Europe Pmc Tools#

Configuration File: europe_pmc_tools.json Tool Type: Local Tools Count: 7

This page contains all tools defined in the europe_pmc_tools.json configuration file.

Available Tools#

EuropePMC_get_citations (Type: EuropePMCRESTTool)#

Get one page of citations for an article from Europe PMC. Returns exact hitCount-derived paginati…

EuropePMC_get_citations tool specification

Tool Information:

  • Name: EuropePMC_get_citations

  • Type: EuropePMCRESTTool

  • Description: Get one page of citations for an article from Europe PMC. Returns exact hitCount-derived pagination metadata (total_results, total_pages, has_more, and next_page), so callers can retrieve every citing article. Requires a source (e.g., ‘MED’ for PubMed, ‘PMC’ for PMC) and article ID.

Parameters:

  • source (string) (optional) Source database (e.g., ‘MED’ for PubMed, ‘PMC’ for PMC). Usually ‘MED’ for most articles.

  • article_id (string) (required) Article ID from the source database (e.g., PMID for MED source)

  • page_size (integer) (optional) Number of citations to retrieve (default: 25, max: 1000)

  • page (integer) (optional) Page number for pagination (default: 1)

Example Usage:

query = {
    "name": "EuropePMC_get_citations",
    "arguments": {
        "article_id": "example_value"
    }
}
result = tu.run(query)

EuropePMC_get_full_text (Type: EuropePMCStructuredFullTextTool)#

Retrieve and parse full-text XML from Europe PMC into structured sections. Returns title, abstrac…

EuropePMC_get_full_text tool specification

Tool Information:

  • Name: EuropePMC_get_full_text

  • Type: EuropePMCStructuredFullTextTool

  • Description: Retrieve and parse full-text XML from Europe PMC into structured sections. Returns title, abstract, body sections (introduction, methods, results, discussion, conclusions), figure captions, table captions, and references as structured JSON. Accepts a PMC ID directly or a PMID (auto-resolved to PMC ID via Europe PMC search). Only works for open-access articles available in PubMed Central.

Parameters:

  • pmcid (string) (optional) PubMed Central ID (e.g., ‘PMC7096075’ or ‘7096075’). Preferred identifier.

  • pmid (string) (optional) PubMed ID (e.g., ‘32226684’). Will be auto-resolved to a PMC ID via Europe PMC search. Use when PMC ID is not known.

  • max_section_chars (integer) (optional) Maximum characters per section in the output. Longer sections are truncated.

Example Usage:

query = {
    "name": "EuropePMC_get_full_text",
    "arguments": {
    }
}
result = tu.run(query)

EuropePMC_get_fulltext (Type: EuropePMCFullTextFetchTool)#

Fetch a PMC article’s full text with deterministic fallbacks: Europe PMC fullTextXML → NCBI PMC O…

EuropePMC_get_fulltext tool specification

Tool Information:

  • Name: EuropePMC_get_fulltext

  • Type: EuropePMCFullTextFetchTool

  • Description: Fetch a PMC article’s full text with deterministic fallbacks: Europe PMC fullTextXML → NCBI PMC OAI (JATS) → NCBI efetch (XML) → PMC HTML. Designed for machine consumption: returns structured provenance (source/format/content_type) and a retrieval_trace of every attempt.

Parameters:

  • fulltext_xml_url (string) (optional) Direct Europe PMC fullTextXML URL (optional).

  • pmcid (string) (optional) PMC ID (e.g., ‘PMC5234727’ or ‘5234727’).

  • source_db (string) (optional) Europe PMC source database (e.g., ‘MED’ or ‘PMC’). Used with article_id to derive a Europe PMC fullTextXML URL.

  • article_id (string) (optional) Europe PMC article ID within source_db (e.g., PMID for source_db=’MED’).

  • output_format (string) (optional) Return ‘text’ (default; extracted plain text) or ‘raw’ (raw XML/HTML).

  • include_raw (boolean) (optional) If true, also returns a bounded raw field (useful for downstream parsing).

  • max_chars (integer) (optional) Hard cap on characters returned in text (or content when output_format=’raw’).

  • max_raw_chars (integer) (optional) Hard cap on characters returned in raw when include_raw=true.

