BioContext7

BioContext7 unifies 47,000+ tools from bio.tools, nf-core, and Bioconda into a single searchable index with EDAM semantic annotations. It generates production-ready pipelines in Nextflow, Snakemake, and WDL from natural language descriptions, with LSP-validated self-healing loops that automatically fix errors.

How It Works

User: "Create an RNA-seq pipeline"
         |
BioContext7:
1. Queries registries (bio.tools, nf-core, Bioconda)
2. Resolves EDAM ontology → tool chain
3. Compiles to Nextflow/Snakemake/WDL
4. Validates with LSP → auto-fixes errors
5. Runs stub tests → verifies execution
         |
Output: Production-ready pipeline

Key Features

• 47k+ tools from bio.tools with EDAM semantic search
• Self-healing pipelines — LSP validation + automatic error correction
• Multi-target compilation — Nextflow DSL2, Snakemake, WDL
• 5 MCP tools — resolve-library-id, get-library-docs, find-skills, recommend-tools, report-snippet-quality
• find-skills — single-call semantic search with bc7score quality scoring, install commands, health signals, compact mode, and platform filtering
• Pipeline composition — chain tools by input/output compatibility
• GA4GH Standards — Beacon, VRS, Phenopackets, WES integration
• Protein databases — UniProt sequence and annotation retrieval
• Metabolomics — HMDB, Metabolomics Workbench, MassBank, LIPID MAPS

Registry Coverage