← 返回 Skills 市场
Intelligent Search Agent
作者
bowen-dotcom
· GitHub ↗
· v1.0.2
· MIT-0
1129
总下载
1
收藏
3
当前安装
3
版本数
在 OpenClaw 中安装
/install aisa-search-skill
功能描述
Intelligent search for agents. Multi-source retrieval across web, scholar, Tavily, and Perplexity Sonar models.
使用说明 (SKILL.md)
\r \r
OpenClaw Search\r
\r Intelligent search for autonomous agents, powered by AIsa.\r \r One API key gives you:\r
- Structured web search\r
- Scholar search\r
- Hybrid scholar search\r
- Tavily search and extraction tools\r
- Perplexity Sonar answer-generation endpoints with citations\r \r
What This Skill Is Best For\r
\r
Fast web lookup\r
Search the latest AI infrastructure launches and summarize the top sources.\r
```\r
\r
### Scholar-backed research\r
```text\r
Find recent papers on multimodal reasoning from 2024 onward.\r
```\r
\r
### Citation-rich answers\r
```text\r
Use Sonar Pro to answer which open-source agent frameworks are gaining traction and cite sources.\r
```\r
\r
### Deep research reports\r
```text\r
Use Sonar Deep Research to produce a thorough market map of AI browser agents.\r
```\r
\r
## Quick Start\r
\r
```bash\r
export AISA_API_KEY="your-key"\r
```\r
\r
## Search APIs\r
\r
### Web Search\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/web?query=AI+frameworks&max_num_results=10" \\r
-H "Authorization: Bearer $AISA_API_KEY"\r
```\r
\r
### Scholar Search\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/scholar?query=transformer+models&max_num_results=10" \\r
-H "Authorization: Bearer $AISA_API_KEY"\r
\r
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/scholar?query=LLM&max_num_results=10&as_ylo=2024&as_yhi=2025" \\r
-H "Authorization: Bearer $AISA_API_KEY"\r
```\r
\r
### Hybrid Scholar Search\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/scholar/search/smart?query=machine+learning+optimization&max_num_results=10" \\r
-H "Authorization: Bearer $AISA_API_KEY"\r
```\r
\r
## Perplexity Sonar APIs\r
\r
The deprecated `/search/full` and `/search/smart` nodes have been removed from this skill.\r
\r
The replacement flow is the Perplexity API family:\r
\r
| Endpoint | Use case |\r
|----------|----------|\r
| `/perplexity/sonar` | Lightweight, cost-effective search answers with citations |\r
| `/perplexity/sonar-pro` | Better for complex queries and multi-step follow-ups |\r
| `/perplexity/sonar-reasoning-pro` | Stronger analytical reasoning with web search |\r
| `/perplexity/sonar-deep-research` | Exhaustive research and long-form reports |\r
\r
These descriptions are based on the AIsa docs:\r
- [Sonar](https://docs.aisa.one/reference/post_perplexity-sonar)\r
- [Sonar Pro](https://docs.aisa.one/reference/post_perplexity-sonar-pro)\r
- [Sonar Reasoning Pro](https://docs.aisa.one/reference/post_perplexity-sonar-reasoning-pro)\r
- [Sonar Deep Research](https://docs.aisa.one/reference/post_perplexity-sonar-deep-research)\r
\r
### Sonar\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/perplexity/sonar" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{\r
"model": "sonar",\r
"messages": [\r
{"role": "user", "content": "What changed in the AI agent ecosystem this week?"}\r
]\r
}'\r
```\r
\r
### Sonar Pro\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/perplexity/sonar-pro" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{\r
"model": "sonar-pro",\r
"messages": [\r
{"role": "user", "content": "Compare the top browser-use agent frameworks and cite the key differences."}\r
]\r
}'\r
```\r
\r
### Sonar Reasoning Pro\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/perplexity/sonar-reasoning-pro" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{\r
"model": "sonar-reasoning-pro",\r
"messages": [\r
{"role": "user", "content": "Analyze whether small vertical AI agents can defend against general-purpose copilots."}\r
]\r
}'\r
```\r
\r
### Sonar Deep Research\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/perplexity/sonar-deep-research" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{\r
"model": "sonar-deep-research",\r
"messages": [\r
{"role": "user", "content": "Create a deep research report on AI coding agents in 2026, including product categories, pricing, and risks."}\r
]\r
}'\r
```\r
\r
## Tavily APIs\r
\r
```bash\r
curl -X POST "https://api.aisa.one/apis/v1/tavily/search" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{"query":"latest AI developments"}'\r
\r
curl -X POST "https://api.aisa.one/apis/v1/tavily/extract" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{"urls":["https://example.com/article"]}'\r
\r
