← 返回 Skills 市场
106
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install aisa-multi-search-engine
功能描述
Multi-source search engine powered by AIsa API. Combines Tavily web search, Scholar academic search, Smart hybrid search, and Perplexity deep research — all...
使用说明 (SKILL.md)
AIsa Multi Search Engine v1.0.0
Multi-source search engine powered by AIsa API. One API key unlocks web search, academic search, Tavily integration, and Perplexity deep research with confidence-scored results.
Quick Start
export AISA_API_KEY="your-key-here"
Get your API key at aisa.one.
Search Tools
This plugin registers seven agent tools, each backed by a different AIsa API endpoint.
| Tool | Description | AIsa Endpoint |
|---|---|---|
aisa_web_search |
Structured web search results | /scholar/search/web |
aisa_scholar_search |
Academic paper search with year filtering | /scholar/search/scholar |
aisa_smart_search |
Intelligent hybrid search (web + academic) | /scholar/search/smart |
aisa_tavily_search |
Advanced web search with depth, topic, time, domain filters | /tavily/search |
aisa_tavily_extract |
Extract clean content from URLs | /tavily/extract |
aisa_perplexity_search |
Deep research via Perplexity Sonar models | /sonar, /sonar-pro, etc. |
aisa_multi_search |
Parallel multi-source search with confidence scoring | Multiple endpoints |
Usage Examples
Web Search
Search the web for structured results.
Use aisa_web_search to find "latest developments in AI agents 2025"
Academic Search
Search scholarly papers with optional year filtering.
Use aisa_scholar_search to find papers on "transformer architecture" from 2024 to 2025
Smart Search
Intelligent hybrid search combining web and academic sources.
Use aisa_smart_search to research "autonomous AI agents"
Tavily Search (Advanced)
Advanced web search with filtering options.
Use aisa_tavily_search for "AI startup funding" with topic "news" and time_range "month"
Tavily Extract
Extract full content from specific URLs.
Use aisa_tavily_extract to get content from ["https://example.com/article"]
Perplexity Deep Research
Use Perplexity Sonar models for synthesized answers with citations.
Use aisa_perplexity_search to answer "What is the current state of quantum computing?" with model "sonar-pro"
| Model | Best For |
|---|---|
sonar |
Quick answers, fast response |
sonar-pro |
Detailed answers with more citations |
sonar-reasoning-pro |
Complex reasoning and analysis |
sonar-deep-research |
Exhaustive multi-step research |
Multi-Source Search (Verity-Style)
Parallel search across all sources with confidence scoring and AI synthesis.
Use aisa_multi_search to comprehensively research "Is quantum computing ready for enterprise use?"
Confidence Scoring
The aisa_multi_search tool provides a deterministic confidence score based on four factors.
| Factor | Weight | Description |
|---|---|---|
| Source Availability | 40% | How many search sources returned results |
| Result Quality | 35% | Volume of results relative to expectations |
| Source Diversity | 15% | Mix of academic and web sources |
| Recency | 10% | Bonus for having any successful sources |
| Score | Level | Interpretation |
|---|---|---|
| 90-100 | Very High | Strong consensus across sources |
| 70-89 | High | Good agreement, reliable sources |
| 50-69 | Medium | Mixed signals, verify independently |
| 30-49 | Low | Conflicting or sparse sources |
| 0-29 | Very Low | Insufficient data |
Python CLI Client
A standalone Python client is included for command-line use.
# Web search
python3 scripts/search_client.py web --query "latest AI news" --count 10
# Academic search
python3 scripts/search_client.py scholar --query "transformer architecture" --year-from 2024
# Smart search
python3 scripts/search_client.py smart --query "autonomous agents" --count 10
# Tavily search
python3 scripts/search_client.py tavily --query "AI developments" --depth advanced
# Extract content from URLs
python3 scripts/search_client.py extract --urls "https://example.com/article"
# Perplexity deep research
python3 scripts/search_client.py sonar --query "quantum computing" --model sonar-pro
# Multi-source search with confidence scoring
python3 scripts/search_client.py verity --query "Is quantum computing ready?"
API Endpoints Reference
All endpoints use the AIsa API base URL https://api.aisa.one/apis/v1.
| Endpoint | Method | Description |
|---|---|---|
/scholar/search/web |
POST | Web search with structured results |
/scholar/search/scholar |
POST | Academic paper search |
/scholar/search/smart |
POST | Intelligent hybrid search |
/scholar/explain |
POST | Generate result explanations with confidence |
/tavily/search |
POST | Tavily web search |
/tavily/extract |
POST | Extract content from URLs |
/tavily/crawl |
POST | Crawl web pages |
/tavily/map |
POST | Generate site maps |
/sonar |
POST | Perplexity Sonar (fast) |
/sonar-pro |
POST | Perplexity Sonar Pro (detailed) |
/sonar-reasoning-pro |
POST | Perplexity Sonar Reasoning Pro |
/sonar-deep-research |
POST | Perplexity Sonar Deep Research |
Configuration
The plugin accepts the following configuration options.
