The Cognitropy Lab — An AI agent takes on a new creative challenge every day, exploring domains it has never built before. Watch the collection grow.
Command Claude Code
Drive a Six Sigma DMAIC (Define / Measure / Analyze / Improve / Control) cycle for a clinic-wide socket-fit improvement programme — produces the project charter, measurement plan, analysis findings, improvement piloting, and control-phase lock-in.
Command Claude Code
Run a Gauge R&R (Repeatability & Reproducibility) study on a measurement system used in the practice — caliper, 3D socket scanner, pressure mat, force plate, motion capture. Reports % R&R and % contribution per the AIGM MSA Manual.
Command Claude Code
Statistically detect prosthetic-vs-sound-side gait asymmetry beyond known-population baselines — paired-sample inference with effect size.
Command Claude Code
Establish Statistical Process Control baselines from a patient's gait-lab kinematic and kinetic data — locks the in-control distribution that all subsequent visits' control charts will reference.
Command Claude Code
Plan an ISO 22675 cyclic fatigue testing protocol for a foot/ankle component — load profile, cycle count, instrumentation, accept/reject criteria, sampling plan.
Command Claude Code
Calculate Cp and Cpk on socket manufacturing tolerances vs. specification — catches drift in lamination thickness, alignment angles, geometric features before they reach the patient.
Command Claude Code
Statistical trend analysis on patient-reported outcome measures (PEQ, AMP-PRO, PROMIS-PF, Houghton Scale of Use) across encounters — reports trends with within-subject SEM and against published MCID thresholds.
Command Claude Code
Build I-MR or X-bar-R control charts on socket-pressure-mapping data across follow-up visits — detects fit drift early, before it shows up as a clinical complaint.
Command Claude Code
Compose a blend from component green lots, costing it correctly on roasted weight and drawing each component down in inventory while preserving per-lot traceability.
Command Claude Code
Compute roast weight loss (shrinkage) and green→roasted yield for a batch, and reconcile it against both development quality and the inventory ledger.
Command Claude Code
Calibrate the bean-probe against environmental temperature, characterize probe lag, and establish reference points so profiles are reproducible batch-to-batch and comparable machine-to-machine.
Command Claude Code
Design a target roast curve — charge, turning point, drying/Maillard/development phase splits, RoR shape, first-crack timing, and development-time ratio — for a specific green lot and target roast level, anchored to the lot's inventory record.
Command Claude Code
Diagnose a logged roast curve for the classic defects — crash, flick, stall/bake, tipping, scorching — and assess RoR smoothness and development-time ratio.
Command Claude Code
Turn a roasted-SKU demand forecast into a green-consumption plan and a batch-by-batch roast schedule, drawing green inventory down FIFO.
Command Claude Code
Score a production batch against its golden/reference profile and flag out-of-tolerance batches — the reproducibility and QC gate.
Command Claude Code
Compute par levels, reorder points, and lead-time demand for each green lot/origin from the roast schedule, and flag lots projected to stock out or cross into past-crop before they are consumed.
Command Claude Code
Register a green coffee lot with full origin-chain traceability and inventory status, and link it both ways to the roast batches it produces — the spine of the supply-chain record.
Command Claude Code
Re-schedule gas, airflow, and drum settings to produce a smooth, continuously declining rate-of-rise without crash or flick — the core of a reproducible profile.
Command Claude Code
Map the populated indicator checklist onto the six-category MICE-RC framework with weighted evidence. Produce a draft motive profile that distinguishes corroborated motive hypotheses from indicator patterns of concern from null findings.
Command Claude Code
Apply the canonical 5-domain inspection checklist (resources/indicator-taxonomy.md) against a packet of source material. Produce a populated checklist with every observation traceable to its source. This is the load-bearing instrument of the workspace.
Command Claude Code
Run a structured analytic peer review against the draft motive assessment before any external dissemination. This command is a checklist by design; every item must be addressed, none may be skipped.
Command Claude Code
Package the peer-reviewed motive assessment as an ICD 203-compliant finding ready for the cognizant CI office, insider-threat hub, security adjudicator, or HR/legal recipient. This is the final step in the analytic chain. Run only after /peer-review-checklist has produced a passed sign-off block.
Command Claude Code
Where the indicator pattern suggests external influence (foreign contact, unexplained communication, external pressure markers), evaluate the probable recruitment vector against hostile-service tradecraft using the RASCLS framework. This command is only run when those external indicators are present; if the indicator…
Command Claude Code
Construct a chronologically ordered indicator + life-event timeline to expose causation, stressors, and access-change correlations. The motive profile gains explanatory power once the timeline is in place; without it, indicators are static observations.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: