aisa-multi-source-search
от aisapay
Search & ResearchКратко
Intelligent search for agents.
Описание
Что делает этот навык
Queries web, academic, and Tavily sources in parallel and synthesizes results into a confidence-scored answer. Uses a two-phase approach: parallel retrieval followed by meta-analysis via AIsa Explain. Each response includes a score from 0 to 100 based on source quality, cross-source agreement, recency, and relevance.
Combining web, academic, and external validation sources in one API call with deterministic confidence scoring removes the need to manually reconcile results from separate search tools.
Когда его использовать
- Finding recent academic papers on a topic with year range filters
- Aggregating web coverage of an industry event or funding round
- Comparing reviews and benchmarks across competing frameworks
- Verifying a technical claim across multiple source types before citing it
- Crawling and extracting content from a list of competitor URLs
Пример сценария
User asks: Is quantum computing ready for enterprise use?
- 1
Sends the query to scholar, web, smart, and Tavily endpoints in parallel
- 2
Collects structured results from all four streams
- 3
Passes combined results to AIsa Explain for meta-analysis
- 4
Calculates a confidence score based on source quality, agreement, recency, and relevance
- 5
Returns synthesized answer with citations and the numeric confidence score
A confidence-scored summary (e.g. 72/100) with sourced claims and a synthesis of current consensus across academic and web sources
Требования
- AISA_API_KEY — sign up at aisa.one to obtain
- Credits loaded on the aisa.one account (pay-as-you-go pricing)
Установка
clawhub install aisapay/aisa-multi-source-searchВыполните команду CLI или вставьте промпт в Claude Code, Cursor, Codex — любой кодинг-агент, которым пользуетесь.