Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Germline ACMG/AMP classification, genes, disease associations, literature and API-key case analysis.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Works in Claude on the web, desktop and mobile:
https://api.helena.bio/folklore/v1/mcpRun this in your terminal
claude mcp add --transport http folklore https://api.helena.bio/folklore/v1/mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"folklore": {
"url": "https://api.helena.bio/folklore/v1/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"folklore": {
"type": "http",
"url": "https://api.helena.bio/folklore/v1/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
ChatGPT connects only to servers on the internet:
https://api.helena.bio/folklore/v1/mcpMenu names can differ slightly by plan and over time.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Tool names and descriptions exactly as the server reported them on Nov 7, 2026. The maker wrote them, not us; we show them as plain text.
classify_germline_variantClassify one GRCh38 germline nuclear SNV or simple indel smaller than 50 bp under ACMG/AMP. Accepts genomic, coding or protein HGVS, coordinates, SPDI, rsID or a returned canonical key. Returns normalized identity, automated classification, applied criteria, ClinVar assertions, p
search_variant_literatureResolve one GRCh38 germline variant and retrieve publications from the PubMed-derived genetics corpus. Exact variant mentions rank ahead of broader gene associations. Returns match types, source links, corpus provenance and retrieval time.
get_publication_detailsRetrieve one PMID from the PubMed-derived genetics corpus, including title, abstract, authors, journal, publication types, MeSH terms, gene and variant mentions, retraction status and PubMed/PMC links.
search_literature_corpusSearch the genetics Literature Corpus by question, PMID, DOI, PMCID, gene, variant, phenotype, HPO or OMIM concept. Known publication identifiers provide exact anchors. Returns ranked publications, match types, scores and source links.
support_helenaRetrieve Helena Good campaign, rules, manifest and MCP connection URLs.
get_gene_disease_associationsFind diseases associated with an exact gene symbol or HGNC identifier. Returns ClinGen gene-disease validity assertions, relation-specific inheritance, source reports and snapshot provenance, including conflicting and limited assertions.
search_disease_genesFind genes associated with an exact MONDO identifier or disease-name search. Returns ClinGen validity assertions with inheritance, source reports and snapshot provenance. Name searches can return multiple disease identities.
resolve_geneResolve a human gene symbol, previous symbol, alias, HGNC, Ensembl or NCBI Gene identifier. Returns exact, alias, ambiguous, not_found or withdrawn status with candidates and identity_snapshot_id. Use a selected HGNC gene_id and snapshot_id with get_gene_profile. Gene-plus-HGVS a
get_gene_profileRead the selected gene_id and immutable snapshot_id from resolve_gene. Returns overview, disease validity, dosage, mechanisms, transcripts, native assemblies, constraint, ClinVar summaries and consequences, expression, GWAS, phenotypes and protein/pathways with source assertions,
get_gene_sectionRead a gene-profile section using gene_id, snapshot_id and its returned opaque cursor. Returns source records, assembly/release, coverage, totals, count units and provenance. snapshot_unavailable requires a fresh resolution/profile request.
search_gene_literatureRetrieve a live publication capture for a resolved gene_id and snapshot_id. Returns source links, match labels, capture time/hash, degradation and gene/snapshot/sort-bound cursors. This capture is separate from the immutable reference snapshot and contains no abstracts.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026
The registry name was verified with the GitHub account helena-bioinformatics, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
Responded in 447 ms (HTTP 200)
No sign-in needed
Read 11 tool descriptions and found no hidden instructions to the AI
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 3 of 3 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.
The GitHub repo last changed –, and the latest version (2.0.0) reached the registry yesterday.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
In the same category: Spec Driven Development, Remote Desktop Commander, OpenTakeoff
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Construction takeoff for AI agents: load plans, set scale, measure, count, export with provenance.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
One hosted MCP server for public scientific queries and authenticated case workflows.
Endpoint: https://api.helena.bio/folklore/v1/mcp
Version 2.0.0
Connect your MCP client directly to this Streamable HTTP endpoint. No local Folklore server is required. Anonymous callers receive 11 public tools. Send your Folklore API key in the Authorization Bearer header to access 18 additional case, source and result tools at the same endpoint. Invalid supplied credentials fail authentication; they do not select the anonymous catalog.
Create a key in your Folklore account for the selected account and sources. Keep it in your client's secret store, never in prompts, tool arguments or URLs. The selected account's native subscription capabilities, usage limits and eligible credits govern processing. A key does not grant funded compute.
See case workflows, the 18 authenticated tool schemas and the public task recipes.