  • timeout (integer) (optional) HTTP timeout in seconds.

Example Usage:

query = {
    "name": "EuropePMC_get_fulltext",
    "arguments": {
    }
}
result = tu.run(query)

EuropePMC_get_fulltext_snippets (Type: EuropePMCFullTextSnippetsTool)#

Fetch an article’s full text (best-effort) and return bounded text snippets around provided terms…

EuropePMC_get_fulltext_snippets tool specification

Tool Information:

  • Name: EuropePMC_get_fulltext_snippets

  • Type: EuropePMCFullTextSnippetsTool

  • Description: Fetch an article’s full text (best-effort) and return bounded text snippets around provided terms. Uses deterministic fallbacks: Europe PMC fullTextXML → NCBI PMC OAI (JATS) → NCBI efetch (XML) → PMC HTML. Returns a machine-readable retrieval_trace so callers can see exactly what succeeded/failed.

Parameters:

  • fulltext_xml_url (string) (optional) Direct Europe PMC fullTextXML URL (recommended when you already have it from EuropePMC_search_articles).

  • pmcid (string) (optional) PMC ID (e.g., ‘PMC11237425’ or ‘11237425’). If provided, the tool derives the Europe PMC fullTextXML URL.

  • source_db (string) (optional) Europe PMC source database (e.g., ‘MED’ or ‘PMC’). Used with article_id to derive the fullTextXML URL.

  • article_id (string) (optional) Europe PMC article ID within source_db (e.g., PMID for source_db=’MED’).

  • terms (array) (optional) Terms to search for in the extracted full text (case-insensitive).

  • keywords (array) (optional) Alias for terms (case-insensitive). Provided for backwards/UX compatibility.

  • window_chars (integer) (optional) Context window size (characters) before and after each match.

  • max_snippets_per_term (integer) (optional) Maximum number of snippets returned per term.

  • max_total_chars (integer) (optional) Hard cap on total characters returned across all snippets.

Example Usage:

query = {
    "name": "EuropePMC_get_fulltext_snippets",
    "arguments": {
    }
}
result = tu.run(query)

EuropePMC_get_references (Type: EuropePMCRESTTool)#

Get one page of references (bibliography) for an article from Europe PMC. Returns exact hitCount-…

EuropePMC_get_references tool specification

Tool Information:

  • Name: EuropePMC_get_references

  • Type: EuropePMCRESTTool

  • Description: Get one page of references (bibliography) for an article from Europe PMC. Returns exact hitCount-derived pagination metadata (total_results, total_pages, has_more, and next_page), so callers can retrieve every cited article. Requires a source (e.g., ‘MED’ for PubMed, ‘PMC’ for PMC) and article ID.

Parameters:

  • source (string) (optional) Source database (e.g., ‘MED’ for PubMed, ‘PMC’ for PMC). Usually ‘MED’ for most articles.

  • article_id (string) (required) Article ID from the source database (e.g., PMID for MED source)

  • page_size (integer) (optional) Number of references to retrieve (default: 25, max: 1000)

  • page (integer) (optional) Page number for pagination (default: 1)

Example Usage:

query = {
    "name": "EuropePMC_get_references",
    "arguments": {
        "article_id": "example_value"
    }
}
result = tu.run(query)

EuropePMC_search_articles (Type: EuropePMCTool)#

Search Europe PMC (PubMed/MEDLINE plus PMC full text and preprints) for published biomedical and …

EuropePMC_search_articles tool specification

Tool Information:

  • Name: EuropePMC_search_articles

  • Type: EuropePMCTool

  • Description: Search Europe PMC (PubMed/MEDLINE plus PMC full text and preprints) for published biomedical and clinical literature, with abstracts and metadata. Use it for evidence on how a disease, injury or complication is treated, surgical and procedural techniques, trial results, reviews and outcomes; for case reports alone use EuropePMC_search_case_reports. Returns the top limit results ranked by relevance, which is normally a SLICE of the matches: metadata.total_results carries Europe PMC’s own hitCount (every article matching the query) while metadata.count is only how many were returned, and a top-level truncated: true plus truncation_note appear whenever matches were left behind. Do not read a small count as ‘few articles exist’ — check total_results. Europe PMC supports fielded queries for indexed full text (e.g., BODY:”term”) but only for records where Europe PMC has full text indexed (HAS_FT:Y). Use require_has_ft/fulltext_terms to avoid confusing ‘has a full-text link somewhere’ with ‘full text searchable in Europe PMC’.

