Dataverse Tools#

Configuration File: dataverse_tools.json Tool Type: Local Tools Count: 2

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

Available Tools#

Dataverse_get_dataset (Type: BaseRESTTool)#

Get detailed metadata for a specific Harvard Dataverse dataset by its persistent DOI identifier. …

Dataverse_get_dataset tool specification

Tool Information:

  • Name: Dataverse_get_dataset

  • Type: BaseRESTTool

  • Description: Get detailed metadata for a specific Harvard Dataverse dataset by its persistent DOI identifier. Returns comprehensive information including dataset ID, version details, citation metadata (title, authors, description, subjects, keywords), terms of use, and publication dates. The persistent ID is the DOI from search results (e.g., ‘doi:10.7910/DVN/DUWBBU’). Use Dataverse_search_datasets first to find the global_id, then pass it here as persistentId.

Parameters:

  • persistentId (string) (required) Persistent DOI identifier for the dataset (e.g., ‘doi:10.7910/DVN/DUWBBU’, ‘doi:10.7910/DVN/A6RYXI’). Found in global_id field of search results.

Example Usage:

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

Dataverse_search_datasets (Type: BaseRESTTool)#

Search Harvard Dataverse for datasets, files, or dataverses. Returns one result page with an exac…

Dataverse_search_datasets tool specification

Tool Information:

  • Name: Dataverse_search_datasets

  • Type: BaseRESTTool

  • Description: Search Harvard Dataverse for datasets, files, or dataverses. Returns one result page with an exact total and offset continuation metadata. Follow metadata.next_start while metadata.has_more is true. If Harvard’s AWS WAF challenges the API client, the tool returns an explicit error instead of a false empty success.

Parameters:

  • q (string) (required) Search query string for dataset title, description, or keyword (e.g., ‘climate change’, ‘genomics’, ‘RNA-seq cancer’)

  • type (string) (optional) Filter by type: ‘dataset’, ‘file’, or ‘dataverse’ (default: returns all types)

  • per_page (integer) (optional) Number of results per page (1-1000; default 10). Dataverse silently replaces 0 with 10, so the wrapper rejects values below 1.

  • start (integer) (optional) Zero-based result offset. Use metadata.next_start rather than assuming every page is full.

Example Usage:

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