{"paper":{"title":"MatClaw: An Autonomous Code-First LLM Agent for End-to-End Materials Exploration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"An LLM agent autonomously handles end-to-end materials simulations by writing and executing its own code.","cross_cats":["cs.SE"],"primary_cat":"cond-mat.mtrl-sci","authors_text":"Boris I. Yakobson, Chenmu Zhang","submitted_at":"2026-04-03T03:32:15Z","abstract_excerpt":"Existing LLM agents for computational materials science are constrained by pipeline-bounded architectures tied to specific simulation codes and by dependence on manually written tool functions that grow with task scope. We present MatClaw, a code-first agent that writes and executes Python directly, composing any installed domain library to orchestrate multi-code workflows on remote HPC clusters without predefined tool functions. To sustain coherent execution across multi-day workflows, MatClaw uses a four-layer memory architecture that prevents progressive context loss, and retrieval-augmente"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our results demonstrate that the gap between guided and fully autonomous computational materials research is narrower than ever before: LLMs already handle code generation and scientific interpretation reliably, and the rapid improvement in their capabilities will accelerate materials discovery beyond what manual workflows can achieve.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the tacit domain knowledge gaps (simulation timescales, equilibration protocols, sampling strategies) can be consistently bridged by literature self-learning and expert-specified constraints without introducing systematic errors or requiring extensive ongoing human oversight.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"MatClaw is a code-first LLM agent that autonomously executes end-to-end materials workflows by generating and running Python scripts on remote clusters, achieving reliable code generation via memory architecture and RAG while requiring guided interventions for tacit knowledge.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"An LLM agent autonomously handles end-to-end materials simulations by writing and executing its own code.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c87d9f093a7b5ff3773f46f84f9863c250f96afe67a6bd39a99d800e9c5376d8"},"source":{"id":"2604.02688","kind":"arxiv","version":3},"verdict":{"id":"12603f99-00e8-4f8a-8ada-ddeff33bddd4","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T18:58:38.223961Z","strongest_claim":"Our results demonstrate that the gap between guided and fully autonomous computational materials research is narrower than ever before: LLMs already handle code generation and scientific interpretation reliably, and the rapid improvement in their capabilities will accelerate materials discovery beyond what manual workflows can achieve.","one_line_summary":"MatClaw is a code-first LLM agent that autonomously executes end-to-end materials workflows by generating and running Python scripts on remote clusters, achieving reliable code generation via memory architecture and RAG while requiring guided interventions for tacit knowledge.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the tacit domain knowledge gaps (simulation timescales, equilibration protocols, sampling strategies) can be consistently bridged by literature self-learning and expert-specified constraints without introducing systematic errors or requiring extensive ongoing human oversight.","pith_extraction_headline":"An LLM agent autonomously handles end-to-end materials simulations by writing and executing its own code."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.02688/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"266d830691709531415121b10072912a3c1df0c75e8e780509e311fc45f438dd"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}