DigiLIB
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_papersA | Search academic literature across OpenAlex, arXiv, Semantic Scholar, and Crossref. Args: keywords: key terms, e.g. ["transformer attention"]. Craft 2-5 focused queries (expand acronyms, add synonyms, use technical phrasing) for the best coverage. year_from: restrict to papers from this year onwards. Applied by OpenAlex and Crossref only. year_to: restrict to papers up to this year. Applied by OpenAlex and Crossref only. size: results per source per keyword (default 10, max 40). Returns: A numbered plain-text list of deduplicated results with title, authors, year, source, link, and abstract per entry. |
| search_booksA | Search books across Open Library and Google Books. Args: keywords: key terms; broader, book-friendly phrasing works best. size: results per source per keyword (default 10). Returns: A numbered plain-text list of deduplicated book results with title, authors, year, source, link, and description per entry. |
| export_research_toolA | Build the LaTeX source for a research export (literature review + conversation summary + citations in APA/Chicago/MLA) and return it. This tool does NOT render a PDF - it returns the raw LaTeX document source. You, the assistant calling this tool, are responsible for rendering it yourself:
Args: conversation_summary: plain-text summary of the conversation, written by the calling LLM. literature_recommended: list of plain-text lines for the Literature Review section (one bullet per line). liked_articles: citation entries, each a dict with title/authors/year/link keys. citation_formats: subset of ["APA", "Chicago", "MLA"]. Returns: The complete LaTeX document source as a string. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| digilib_researcher | Return the DigiLIB researcher system prompt (persona/instructions). |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| researcher_prompt_resource | Raw contents of the DigiLIB researcher system prompt. |
TDQS
Scored across 3 tools
search_papers and search_books target clearly distinct resource types, and export_research_tool is a separate export action. No two tools overlap in purpose.
All tools use snake_case with a leading verb, giving a predictable pattern. export_research_tool adds a slightly redundant '_tool' suffix, a minor deviation from the parallel search_* names.
Three tools map directly to the server's focused workflow: search papers, search books, and export research. Each tool earns its place and the set is tight rather than thin.
The core workflow from literature and book search to LaTeX export is covered. Minor gaps exist around dedicated citation lookup, BibTeX export, or full-text retrieval, but search results supply metadata and export handles the output need.