Awesome Auto Research Hub Papers · Datasets · Projects
← Back to papers

Harnessing AtomisticSkills for Agentic Atomistic Research

arXiv 2026 60.4 method, system

TLDR

Introduces AtomisticSkills, an open-source framework enabling LLM agents to conduct atomistic research via modular, hierarchical skills.

Reasoning

Strengths include a modular, extensible framework with over 100 curated skills and validation across diverse scientific campaigns. Weaknesses are that the abstract lacks details on quantitative performance metrics or comparisons to baselines.

Read-first score

Read-first score 60.4, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 78.

Recency 8%
100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Methodology quality 25%
70

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=analysis,benchmark,validation

Topical relevance 42%
65

Uses existing LLM keyword relevance scores normalized to 0-100. AI scientist,automated scientific discovery,autonomous research agent,automated research,literature review agent,survey generation,automated experimentation,experiment design agent,AI for scientific research,paper writing agent,research automation,scientific discovery agent

Reproducibility 25%
30

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Field roles

FrontierBridgeMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 74.

Keyword Scores

automated scientific discovery
9
autonomous research agent
9
AI for scientific research
9
scientific discovery agent
9
automated research
8
automated experimentation
8
research automation
8
AI scientist
7
experiment design agent
7
literature review agent
3
survey generation
1
paper writing agent
0

Deep Analysis

Innovations

  • Hierarchical decomposition of scientific workflows into agent skills and tools for modular, extensible, and plug-and-play research capabilities
  • Integration of over 100 human-curated multidisciplinary skills spanning database access, thermodynamics, kinetics, MLIPs, and DFT
  • Open-source harness framework that empowers general-purpose AI coding agents to conduct atomistic research across materials science, chemistry, and drug discovery

Methodology

AtomisticSkills hierarchically decomposes scientific workflows into agent skills and tools, integrating more than 100 human-curated multidisciplinary skills (database access, thermodynamics, kinetics, MLIPs, DFT). It empowers general-purpose AI coding agents to perform atomistic research, with validation via functional coverage against literature and orchestration across diverse scientific campaigns.

Key Results

The framework demonstrated robust orchestration across six diverse campaigns: generative design of Li-ion solid-state electrolytes, high-throughput screening of MOFs for CO2 capture, autonomous MLIP benchmarking and fine-tuning, multi-stage structure-based virtual screening for drug design, multimodal XRD pattern analysis, and screening of Fe-oxide catalysts for OER, along with validated functional coverage against scientific literature.

Tags

chem-phmtrl-sciAIcomp-ph