- Home
- Skills
- Kjgarza
- Marketplace Claude
- Searching Academic Outputs With Dimensions
searching-academic-outputs-with-dimensions_skill
- Python
2
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill kjgarza/marketplace-claude --skill searching-academic-outputs-with-dimensions- dsl-reference.md5.8 KB
- SKILL.md5.8 KB
Overview
This skill queries the Dimensions academic database to find publications, grants, patents, clinical trials, datasets, and researcher profiles using the shortwing CLI. It helps product teams, researchers, and analysts locate empirical evidence and scholarly outputs to inform decisions, validate hypotheses, and map research trends. Authentication requires a Dimensions API key configured via environment variables or a local config file.
How this skill works
You compose a Dimensions Search Language (DSL) query and pipe it to shortwing, which sends the request to the Dimensions API and returns structured results. Queries follow the pattern: search <source> [for "terms"] [where <filters>] return <result> [limit N]. The DSL supports field selection, filters, aggregations, sorting, and boolean operators for precise retrieval.
When to use it
- To find peer-reviewed studies or empirical evidence supporting a product decision
- When conducting a literature review or building a reading list on a topic
- To identify funding patterns, major funders, and grant amounts in a field
- To discover researchers, their affiliations, and collaboration networks
- To locate related patents or clinical trials tied to a technology or intervention
Best practices
- Use targeted indexes (title_only or title_abstract_only) for precision before broadening to full-text
- Combine phrase searches and filters (year, times_cited) to prioritize recent and influential work
- Return only required fields (use + to combine) to simplify parsing and reduce payload
- Use boolean operators in UPPERCASE inside quotes for complex term logic (AND/OR/NOT)
- Limit results during exploratory searches and increase limit for comprehensive exports
Example use cases
- Search publications for "machine learning" and sort by times_cited to find seminal papers
- Search grants where funders.name~"NIH" to list major funded projects and funding_usd
- Search clinical_trials for "diabetes" to compile trial phases and outcomes
- Search researchers for "Jane Doe" to retrieve profiles and current research organizations
- Search patents where filing_year>=2023 to monitor recent filings in a technology area
FAQ
Obtain an API key from your Dimensions account and set DIMENSIONS_KEY or create ~/.dimensions/dsl.ini with credentials.
How do I limit result size?
Append limit N to your query, e.g. return publications limit 10.
Can I filter by citation count or year?
Yes. Use where clauses like times_cited>100 or year in [2020:2024].