Analytics
fairdata.ai
Evaluate research datasets for FAIR compliance and AI-readiness using FAIRdata.ai.
ENDPOINT 1
https://fairdata.ai/api/mcp
MCP server metadata
- Name
- fairdata-ai
- Version
- 1.0.0
FAIRdata.ai describes public research datasets for machine use. Start with `get_ruleset` to learn what 'Machine-Ready' means here, then `search_collections` to find a themed collection, `explain_collection_qualification` to see the evidence behind it, and `get_collection_manifest` to hand a pinned, versioned list to a workflow. Qualification is an automated evidence check on metadata, licence signals and file access. It is not permission to train, not a peer review, and not a privacy or consent assessment — use `check_usage_constraints` before advising anyone that data may be used, and report the unknowns it returns.
Known tools 24
get_rulesetThe qualification ruleset in force: every criterion, its threshold, the evidence type behind it, what an unknown value means, and the limits of the verdict.
Inferred read-onlyget_collectionOne collection in full: what it is, what defines membership, its aggregates, its licence position, its freshness, and where to fetch it in each available format.
Inferred read-onlycompare_collectionsCompare two to five collections on the dimensions that decide whether one is usable for a task: size, licence clarity and mix, FAIR and curation medians, file accessibility, assessment freshness, and modality coverage.
Inferred read-onlylist_collection_datasetsThe member datasets of a collection, with licence, scores, file counts and links.
Inferred read-onlyget_collection_manifestThe versioned manifest for a collection — the artefact to pin a workflow to.
Inferred read-onlyexplain_collection_qualificationThe full criteria matrix for a collection: each criterion, its threshold, the observed value, pass/fail/unknown, and the evidence with its evaluator and version.
Inferred read-onlysearch_datasets_in_corpusKeyword search over the datasets that clear the Machine-Ready gate.
Inferred read-onlyget_datasetOne dataset from the qualifying corpus: metadata, licence, scores, files, the collections it belongs to, and its qualification result.
Inferred read-onlyexplain_dataset_qualificationPer-criterion evidence for one dataset: threshold, observed value, status, evidence origin, evaluator and version.
Inferred read-onlycreate_ingestion_planA concrete, ordered plan for loading a collection: which access level applies (metadata only, file manifest, or prepared records), how much data is involved, what to verify, and runnable code for the requested language.
Inferred read-onlycreate_reproducibility_receiptProduce a fairdata.lock.json recording exactly which collection version and datasets a workflow consumed, with the manifest checksum needed to detect later change.
Inferred read-onlycheck_usage_constraintsWhat is and is not known about permission to reuse a dataset or collection.
Inferred read-onlyget_recordThe full FAIRdata.ai enriched record for a DOI: FAIR scores, enrichments with provenance, extracted paper context and links to every serialisation.
Inferred read-onlyget_record_formatA record serialised as schema.org, Croissant, RO-Crate, FAIR Data Point DCAT3, enriched DataCite, or the GDS AI-readiness assessment.
Inferred read-onlylist_recordsBrowse the FAIRdata.ai registry — every assessed dataset, not only the qualifying subset.
Inferred read-onlyget_ai_readinessThe GDS/DSIT 4-pillar AI-readiness assessment for a dataset: per-pillar scores and individual check results.
Inferred read-onlyget_filesDirect, downloadable files for a dataset: URL, size, media type and any published checksum.
Inferred read-onlyget_data_qualityA schema-level quality grade for a dataset's tabular files: completeness, structure and size.
Inferred read-onlyfind_similar_datasetsDatasets most similar to a DOI by cosine similarity over pre-computed text embeddings.
Inferred read-onlysearch_datasetsMeaning-based search across the whole assessed registry, including datasets that do not clear the Machine-Ready gate.
Inferred read-onlyget_linked_papersPublications linked to a dataset — papers that cite, supplement or document it.
Inferred read-onlyassess_datasetRun the full FAIR and AI-readiness assessment pipeline for a DOI that is not yet in the registry.
Inferred read-onlyfind_training_dataDeprecated: use search_collections and get_collection instead, which report the same sets with their licence composition and evidence.
Inferred read-onlyCONNECT WITH APPROVAL
Client installation
Review this server and its permissions before adding it. Secret placeholders must be set locally.
Codex
~/.codex/config.toml
[mcp_servers.fairdata-ai]
url = "https://fairdata.ai/api/mcp"
enabled = trueClaude Code
.mcp.json
{
"mcpServers": {
"fairdata-ai": {
"type": "http",
"url": "https://fairdata.ai/api/mcp"
}
}
}Claude Desktop
Settings → Connectors → Add custom connector
Name: fairdata-ai
Remote MCP URL: https://fairdata.ai/api/mcpAdd this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.
Cursor
.cursor/mcp.json
{
"mcpServers": {
"fairdata-ai": {
"url": "https://fairdata.ai/api/mcp"
}
}
}Visual Studio Code
.vscode/mcp.json
Add to Visual Studio Code{
"servers": {
"fairdata-ai": {
"type": "http",
"url": "https://fairdata.ai/api/mcp"
}
}
}Generic MCP
Client-specific MCP configuration
{
"name": "fairdata-ai",
"transport": "streamable-http",
"url": "https://fairdata.ai/api/mcp"
}MCP Inspector
Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.
TRUST AND VERIFICATION EVIDENCE
Trust Data Available
BuiltWith Trust API v2 evidence for fairdata.ai was fetched 2026-09-18T09:22:35.692Z.
fairdata.ai is assessed as Trusted: Domain runs a meaningful technology spend, consistent with a real business.
Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.