dsh-academic-paper-search

When you are drafting a survey, verifying a citation, or tracing a literature trail inside a DSH workflow, constantly switching to a browser or a separate terminal breaks your focus. dsh-academic-paper-search is a thin adapter Bundle: it installs the official dsh-mcp-client and mounts the existing nature-academic-search Python MCP service under the academic_search namespace, so the model calls a single unified tool set without duplicating any search logic locally.

Out of the box, paper search spans Crossref, PubMed, arXiv, OpenAlex, and Europe PMC simultaneously, with automatic deduplication by DOI, PMID, PMCID, arXiv ID, and OpenAlex ID. Go deeper by pulling references and cited-by citation graphs, querying ClinicalTrials.gov records through a dedicated entity type, or enriching metadata via Semantic Scholar. All capabilities run through one Python MCP runtime; the Bundle only exposes the tools to DSH and never ships a second copy of the search implementation.

It suits researchers and developers who need frequent literature retrieval, citation verification, or citation-graph exploration within DSH. Keep in mind that it does not connect to Google Scholar, Web of Science, Scopus, Embase, CNKI, or Wanfang, and it does not automatically treat model output as a verified citation. Every result carries sources_succeeded, sources_skipped, errors, and verification-status fields so you can judge the reliability of each record yourself.