Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Intelligent search with multi-source retrieval and confidence scoring. Orchestrates web, academic, and Tavily searches with two-phase strategy (discovery + reasoning). Provides deterministic confidence scores (0-100) and cross-validation.
What This Skill Does
Queries web, academic, and Tavily sources in parallel, then runs meta-analysis to produce a confidence-scored answer. A single API key covers all search modes. Confidence scores (0-100) are weighted by source quality, cross-source agreement, recency, and relevance.
Combining academic, web, and Tavily retrieval in one call with deterministic confidence scoring removes the need to integrate, query, and deduplicate multiple search APIs separately.
When to use it
- Finding academic papers on a specific topic filtered by publication year
- Aggregating recent news about an industry or technology trend
- Comparing reviews and discussions of competing software frameworks
- Checking whether a factual claim has consensus across academic and web sources
- Extracting and synthesizing content from a set of known URLs
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Are transformer-based architectures still dominant for LLMs in 2025?
- 1Sends query to scholar, web, smart, and Tavily streams simultaneously
- 2Collects and merges results from all four sources
- 3Passes combined results to AIsa Explain for meta-analysis
- 4Calculates a confidence score weighted by source quality, agreement, recency, and relevance
- 5Returns a synthesized answer with citations and a 0-100 confidence score
A confidence-scored summary citing academic papers and web sources, with score indicating the strength of cross-source consensus
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.