Connect your MCP client to:
https://api.helena.bio/folklore/v1/mcp
Access is anonymous. Call classify_germline_variant:
{"name":"classify_germline_variant","arguments":{"assembly":"GRCh38","query":"NM_007294.4:c.68_69del"}}
Input: one public GRCh38 nuclear germline SNV or simple indel smaller than 50 bp. Accepted notation includes supported HGVS, SPDI, rsID and genomic coordinates. Preserve transcript versions and reference alleles.
A resolved identity with successful interpretation includes automated ACMG/AMP classification, reported criteria, annotations, source evidence and provenance. Submitted ClinVar assertions retain their own source attribution alongside the Folklore classification. A VUS is a classification outcome.
| Status | Next step |
|---|---|
resolved | Read interpretation status, then classification and evidence. |
ambiguous | Present candidates for an explicit allele choice. |
not_found | Report the source-scoped miss. |
invalid_request | Supply a corrected supported expression. |
unsupported | Select a workflow for the requested variant type or assembly. |
resolution_unavailable | Inspect the typed error and retry policy. |
Interpretation status is success or unavailable. Transport, JSON-RPC, MCP and adapter errors have separate fields. See typed outcomes.
| Tool | Operation |
|---|---|
classify_germline_variant | Resolve, classify and assess one supported germline variant. |
search_variant_literature | Retrieve publications linked to a resolved variant. |
get_publication_details | Retrieve the bibliographic record and abstract for a PMID. |
search_literature_corpus | Search publications by scientific question or public identifier. |
get_gene_disease_associations | Retrieve ClinGen assertions for an exact gene symbol or HGNC identifier. |
search_disease_genes | Retrieve ClinGen assertions by MONDO identifier or disease-name substring. |
resolve_gene | Resolve a human gene symbol, alias or stable identifier. |
get_gene_profile | Read the selected HGNC gene at its returned snapshot. |
get_gene_section | Continue one profile section using its cursor. |
search_gene_literature | Retrieve a separate live publication capture for the selected gene. |
ClinGen results preserve assertion-level disease identity, inheritance, validity assessment, panel, date and source links. Profile sections retain provider, assembly, release, units and pagination. Gene literature has its own capture identity. Follow gene-disease and gene-profile contracts for exact inputs.
support_helena provides links to the separate Helena Good service when the user asks about supporting Helena. An optional MCP App renders variant results.
The hosted server uses stateless Streamable HTTP and MCP 2026-07-28. Requests carry matching protocol metadata and routing headers. Supported methods include server/discover, tools/list, tools/call, prompts/list, prompts/get, resources/list and resources/read.
The compatibility guide records protocol/client observations. All clients sharing egress use the same rate budget.
Install the portable Agent Skill using the installation guide. It routes pathogenicity, VUS, ClinVar, population-frequency, gene-disease and literature tasks to the hosted tools. The skill source includes OpenAI interface metadata.
Workflow prompts compose tool operations for classification, VUS assessment, ACMG/AMP explanation, identity verification and variant-literature comparison. Each accepts one public variant expression.
The Apache-2.0 adapter implements MCP transport, public API clients and validated response projection. Folklore's scientific authorities produce normalization, annotation, evidence aggregation and classification.
python3.12 -m venv .venv
. .venv/bin/activate pip install -e '.[dev]' FOLKLORE_MCP_ENABLED=true \ FOLKLORE_LITERATURE_ENABLED=true \ FOLKLORE_GENE_DISEASE_ENABLED=true \ FOLKLORE_PUBLIC_GENE_ENABLED=true \ folklore-mcp
Capabilities are enabled through the flags above. Gene literature requires both public-gene and literature capabilities. The default upstream is https://api.helena.bio; local contract tests support localhost and 127.0.0.1. The standalone HTTP image uses Dockerfile.adapter; Dockerfile provides the pinned stdio bridge.
pytest
ruff check . ruff format --check . python3 ops/reconcile_discovery.py
The read-only reconciler compares runtime, Server Card and Official Registry state, with separate aggregator observations. python3 ops/public_smoke.py checks discovery, prompts and resources.
Public measurement sources: variant interpretation, 100-case agent discovery, 60-query search discovery and genomic discovery cohort.
io.github.helena-bioinformatics/folkloreThe immutable MCP adapter 1.6.0 archive is 10.5281/zenodo.23817704, from signed tag folklore-mcp-v1.6.0 at 55a332c6dd6bd4650e921c12ac145b8e0f90476a. The Biorouter-tag archive is 10.5281/zenodo.23818207.
Apache License 2.0. See LICENSE and NOTICE.
Automated variant classifications support qualified professional review and are not patient diagnoses or treatment recommendations.