| Option | Type | Default | Description |
|---|---|---|---|
aisaApiKey |
string | — | AIsa API key (or set AISA_API_KEY env var) |
defaultMaxResults |
integer | 10 | Default max results per search (1-100) |
defaultSearchDepth |
string | basic |
Default Tavily search depth |
Architecture
┌─────────────────────────────────────────────────────────────┐
│ User Query │
└─────────────────────────────────────────────────────────────┘
│
┌───────────────┼───────────────┐
▼ ▼ ▼
┌─────────┐ ┌─────────┐ ┌─────────┐
│ Scholar │ │ Web │ │ Smart │
└─────────┘ └─────────┘ └─────────┘
│ │ │
└───────────────┼───────────────┘
▼
┌─────────────────┐
│ Tavily Search │
└─────────────────┘
│
▼
┌─────────────────┐
│ AIsa Explain │
│ (Meta-Analysis) │
└─────────────────┘
│
▼
┌─────────────────┐
│ Confidence Score│
│ + Synthesis │
└─────────────────┘
Pricing
| API | Approximate Cost |
|---|---|
| Web search | ~$0.001 |
| Scholar search | ~$0.002 |
| Smart search | ~$0.002 |
| Tavily search | ~$0.002 |
| Explain | ~$0.003 |
| Perplexity Sonar | ~$0.005 |
Resources
- AIsa — Get your API key
- AIsa API Reference — Complete endpoint documentation
- ClawHub — Plugin marketplace
License
MIT
安全使用建议
This skill is internally consistent and appears to do what it claims: query AIsa endpoints and synthesize results. Before installing, confirm you trust the AIsa service (https://aisa.one) because any queries, provided URLs, and extracted page contents will be sent to that external API using your AISA_API_KEY. Do not submit private/internal URLs or sensitive documents unless your organizational policy allows sending them to an external service. If you want an extra check, inspect the bundled index.ts and scripts/search_client.py (already included) to verify there is no unexpected behavior, and consider using a scoped API key or monitoring usage/quotas on the AIsa account. If you need stricter guarantees about data handling, request the provider's privacy/data-retention policy before use.
功能分析
Type: OpenClaw Skill
Name: aisa-multi-search-engine
Version: 1.0.2
The skill bundle provides a legitimate multi-source search interface using the AIsa API. Analysis of index.ts and scripts/search_client.py shows standard HTTP communication with the documented API endpoint (api.aisa.one) for web, academic, and deep research queries. No evidence of data exfiltration, malicious execution, or prompt injection was found; the code is well-structured, uses TypeBox for input validation, and adheres to its stated purpose.
能力标签
能力评估
Purpose & Capability
Name/description advertise multi-source search via the AIsa API and the package, SKILL.md, index.ts, and scripts/search_client.py consistently implement calls to aisa.one endpoints. Required binaries (python3 and node) are justified by the included TypeScript plugin and Python CLI. No unrelated credentials or tools are requested.
Instruction Scope
SKILL.md and the included code instruct the agent to call AIsa endpoints, run the Python CLI, and accept queries/URLs. The instructions do not attempt to read other environment variables, system files, or other skills' configs. Note: tavily_extract and crawl functionality will send provided URLs and their content to the external AIsa service — avoid submitting private/internal URLs or sensitive content if you do not trust the remote service or its data-handling policy.
Install Mechanism
There is no install spec (instruction-only runtime), so nothing is downloaded or executed during installation. Code files are bundled, but no remote installers, archives, or third-party package installs are performed by the skill itself. The only dependency shown in package.json is @sinclair/typebox, which is reasonable for schema validation.
Credentials
The skill requires a single credential, AISA_API_KEY, which is the expected credential for an API-backed search plugin. No unrelated secrets or wide-ranging system config paths are requested. The code uses this key exclusively to authenticate to api.aisa.one.
Persistence & Privilege
always:false and there is no behavior that modifies other skills or system-wide settings. The skill does not request permanent presence or elevated privileges beyond normal autonomous invocation capability.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install aisa-multi-search-engine - 安装完成后，直接呼叫该 Skill 的名称或使用
/aisa-multi-search-engine触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.2
**Major update: Expanded multi-source search, new endpoints, and improved configuration.**
- Added support for Node.js and updated Python3 requirement in environment.
- Expanded and clarified tool coverage: web, scholar, smart, Tavily search/extract, Perplexity research, and deterministic confidence scoring.
- New detailed documentation for each agent tool, API endpoints, scoring methodology, configuration options, and usage examples.
- Updated license to MIT and incremented version to 1.0.2.
- Removed redundant or outdated metadata and simplified quick start steps.
- Enhanced compatibility information and platform-specific integration notes.
v1.0.0
Initial release of aisa-multi-search-engine
- Combines web search, scholar search, Tavily, smart search, and Perplexity-style research in one skill.
- Supports multi-source evidence gathering, web/academic research, URL extraction, and synthesis.
- Requires a single AISA API key and Python 3 runtime.
- Designed for research, due diligence, and market scanning workflows.
- Includes clear usage guidelines, example queries, and setup instructions.
元数据
常见问题
AIsa Multi Search Engine 是什么？
Multi-source search engine powered by AIsa API. Combines Tavily web search, Scholar academic search, Smart hybrid search, and Perplexity deep research — all... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 106 次。
如何安装 AIsa Multi Search Engine？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install aisa-multi-search-engine」即可一键安装，无需额外配置。
AIsa Multi Search Engine 是免费的吗？
是的，AIsa Multi Search Engine 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
AIsa Multi Search Engine 支持哪些平台？
AIsa Multi Search Engine 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 AIsa Multi Search Engine？
由 bibaofeng（@bibaofeng）开发并维护，当前版本 v1.0.2。
推荐 Skills