curl -X POST "https://api.aisa.one/apis/v1/tavily/crawl" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{"url":"https://example.com","max_depth":2}'\r
\r
curl -X POST "https://api.aisa.one/apis/v1/tavily/map" \\r
-H "Authorization: Bearer $AISA_API_KEY" \\r
-H "Content-Type: application/json" \\r
-d '{"url":"https://example.com"}'\r
```\r
\r
## Python Client\r
\r
```bash\r
# Structured search\r
python3 {baseDir}/scripts/search_client.py web --query "latest AI news" --count 10\r
python3 {baseDir}/scripts/search_client.py scholar --query "transformer architecture" --count 10\r
python3 {baseDir}/scripts/search_client.py smart --query "autonomous agents" --count 10\r
\r
# Perplexity Sonar family\r
python3 {baseDir}/scripts/search_client.py sonar --query "Summarize this week's AI launches"\r
python3 {baseDir}/scripts/search_client.py sonar-pro --query "Compare AI agent frameworks with citations"\r
python3 {baseDir}/scripts/search_client.py sonar-reasoning-pro --query "Analyze the defensibility of AI copilots"\r
python3 {baseDir}/scripts/search_client.py sonar-deep-research --query "Write a deep research report on AI browser agents"\r
\r
# Optional system instruction\r
python3 {baseDir}/scripts/search_client.py sonar-pro \\r
--query "Map the top coding agent products" \\r
--system "Respond in markdown with a short executive summary first."\r
\r
# Tavily utilities\r
python3 {baseDir}/scripts/search_client.py tavily-search --query "AI developments"\r
python3 {baseDir}/scripts/search_client.py tavily-extract --urls "https://example.com/article"\r
\r
# Multi-source retrieval\r
python3 {baseDir}/scripts/search_client.py verity --query "Is quantum computing ready for enterprise?"\r
```\r
\r
## API Reference\r
\r
| Endpoint | Method | Description |\r
|----------|--------|-------------|\r
| `/scholar/search/web` | POST | Web search with structured results |\r
| `/scholar/search/scholar` | POST | Academic paper search |\r
| `/scholar/search/smart` | POST | Hybrid scholar search |\r
| `/scholar/explain` | POST | Generate result explanations |\r
| `/perplexity/sonar` | POST | Lightweight search answers with citations |\r
| `/perplexity/sonar-pro` | POST | Advanced search answers for complex tasks |\r
| `/perplexity/sonar-reasoning-pro` | POST | Analytical reasoning with web search |\r
| `/perplexity/sonar-deep-research` | POST | Exhaustive research reports |\r
| `/tavily/search` | POST | Tavily search integration |\r
| `/tavily/extract` | POST | Extract content from URLs |\r
| `/tavily/crawl` | POST | Crawl web pages |\r
| `/tavily/map` | POST | Generate site maps |\r
\r
## Parameters\r
\r
### Scholar search query parameters\r
\r
| Parameter | Type | Description |\r
|-----------|------|-------------|\r
| `query` | string | Search query |\r
| `max_num_results` | integer | Max results (default 10) |\r
| `as_ylo` | integer | Year lower bound |\r
| `as_yhi` | integer | Year upper bound |\r
\r
### Perplexity request body\r
\r
This skill sends a minimal OpenAI-style payload:\r
\r
```json\r
{\r
"model": "sonar-pro",\r
"messages": [\r
{"role": "system", "content": "Optional system instruction"},\r
{"role": "user", "content": "Your question"}\r
]\r
}\r
```\r
\r
Use `messages` because the AIsa Perplexity endpoints are presented as "Ask AI" endpoints in the official docs. This skill keeps the payload intentionally small for broad compatibility.\r
\r
## Notes\r
\r
- `/search/full` and `/search/smart` are no longer documented here because you indicated those nodes have been retired.\r
- The existing scholar and Tavily endpoints remain available.\r
- `verity` still focuses on parallel retrieval from scholar, web, hybrid scholar, and Tavily sources.\r
\r
## Full API Reference\r
\r
See [API Reference](https://docs.aisa.one/reference/) for complete endpoint documentation.\r
\r
## Resources\r
\r
- [AIsa Verity](https://github.com/AIsa-team/verity) - Reference implementation of confidence-scored search agent\r
- [AIsa Documentation](https://docs.aisa.one) - Complete API documentation\r
安全使用建议
This skill appears to do what it says: it sends queries and receives search/answer data from api.aisa.one using the AISA_API_KEY. Before installing, confirm you trust the AIsa service and the skill author (owner is an unknown ID here). Consider creating a scoped or billing-limited API key if possible, and avoid sending sensitive or private documents through this skill since queries and any attached content will be transmitted to the external API. If you plan to allow autonomous agent invocation, remember the agent may call the API without prompting—limit scope or approval as needed.
