The bioinformatics build.
2,780 modules in build covering the daily work of molecular biology: alignment, variant calling, structure prediction, single-cell analysis, and more. Each module exposes a machine-readable interface for AI-agent operation and is verified against its reference implementation.
What the records hold
The catalog, by entry type.
Each entry type has its own record format, and every row links into the module catalog: external tools with their source repositories, native tools, trained models, visualizers, and pipelines.
The flagship, in three sizes
One aligner, three builds.
The STAR splice-aware aligner is demanding at full speed. Our build ships in three configurations: fast, standard, and a low-memory build that fits within 500 megabytes of RAM, for workstations and field environments with constrained resources.
All three are planned as free desktop builds and are verified against the original implementation's output.
Under the tools
The grounding layer.
A knowledge graph covering the human and plant genomes, and a graph neural network for genomics, drug discovery, and repurposing, sit beneath the agent layer and read alongside your data.
Reference access for researchers is planned: a limited free tier, with wider access by subscription.
Every bioinformatics entry, packaged for desktop systems and distributed through standard app stores.
Operate the corpus with Claude, Codex, or the DarwinOmics agents. Reference access to the knowledge graph ships in the same channel.
The platform installed within your institution; your data remains within your infrastructure.
The bioinformatics catalog is public.
2,780 modules in build, each with a public record page. Release proceeds module by module as verification completes.