← 返回 Skills 市场
plagtech

Multi-Search

作者 Plag · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
42
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install multi-search-skill
功能描述
Search 8 knowledge domains in one query — live web, 250M academic papers, 150M scholarly works, arXiv preprints, 36M biomedical papers, chemical compounds, g...
使用说明 (SKILL.md)

Multi Domain Search

Search across 8 knowledge domains from a single agent — not 8 web engines, 8 completely different databases. One query can search the live web, academic papers, preprints, biomedical literature, chemical compounds, scholarly works, government datasets, and the dictionary simultaneously.

How to call endpoints

bash {baseDir}/scripts/search.sh METHOD ENDPOINT '{"key":"value"}'

Search strategy

When the user asks you to search for something, decide which domains are relevant and fan out across them. Don't search every domain for every query — be smart about it.

General knowledge questions — start with web_search and web_qna, add scholarly_search if the topic has academic depth.

Scientific or medical questions — lead with biomedical_search and papers_search, supplement with preprints_search for cutting-edge work, then web_search for recent news coverage.

Chemistry questions — start with chemistry_compound, then biomedical_search for related drug/pharma research, then papers_search for fundamental science.

Data and statistics — check demographics_datasets for government data, scholarly_search for published studies with data, web_search for recent reports.

Word meanings and language — use dictionary_define and dictionary_synonyms for precise definitions, then web_search for usage context.

Comprehensive research — fan out across all relevant domains, then synthesize. Start broad, then drill into the most promising domain.

Always tell the user which domains you searched and what each returned. Cite sources with DOIs, PMIDs, URLs, or arXiv IDs.

Available search domains (8 engines, 15 endpoints)

1. Live Web (Tavily)

Web Search — $0.02 Search the live web. Returns ranked URLs with titles and snippets.

bash {baseDir}/scripts/search.sh POST /api/v1/search/web '{"query":"CRISPR gene therapy 2026 breakthroughs"}'

Web Extract — $0.02 Extract clean text from URLs found via web search.

bash {baseDir}/scripts/search.sh POST /api/v1/search/extract '{"urls":["https://example.com/article"]}'

Web Q&A — $0.03 Direct answers from the web with sources. Best for factual questions.

bash {baseDir}/scripts/search.sh POST /api/v1/search/qna '{"query":"How many people use AI tools for research in 2026?"}'

2. Academic Papers (OpenAlex — 250M papers)

Search Papers — $0.002 Full-text search across 250 million academic papers. Returns titles, abstracts, authors, DOIs, citation counts.

bash {baseDir}/scripts/search.sh GET /api/v1/research/papers/search '{"query":"transformer attention mechanisms"}'

By Author — $0.002 Find all papers by a specific researcher.

bash {baseDir}/scripts/search.sh GET /api/v1/research/papers/by-author '{"author":"Geoffrey Hinton"}'

Trending — $0.002 Discover what's trending in any research field right now.

bash {baseDir}/scripts/search.sh GET /api/v1/research/papers/trending '{"topic":"large language models","days":7}'

3. Scholarly Works (Crossref — 150M works)

Search Scholarly — $0.002 Search across 150 million works — books, conference papers, datasets, journal articles, and more. Broader than OpenAlex.

bash {baseDir}/scripts/search.sh GET /api/v1/research/scholarly/search '{"query":"reinforcement learning from human feedback"}'

Journal Info — $0.001 Look up journal metadata, impact, and subject areas by ISSN.

bash {baseDir}/scripts/search.sh GET /api/v1/research/scholarly/journal-info '{"issn":"0028-0836"}'

4. Preprints (arXiv)

Search Preprints — $0.002 Search arXiv for the latest preprints before they're formally published. Cutting-edge research.

bash {baseDir}/scripts/search.sh GET /api/v1/research/preprints/search '{"query":"diffusion models image generation","category":"cs.CV"}'

Recent by Category — $0.002 Browse the latest preprints in any arXiv category.

bash {baseDir}/scripts/search.sh GET /api/v1/research/preprints/recent '{"category":"cs.AI"}'

5. Biomedical (PubMed — 36M papers)

Search Biomedical — $0.002 Search 36 million biomedical papers — medicine, biology, pharmacology, public health, genetics, neuroscience.

bash {baseDir}/scripts/search.sh GET /api/v1/research/biomedical/search '{"query":"mRNA vaccine long term immunity"}'

