Orchestrate closed-loop autonomous optimization of experimental conditions for materials synthesis.
npx skills add https://github.com/hkqai/MatClaw --skill active-learning
Please help me install this Agent Skill. Skill: active-learning Source: https://github.com/hkqai/MatClaw/tree/main/skills/active-learning Command: npx skills add https://github.com/hkqai/MatClaw --skill active-learning
Orchestrates closed-loop autonomous experimental optimization to accelerate materials synthesis and process tuning, removing manual trial-and-error and bridging experiments with computational models. It enables rapid discovery of synthesis conditions and process parameters by coordinating campaign setup, experiment suggestion, automated characterization, and iterative model updates.
Start an ARROWS campaign to synthesize BaTiO3 from BaCO3 and TiO2 across 600–900°C and request the first experiment suggestion.
High-intent search queries and answers about installing and using this skill.
Bayesian optimization for materials synthesis is automated by defining continuous, discrete, or categorical parameter spaces, collecting observations, and building GP surrogates to suggest experiments using EI, UCBe, or PI acquisition strategies. This closed-loop process tuning replaces manual trial-and-error.
Automated XRD analysis can be integrated into autonomous experimentation loops by using CNN-based phase identification and Rietveld refinement to analyze patterns, feeding phase and weight-fraction outputs directly into the campaign recorder to update model state.
Start an ARROWS campaign for solid-state synthesis by initializing thermodynamically-guided campaigns, enumerating balanced precursor sets, ranking them by driving force, and suggesting discrete experiments for specific synthesis targets across defined temperature ranges.
Combine thermodynamic guidance with surrogate-based process optimization using hybrid ARROWS and Bayesian optimization workflows, executing ARROWS→BO, BO→ARROWS, or parallel strategies to exploit thermodynamic guidance and fine-tune complex materials discovery problems.
Active learning for materials discovery supports mixed parameter types by allowing continuous, discrete, and categorical parameter spaces within Bayesian optimization loops, accommodating diverse experimental conditions for thin-film deposition or precursor screening.
Record experimental outcomes and extract reaction rules by feeding results into the campaign recorder, which learns pairwise reaction rules, updates campaign state files, and exports learned reactions and summaries for knowledge extraction.
stanislavjiricekGuide parameter exploration and model fitting for computational neuroscience.
metratonCoordinates metric-driven, iterative software artifact improvement cycles with automated escalation and logging.
learningmatter-mitRefine powder XRD patterns against CIF phases to quantify phase fractions and lattice parameters.
0xHoneyJarExecute autonomous sprint loops with PR lifecycle orchestration and safety guardrails.