功能分析
Type: OpenClaw Skill
Name: aisa-search-skill
Version: 1.0.2
The skill provides a legitimate interface for the AIsa search API, enabling agents to perform web, academic, and AI-powered searches via endpoints at api.aisa.one. The Python client in `scripts/search_client.py` is well-structured, using only standard libraries to handle HTTP requests, and includes features like parallel multi-source retrieval and confidence scoring. No indicators of data exfiltration, malicious execution, or prompt injection were found.
能力评估
Purpose & Capability
Name/description (Intelligent Search across web, scholar, Tavily, Perplexity Sonar) matches the requested binary requirements (curl, python3), the single environment variable (AISA_API_KEY), and the client code which calls api.aisa.one endpoints. There are no unrelated credentials or unexpected dependencies.
Instruction Scope
SKILL.md instructs the agent to call AIsa endpoints and shows curl/python examples that use only AISA_API_KEY. The included Python client implements those calls and does not read other environment variables or local files beyond the API key. There are no instructions to read sensitive local files or to post data to third-party endpoints beyond api.aisa.one.
Install Mechanism
No install spec is present (instruction-only skill plus a Python script). That is low-risk: nothing will be downloaded or extracted during installation. Required binaries are standard (curl, python3).
Credentials
The skill requires a single API credential (AISA_API_KEY), which is appropriate for a client that calls AIsa endpoints. No additional tokens, keys, or config paths are requested.
Persistence & Privilege
always is false and the skill does not request elevated platform privileges or modify other skills. It does not persist credentials to unrelated locations.
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-search-skill - 安装完成后,直接呼叫该 Skill 的名称或使用
/aisa-search-skill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
**OpenClaw Search v1.0.2 — Now supporting Perplexity Sonar APIs and simpler docs**
- Added integration and documentation for Perplexity Sonar, Sonar Pro, Sonar Reasoning Pro, and Sonar Deep Research APIs for answer generation with citations.
- Removed deprecated `/search/full` and `/search/smart` endpoints; Perplexity endpoints are the recommended replacement.
- Simplified and clarified documentation, with concise use cases and streamlined API references.
- Updated metadata (emoji) and quick start instructions.
- The skill now provides multi-source search (web, scholar, Tavily, Perplexity Sonar) under one API key.
v1.0.1
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.
v1.0.0
## Intelligent Search 🔍 - Multi-Source Search with Confidence Scoring
### Overview
Intelligent Search provides advanced search capabilities with multi-source retrieval and deterministic confidence scoring. Inspired by the AIsa Verity project, this skill implements a sophisticated two-phase search orchestration that evaluates source credibility and cross-validation.
### Key Features
- **🌐 Multi-Source Retrieval**: Parallel search across Web, Academic, Smart, and Tavily sources
- **🎯 Confidence Scoring**: Deterministic scoring algorithm (0-100) based on source quality and consensus
- **📊 Meta-Analysis**: Cross-source validation and agreement analysis
- **🏛️ Architectural Clarity**: Clear two-phase orchestration (Discovery + Reasoning)
- **🔍 Tavily Integration**: External search engine for additional validation signals
- **📝 Result Explanation**: Generate summaries and explanations from search results
### Search Architecture
#### Phase 1: Parallel Discovery
Query 4 distinct search streams simultaneously:
- **Scholar**: Deep academic paper retrieval
- **Web**: Structured web search results
- **Smart**: Intelligent hybrid search (web + academic)
- **Tavily**: External validation and broader web coverage
#### Phase 2: Confidence Scoring
Deterministic scoring based on:
- **Source Quality** (40%): Academic > Smart/Web > External sources
- **Agreement Analysis** (35%): Cross-source consensus checking
- **Data Availability** (15%): Volume and recency of results
- **Reliability** (10%): Error-free responses across sources
### Confidence Score Interpretation
| Score Range | Level | Meaning |
|-------------|-------|---------|
| 90-100 | Very High | Strong consensus across high-quality sources |
| 70-89 | High | Good agreement from reliable sources |
| 50-69 | Medium | Mixed signals, verify independently |
| 30-49 | Low | Conflicting sources, use with caution |
| 0-29 | Very Low | Insufficient or contradictory data |
### Files Included
元数据
常见问题
Intelligent Search Agent 是什么?
Intelligent search for agents. Multi-source retrieval across web, scholar, Tavily, and Perplexity Sonar models. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1129 次。
如何安装 Intelligent Search Agent?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install aisa-search-skill」即可一键安装,无需额外配置。
Intelligent Search Agent 是免费的吗?
是的,Intelligent Search Agent 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Intelligent Search Agent 支持哪些平台?
Intelligent Search Agent 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 Intelligent Search Agent?
由 bowen-dotcom(@bowen-dotcom)开发并维护,当前版本 v1.0.2。
推荐 Skills