Folklore Clinical Variant Interpretation MCP
Folklore MCP 1.5.0: read-only GRCh38 germline variant evidence, ClinGen gene-disease associations and biomedical literature.
Cite this software
Description
Folklore Clinical Variant Interpretation MCP
Folklore is Helena Bioinformatics' public, read-only Model Context Protocol (MCP) adapter for source-linked GRCh38 germline variant interpretation, ACMG/AMP decision support, ClinGen gene-disease evidence, and related biomedical literature discovery.
Agents can use the current scientific tool surface to:
- retrieve normalized public variant evidence with
search_variant_evidence; - discover variant-linked publications with
search_variant_literature; - retrieve source-linked publication records with
get_publication_details; - search the maintained literature corpus with
search_literature_corpus; - retrieve ClinGen gene-disease assertions by gene symbol or HGNC ID with
get_gene_disease_associations; - find genes for a disease name or MONDO ID with
search_disease_genes.
Release 1.5.0 provides these six scientific tools and also retains the separate, non-scientific support_helena information helper.
Public access
- MCP endpoint: https://api.helena.bio/folklore/v1/mcp
- Product site: https://folklore.helena.bio
- Integration guide: https://folklore.helena.bio/integrations
- Source code: https://github.com/helena-bioinformatics/folklore-mcp
- Official MCP Registry identity:
io.github.helena-bioinformatics/folklore
The public adapter is licensed under Apache-2.0 and does not require an account, API key, or OAuth.
Scope and safety
Variant queries support GRCh38 germline nuclear SNVs and simple indels shorter than 50 bp. Gene-disease queries accept gene/HGNC or disease/MONDO identifiers and preserve source assertion classifications, inheritance, report links and snapshot provenance. The public MCP does not ingest raw DNA, VCF uploads or batch analyses. The service accepts no patient, phenotype, family, segregation, private-case, credential, or uploaded private content. Results are evidence-discovery and decision-support outputs; they require qualified professional review and must not be treated as a diagnosis or treatment recommendation.