Blatant-Why plugin
A Claude Code agent for protein design: target research, antibody and nanobody generation, screening and candidate ranking.
- 119 ★ on the repository
- Last commit
Extended access
Why this level
- The agent runs costly GPU and cloud compute with your keys.
- A lab order costs money even though it is triple-gated.
Install
- Works in
Manual install
npx blatant-why initThe command from step 2 of the README Quick Start, run in a new project directory.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The agent runs costly GPU and cloud compute with your keys.
- A lab order costs money even though it is triple-gated.
This is third-party code. Review the repository files before installing.
What it does
Blatant-Why sets up a protein design campaign kit in a project: 11 MCP servers, 21 agents, 19 skills and 13 slash commands plus a `CLAUDE.md` orchestration file. The agent researches a target across PDB, UniProt and SAbDab, plans a campaign, runs generation on a local GPU, RunPod, Modal, SLURM or Tamarind Bio cloud, and screens designs for structural quality, sequence liabilities and developability. Candidates are ranked by a composite score. When lab CSVs come back the agent ingests them, compares them with predictions and feeds the calibration into the next round. Skills cover BoltzGen, Protenix and PXDesign.
Who it is for. Computational biology researchers planning antibody and nanobody design campaigns who can run jobs on their own GPUs or in the cloud.
Good fit when
- You need to design VHH nanobodies against PD-L1 and compare candidates.
- A campaign calls for target research in PDB and UniProt before any compute runs.
- Lab results arrived as CSV and must be compared with predictions.
- Compute runs on HPC or cloud while campaign state is kept in the project.
Not a fit when
- You have no GPU, cloud provider or Tamarind keys.
- You need a ready commercial platform with support and guarantees.
- You need a clinically validated result: the project is in beta 0.1.0.
Example request
Design VHH nanobodies against PD-L1, start with target research and show the campaign plan before running compute.Limitations
The project is beta 0.1.0. Node.js 18 or newer, Python 3.11 or newer, `uv` and Claude Code are required, and compute runs on a GPU or with paid providers whose keys go into `.env`. Submitting orders to the Adaptyv Bio lab is triple-gated and costs money. In silico scores serve selection and do not cancel lab validation; a specialist should review the biological conclusions of the agent. The time-saving chart in the README is not backed by measurements. Last commit October 2026.
How to disable. Delete the scaffolded project directory, stop GPU and cloud jobs and revoke provider keys.
FAQ
What does `npx blatant-why init` create?
A project scaffold: 11 MCP servers, 21 agents, 19 skills, 13 slash commands and `CLAUDE.md`. It takes about 30 seconds.
Where does compute run?
On a local GPU by default when one exists. Otherwise the first-run questionnaire offers RunPod, Modal, SLURM or Tamarind cloud.
How do I start?
Run `claude` and describe the task, for example nanobodies against PD-L1. `/by:plan-campaign` gives a guided flow and `/by:welcome` suits a first run.
Can an order reach the lab by accident?
The `by-lab` agent for Adaptyv Bio is triple-gated. Orders are paid, so review the candidate list before confirming.
Which databases does the research stage use?
PDB, UniProt and SAbDab. Separate MCP servers `pdb`, `uniprot` and `sabdab` handle queries, while the `by-research` agent covers literature and target analysis.
Discussion
No questions or reviews yet.
Related
PluginsA Claude Code plugin and MCP server: over 1000 scientific tools for biology, chemistry and medicine, data-check commands and routing skills
MCP server that connects Claude Code and Codex to an open R or Python session in RStudio and Positron, where your data lives
Around 200 skills for Claude Code on bioinformatics, chemistry and biostatistics, from RNA-seq analysis to molecular docking
Life Sciences: Anthropic's marketplace for Claude Code
Life Sciences Marketplace
Anthropic's official plugin marketplace for biology and medicine: PubMed, BioRender, Synapse, ChEMBL, ClinicalTrials.gov and data analysis skills
Monte Carlo plugin for Claude Code and Cursor: table checks before queries, data incident triage and monitor creation.
AWS hcls-agents plugin gives Claude Code and Codex skills for genomics, clinical trials and biomarkers plus MCP servers.
README in short
The whole pipeline runs inside Claude Code with no platform or dashboard: open-source tools are used directly and compute runs on your own resources.