AIsa Tavily Search

该 Skill 借助 AIsa 的检索接口执行网页、多源和近 30 天范围的搜索研究，面向检索与总结、竞品扫描、趋势发现等场景，可把多源检索整理为研究简报或结构化输出。运行依赖 node 脚本与 AISA_API_KEY，脚本为 scripts/search.mjs 与 scripts/extract.mjs；当部分数据源超时时要求如实说明。

bibaofeng

热度 47.07

最近更新 2026-05-11 17:45:14

分类：搜索依赖：代码执行 / 第三方 API应用场景：研究 / 摘要总结 / 市场分析

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add aisa-tavily-search

Metadata

name: aisa-tavily-search
description: Run web, multi-source, or last-30-days research through AIsa. Use when: the user needs search, synthesis, competitor scans, or trend discovery. Supports research-ready outputs and structured retrieval.
author: AIsa
version: 1.0.3
license: Apache-2.0
homepage: https://aisa.one
source: https://github.com/baofeng-tech/agent-skills-io/tree/main/targetSkills/aisa-tavily
user-invocable: true
primaryEnv: AISA_API_KEY
requires: {"bins":["node"],"env":["AISA_API_KEY"]}
metadata: {"aisa":{"emoji":"🔎","requires":{"bins":["node"],"env":["AISA_API_KEY"]},"primaryEnv":"AISA_API_KEY","compatibility":["openclaw","claude-code","hermes"]},"openclaw":{"emoji":"🔎","requires":{"bins":["node"],"env":["AISA_API_KEY"]},"primaryEnv":"AISA_API_KEY"}}

AIsa Tavily Search

Run web, multi-source, or last-30-days research through AIsa. Use when: the user needs search, synthesis, competitor scans, or trend discovery. Supports research-ready outputs and structured retrieval.

When to use

The user needs web, multi-source, or last-30-days research.
The user wants competitor scans, trend discovery, or structured search output.
The user wants one skill to cover multiple retrieval surfaces.

High-Intent Workflows

Search and summarize recent evidence.
Compare two tools or companies using recent signals.
Turn multi-source retrieval into a research brief.

Setup

AISA_API_KEY is required for AIsa-backed API access.
Use repo-relative scripts/ paths from the shipped package.

Example Requests

Research OpenAI Agents SDK over the last 30 days
Compare OpenClaw and Codex using recent public discussion
Search recent sentiment around a product launch

Guardrails

Do not present test-only helpers as public features.
Do not claim sources that were not actually queried.
If some providers time out, report that honestly.