What problem does it solve?
PubMed searches and cross-database linking are time-consuming and error-prone when you need accurate PMIDs, abstracts, and entity mappings (genes, proteins, sequences, and PubChem chemicals) for scientific research and clinical literature triage.
Core Features & Use Cases
- Structured literature retrieval: Search PubMed with NCBI query syntax, fetch article abstracts/metadata in batches, and optionally retrieve open-access full text from PMC in the BioC OA subset.
- Cross-database entity linking: Resolve biological entities discussed in a paper by traversing NCBI ELink relationships and then summarize resolved IDs into human-readable metadata.
- Accuracy and scaling guardrails: Verify biomedical spelling before searching, match messy citations to PMIDs, and cache large PMID sets for bulk processing to avoid rate limits and context overflow.
- Use Case: Given a topic query (e.g., a gene + disease), retrieve candidate papers, extract abstracts, link each paper to associated genes/compounds/sequences, then produce a consolidated, grounded list of entity metadata.
Quick Start
Use PubMed to find up to five relevant PMIDs for a topic query using npx-style browsing via the skill’s command interface.
