TwoSCI
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| WOS_API_KEY | No | Web of Science Starter API key used as a fallback when WOS_EXPANDED_API_KEY is not configured. | |
| OPENALEX_EMAIL | No | Contact email for the OpenAlex polite pool. | |
| CROSSREF_MAILTO | No | Contact email for the Crossref polite pool. | |
| ELSEVIER_API_KEY | No | Elsevier API key for Scopus detail and abstract entitlement. | |
| WOS_EXPANDED_API_KEY | No | Web of Science Expanded API key, the preferred WoS product. | |
| SCOPUS_SEARCH_API_KEY | No | Optional dedicated Scopus Search API key. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_papersC | Search academic papers from multiple sources including arXiv, Web of Science, etc. |
| search_arxivC | Search academic papers specifically from arXiv preprint server |
| search_webofscienceB | Search Web of Science. Defaults to Expanded when its key is configured, otherwise Starter; explicitly selecting a product is optional. |
| get_webofscience_related_recordsB | Get references, citing records, or related records through Web of Science Expanded API |
| search_pubmedC | Search biomedical literature from PubMed/MEDLINE database using NCBI E-utilities API |
| search_biorxivB | Search bioRxiv preprint server for biology papers |
| search_medrxivC | Search medRxiv preprint server for medical papers |
| search_semantic_scholarC | Search Semantic Scholar for academic papers with citation data |
| search_iacrC | Search IACR ePrint Archive for cryptography papers |
| download_paperC | Download PDF file of an academic paper |
| download_public_paperC | Download a public paper from Publisher, Google Scholar, or Sci-Hub using bounded page fallback and local PDF validation |
| get_paper_markdownB | Explicitly retrieve bounded, untrusted ScrapingAnt Markdown for a Publisher, Scholar, or Sci-Hub paper reference |
| search_google_scholarC | Search Google Scholar for academic papers using web scraping |
| get_paper_by_doiC | Retrieve paper information using DOI from available platforms |
| discover_paper_accessA | Discover a public publisher PDF candidate for one DOI without claiming open-license or complete download success |
| search_scihubA | Controlled, opt-in Sci-Hub DOI adapter (disabled by default); accepts DOI/doi.org URLs only, tries direct mirrors before bounded ScrapingAnt fallback, and does not grant access rights. |
| check_scihub_mirrorsC | Check the health status of all Sci-Hub mirror sites |
| get_platform_statusC | Check the status and capabilities of available academic platforms |
| search_sciencedirectC | Search academic papers from Elsevier ScienceDirect database (requires API key) |
| search_springerA | Search academic papers from Springer Nature database. Uses Metadata API by default (all content) or OpenAccess API when openAccess=true (full text available). Same API key works for both. |
| search_scopusB | Search the Scopus abstract and citation database (requires Elsevier API key). COMPLETE view returns abstracts and is paginated. |
| search_crossrefC | Search academic papers from Crossref database. Free API with extensive scholarly metadata coverage across publishers. |
| search_openalexC | Search OpenAlex for fast free discovery and one-hop citation enrichment |
| create_research_briefA | Create a research brief from a seed paper plus the user intent. Resolves the seed with WoS Expanded (when configured), Scopus, OpenAlex, and Crossref; returns its abstract, keywords, references, citing sample, and a brief_id for later searches. |
| search_literatureA | Unified deduplicated literature search across WoS (Expanded by default when configured), Scopus, OpenAlex, and Crossref. Returns candidates with source provenance and a search_id. Use layer=metadata for cheap triage, abstract for screening, evidence for deeper evidence packets. Screening should be delegated to subagents in batches. |
| get_literature_recordsA | Hydrate stored search results by record_id, or resolve DOIs, without re-running discovery. By default returns the full abstract and references already held in the session. Set refresh=true only when fresh provider metadata is required. |
| expand_citation_graphA | Expand one hop from seed papers through references, citing papers, or related records. Uses WoS Expanded first when the seed has a WoS UID, then OpenAlex, then Crossref. Returns graph nodes, typed edges, and a search_id for export or hydration. |
| save_screeningB | Persist include/exclude/maybe decisions and reasons from abstract-screening subagents. The decision ledger is attached to the search_id and can be exported or reviewed later. |
| export_literatureB | Export a search result set to BibTeX, RIS, or JSON. Export is user-initiated and is the intended way to retain WoS/Scopus records outside the process. |
| get_source_statusA | Report configured sources, WoS default product, quota/rate status, OpenAlex/Crossref cache state, and in-memory brief/search session counts. Set validate=true only when an actual key validation request is acceptable. |
| get_citationsA | Retrieve citation data (citation count, references, venue) for a paper by DOI using Semantic Scholar |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 31 tools
Per-source searches (arxiv, pubmed, scopus, crossref, etc.) are clearly distinct, but the set also contains three overlapping entry points for general discovery: search_papers, search_literature, and the union of all source-specific tools, with no guidance on when to prefer each. The download path is similarly crowded (download_paper, download_public_paper, get_paper_markdown, discover_paper_access, get_paper_by_doi), and get_platform_status vs get_source_status plus get_citations vs get_literature_records are near-duplicates that an agent can easily pick wrongly.
Every name is snake_case with a leading verb (search_, get_, download_, create_, save_, export_, check_, expand_, discover_) followed by a clear object. The verb varies by semantics rather than arbitrarily, so patterns like search_<source> and get_<resource> are immediately predictable.
31 tools exceeds a comfortable surface for an agent, and the count is inflated by redundancy rather than coverage: twelve source-specific search tools sit alongside two generic search tools (search_papers, search_literature) that already aggregate them. Several download/status tools could be consolidated or folded into options on existing tools.
The literature workflow is covered end to end: discovery across many sources, hydration by record_id/DOI, one-hop citation expansion, screening persistence, export to BibTeX/RIS/JSON, PDF retrieval, and source health/status. Minor gaps exist around research briefs (create_research_brief exists but no get/list/delete brief) and no way to inspect or clean up stored search sessions beyond the source-status counts.