specific sub-tasks to the LLM best suited for the job.
- Cross-Agent Permissions: Utilizes granular
subagents.allowAgentsconfigurations to maintain secure communication between the main, search, and analyst agents.
SKILL.md
Loading
Related Openclaw Skills
Yuanyuan Blueprint Workshop
A structured 5-step workshop designed to extract tacit human know-how and transform it into a buildable, testable AI Agent Blueprint.
MiniMax AI Hot Trend
A specialized data extraction tool for gathering the latest public model information, pricing updates, and documentation from the MiniMax AI platform.
Compound
A professional CLI-based financial engine for calculating compound interest, investment growth, and savings goals.
Internet Speed Test
A lightweight Python-based tool to measure current download and upload throughput using Cloudflare speed endpoints.
Analyze Gas Optimization
A specialized tool for analyzing and optimizing Aptos Move smart contracts to minimize gas consumption and transaction costs.
MediaIO Nano Banana Pro Image Generator
A high-performance AI image generation skill leveraging Nano Banana Pro for state-of-the-art visual quality and multimodal reasoning.
Featured*
