Tristan Farmer
Community@001tmf
Nature has time we have GPUs
Agent Skills by Tristan Farmer
Showing 19 vetted skills indexed across 1 GitHub repositories.
001TMFby-epitope-analysis
Identify and score epitope interface residues from co-crystal structures for binder design.
001TMFby-database
Queries PDB, UniProt, and SAbDab via MCP servers to produce structured CSV outputs for target characterization and scaffold selection.
001TMFby-research
Automate cross-database target research for antibody campaigns with an 8-phase pipeline.
001TMFby-knowledge
Store and query campaign knowledge in a persistent graph.
001TMFby-scoring
Compute and interpret BY ipSAE, ipTM, and composite scores from PAE matrices.
001TMFby-screening
Apply hard filters and composite ranking to protein designs, outputting PASS/MARGINAL/FAIL verdicts as CSV.
001TMFby-display
Render consistent BY terminal outputs using canonical display templates.
001TMFby-causal-reasoning
Generate ranked, evidence-grounded hypotheses from diagnostic features.
001TMFby-design-workflow
Route design intents to engines and generate routing and handoff JSON artifacts.
001TMFby-campaign-optimizer
Generates next-round YAML configs from scored protein designs using Random Forest regression.
001TMFby-session
Initialize BY sessions by configuring the environment and running the first-run questionnaire.
001TMFby-deploy-compute
Deploy BY design tools to local, RunPod, Modal, or HPC compute surfaces.
001TMFby-failure-diagnosis
Run Mann-Whitney U tests on PASS/FAIL designs to diagnose failure modes.
001TMFpxdesign
Generate de novo protein binders against target structures via CLI workflows.
001TMFby-hypothesis-debate
Coordinate adversarial multi-agent debate to select a protein/antibody design strategy.
001TMFboltzgen
Design antibody, nanobody, and de novo binder backbones conditioned on target structures.
001TMFprotenix
Predict protein, complex, and protein-ligand structures from sequence inputs with confidence metrics.
001TMFby-experiment-results
Ingest wet-lab readouts and calibrate in-silico predictions against lab outcomes.
001TMFby-campaign-manager
Coordinate protein design campaign planning, sizing, tracking, and resume-after-interruption.