{"paper":{"title":"MolClaw: An Autonomous Agent with Hierarchical Skills for Drug Molecule Evaluation, Screening, and Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A three-tier hierarchical skill architecture lets an AI agent reliably handle complex multi-step drug molecule workflows.","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Bowen Zhou, Haoran Sun, Haoyang Su, Lei Bai, Lilong Wang, Lisheng Zhang, Qikui Yang, Qingsong Li, Wei Tang, Wenjie Lou, Xiangyu Sun, Xiaosong Wang, Yankai Jiang, Yingnan Han, Yuehui Qian, Zhengwei Xie, Zhenyu Tang","submitted_at":"2026-04-02T09:27:36Z","abstract_excerpt":"Computational drug discovery, particularly the complex workflows of drug molecule screening and optimization, requires orchestrating dozens of specialized tools in multi-step workflows, yet current AI agents struggle to maintain robust performance and consistently underperform in these high-complexity scenarios. Here we present MolClaw, an autonomous agent that leads drug molecule evaluation, screening, and optimization. It unifies over 30 specialized domain resources through a three-tier hierarchical skill architecture (70 skills in total) that facilitates agent long-term interaction at runti"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"MolClaw achieves state-of-the-art performance across all metrics on MolBench, and ablation studies confirm that gains concentrate on tasks that demand structured workflows while vanishing on those solvable with ad hoc scripting.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the three-tier hierarchical skill architecture is the primary driver of improved performance rather than other factors such as prompt engineering, tool selection, or benchmark-specific tuning.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"MolClaw deploys a hierarchical skill system (tool, workflow, and discipline levels) to achieve state-of-the-art results on MolBench tasks requiring 8 to 50+ sequential tool calls in drug discovery.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A three-tier hierarchical skill architecture lets an AI agent reliably handle complex multi-step drug molecule workflows.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d1e12e66c4b986554c223458a4a9e5a3c80a0246a2dd2d347c13bca0394859f5"},"source":{"id":"2604.21937","kind":"arxiv","version":2},"verdict":{"id":"c2f5bb3f-0ab3-4d4e-a7b9-4041af74b949","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T21:23:01.068824Z","strongest_claim":"MolClaw achieves state-of-the-art performance across all metrics on MolBench, and ablation studies confirm that gains concentrate on tasks that demand structured workflows while vanishing on those solvable with ad hoc scripting.","one_line_summary":"MolClaw deploys a hierarchical skill system (tool, workflow, and discipline levels) to achieve state-of-the-art results on MolBench tasks requiring 8 to 50+ sequential tool calls in drug discovery.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the three-tier hierarchical skill architecture is the primary driver of improved performance rather than other factors such as prompt engineering, tool selection, or benchmark-specific tuning.","pith_extraction_headline":"A three-tier hierarchical skill architecture lets an AI agent reliably handle complex multi-step drug molecule workflows."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.21937/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":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}