Related Articles — $0.002 Find conceptually similar papers to a known PMID.

bash {baseDir}/scripts/search.sh GET /api/v1/research/biomedical/related '{"pmid":"33116299"}'

6. Chemistry (PubChem)

Compound Search — $0.002 Search PubChem by compound name, formula, or CID. Returns structure, properties, identifiers.

bash {baseDir}/scripts/search.sh GET /api/v1/research/chemistry/compound '{"query":"ibuprofen"}'

Similar Compounds — $0.002 Find structurally similar compounds for drug discovery and chemical research.

bash {baseDir}/scripts/search.sh GET /api/v1/research/chemistry/similarity '{"cid":"2244"}'

7. Government Data (Data.gov + Census)

Dataset Search — $0.001 Search open government datasets across Data.gov.

bash {baseDir}/scripts/search.sh GET /api/v1/research/demographics/datasets '{"query":"renewable energy production by state"}'

Census Data — $0.001 US Census data by state, county, or zip code. Population, income, education, housing.

bash {baseDir}/scripts/search.sh GET /api/v1/research/demographics/census '{"location":"California"}'

8. Dictionary

Define — $0.001 Full definition with phonetics, part of speech, and examples.

bash {baseDir}/scripts/search.sh GET /api/v1/research/dictionary/define '{"word":"epistemology"}'

Synonyms — $0.001 Synonyms and antonyms. Also useful for expanding search queries across other domains.

bash {baseDir}/scripts/search.sh GET /api/v1/research/dictionary/synonyms '{"word":"innovative"}'

Cost per domain

Domain Endpoints Cost per call
Live Web 3 $0.02–$0.03
Academic Papers 3 $0.002
Scholarly Works 2 $0.001–$0.002
Preprints 2 $0.002
Biomedical 2 $0.002
Chemistry 2 $0.002
Government Data 2 $0.001
Dictionary 2 $0.001

A typical multi-domain search across 3-4 domains costs $0.006–$0.04.

Data sourced from Tavily, OpenAlex, Crossref, arXiv, PubMed, PubChem, Data.gov, US Census Bureau.

安全使用建议
Install only if you are comfortable sending search terms and extracted URLs to the configured research gateway and paying the listed per-call costs. Avoid using it for secrets, proprietary text, regulated personal data, or sensitive investigations unless you trust the gateway and have reviewed the billing setup.
能力标签
cryptofinancial-authorityrequires-walletcan-make-purchasesrequires-paid-servicerequires-sensitive-credentials
能力评估
Purpose & Capability
The stated purpose is multi-domain research search, and the artifacts consistently implement that through documented search endpoints covering web, scholarly, biomedical, chemistry, government data, and dictionary sources.
Instruction Scope
The skill tells the agent to choose relevant domains and avoid searching every domain, but it does not require per-call user confirmation before external paid searches.
Install Mechanism
Installation is a normal OpenClaw skill install with one shell helper script; no post-install hooks, package dependencies, or hidden install behavior were found.
Credentials
It requires curl, python3, outbound network access, and a RESEARCH_API_KEY for a paid gateway, which fits the search purpose but means user queries and URLs leave the local environment.
Persistence & Privilege
The script performs request construction and curl calls only; it does not write files, install persistence, spawn background workers, read local credential stores, or modify local state.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install multi-search-skill
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /multi-search-skill 触发
  4. 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
v1.0.0
multi-domain-search 1.0.0 - Initial release enabling unified search across 8 major knowledge domains: live web, academic papers, scholarly works, preprints, biomedical, chemistry, government data, and dictionary. - Provides 15 endpoints for targeted, domain-specific queries via a single agent. - Includes clear search strategies and cost breakdowns per domain. - Supports smart selection of relevant domains per user query. - Delivers source citations in responses for transparency and reliability.
元数据
Slug multi-search-skill
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Multi-Search 是什么?

Search 8 knowledge domains in one query — live web, 250M academic papers, 150M scholarly works, arXiv preprints, 36M biomedical papers, chemical compounds, g... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 42 次。

如何安装 Multi-Search?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install multi-search-skill」即可一键安装,无需额外配置。

Multi-Search 是免费的吗?

是的,Multi-Search 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Multi-Search 支持哪些平台?

Multi-Search 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 Multi-Search?

由 Plag(@plagtech)开发并维护,当前版本 v1.0.0。

💬 留言讨论