Parameters:

  • query (string) (required) Search query for Europe PMC. Supports Lucene-like fielded syntax (e.g., BODY:”term”, INTRO:”term”, METHODS:”term”).

  • limit (integer) (optional) Number of articles to return. This sets the maximum number of articles retrieved from Europe PMC; anything beyond it is silently left behind, so compare it against metadata.total_results in the response (Europe PMC’s hitCount) before treating the returned set as complete. Europe PMC accepts up to 1000 per request.

  • require_has_ft (boolean) (optional) If true, appends HAS_FT:Y to the query to restrict results to records where Europe PMC has full text indexed. This is NOT the same as having a link to free full text elsewhere (e.g., PMC) and is often the key reason body-only keywords cannot be found via Europe PMC search.

  • fulltext_terms (array) (optional) Optional list of terms that must occur in the indexed full text (Europe PMC BODY field). When provided, the tool adds a BODY:”…” clause and automatically enables require_has_ft.

  • enrich_missing_abstract (boolean) (optional) If true, best-effort fills missing abstracts by fetching Europe PMC fullTextXML (bounded to a few results).

  • extract_terms_from_fulltext (array) (optional) Optional list of terms to extract from full text (open access only). When provided, automatically fetches fullTextXML for open-access articles and returns snippets around these terms. Terms are processed in batches of 5 internally, so any number of terms is accepted. Bounded to first 3 OA articles to avoid latency. Returns snippets in a ‘fulltext_snippets’ field for each article.

  • page_size (integer) (optional) Alias for limit — number of results to return

Example Usage:

query = {
    "name": "EuropePMC_search_articles",
    "arguments": {
        "query": "example_value"
    }
}
result = tu.run(query)

EuropePMC_search_case_reports (Type: EuropePMCCaseReportsTool)#

Find published clinical case reports about a disease, rare presentation, complication, injury or …

EuropePMC_search_case_reports tool specification

Tool Information:

  • Name: EuropePMC_search_case_reports

  • Type: EuropePMCCaseReportsTool

  • Description: Find published clinical case reports about a disease, rare presentation, complication, injury or treatment, with their abstracts, from Europe PMC (PubMed/MEDLINE plus PMC). Use it for unusual or rare cases that guidelines and review chapters do not cover: how a similar patient was diagnosed, which surgery or drug was used, and the outcome. Searches the title and abstract of articles typed as Case Reports; every word must appear (Europe PMC does not stem words; British spellings such as ‘tumour’ are also searched in their American form), so give the condition’s name in 2-5 words (e.g. ‘Meckel diverticulum torsion’, ‘anti-GFAP astrocytopathy’, ‘xanthogranulomatous prostatitis’) rather than the whole patient description, and leave out ages, sexes and words like ‘treatment’ or ‘management’. When no case report has every word, the search falls back to any of the words, ranked by relevance, and metadata.matching says so; check those reports against the case. Optional published_after / published_before (a year or YYYY-MM-DD) limit the publication date. Results are ranked by relevance; metadata.total_results is Europe PMC’s count of every matching case report.

Parameters:

  • query (string) (required) The condition, presentation or treatment in plain words (2-5 words work best), e.g. ‘invasive cervical resorption’ or ‘retrocaval ureter’. Quotes make a phrase; any other search syntax (fields, parentheses, AND/OR/NOT, -) is ignored.

  • limit (integer) (optional) Number of case reports to return (default 5). At most 25 are returned; larger values are treated as 25.

  • published_after (string) (optional) Only case reports first published on or after this date: a year (2015) or YYYY-MM-DD.

  • published_before (string) (optional) Only case reports first published on or before this date: a year (2023) or YYYY-MM-DD.

Example Usage:

query = {
    "name": "EuropePMC_search_case_reports",
    "arguments": {
        "query": "example_value"
    }
}
result = tu.run(query)