{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.02358", "kind": "paper", "url": "https://arxiv.org/abs/2608.02358"}, {"id": "gh:declare-lab/scrambletoolbench", "kind": "repo", "url": "https://github.com/declare-lab/ScrambleToolBench"}], "authors": ["Vernon Toh", "Navonil Majumder", "Zhengyuan Liu", "Nancy F. Chen", "Soujanya Poria"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.02358", "source_url": "https://arxiv.org/abs/2608.02358"}, "categories": ["Agents & Tool Use", "General AI", "Reasoning"], "description": {"en": "ScrambleToolBench evaluates agent behavioral reasoning in an interactive terminal environment with obfuscated tools and dynamic challenges like mapping drift and stochastic failures."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.02358", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.02358", "title": "ScrambleToolBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Agency for Science, Technology and Research", "Nanyang Technological University", "International Paper (United States)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.02358", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.02358", "source_url": "https://openalex.org/W7172445016"}, "key": "claire-radar:2608.02358", "modality": null, "name": "ScrambleToolBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_scrambletoolbench_aa15193b", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "34491d218eb05f421b4d3d96f2d613ad7136fce7bb733c7b9b8e6100da1e9861", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.02358"}, "publisher": null, "released": "2026-08-03", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.02358", "source_url": "https://arxiv.org/abs/2608.02358"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-02358", "source": "claire_radar", "source_benchmark_id": "2608.02358", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Agents & Tool Use", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/declare-lab/ScrambleToolBench", "githubScope": "benchmark_repo", "githubStars": 4, "hfDailySubmittedAt": "2026-08-04T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 11, "hfPaperUrl": "https://huggingface.co/papers/2608.02358", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": ["Reasoning"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "3e01134c43707be0ccc50816693ac98caa4b3846a76af61e086562584d3efb0f", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ScrambleToolBench evaluates agent behavioral reasoning in an interactive terminal environment with obfuscated tools and dynamic challenges like mapping drift and stochastic failures.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To address this limitation, we introduce ScrambleToolBench, an interactive terminal benchmark designed to isolate behavioral reasoning."}, "familyId": "bmf_2758dc7261ac", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/declare-lab/ScrambleToolBench", "url": "https://github.com/declare-lab/ScrambleToolBench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/declare-lab/ScrambleToolBench", "url": "https://huggingface.co/papers/2608.02358"}], "scope": "benchmark_repo", "url": "https://github.com/declare-lab/ScrambleToolBench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_scrambletoolbench_aa15193b", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/declare-lab/ScrambleToolBench", "data": null, "hfPaper": "https://huggingface.co/papers/2608.02358", "paper": "https://arxiv.org/abs/2608.02358", "pdf": "https://arxiv.org/pdf/2608.02358", "project": null, "report": "https://arxiv.org/abs/2608.02358"}, "metrics": [], "motivation": "To operate robustly in open-world environments, autonomous agents should be able to infer the behavior of unfamiliar systems through interaction alone, even in the absence of documentation.", "name": "ScrambleToolBench", "oneLine": "ScrambleToolBench evaluates agent behavioral reasoning in an interactive terminal environment with obfuscated tools and dynamic challenges like mapping drift and stochastic failures.", "paperTitle": "ScrambleToolBench: Agents Search Exhaustively Even When Their Own Map Points to the Next Step", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.02358", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.21483102985913266, "value": 4}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.4390696100905538, "value": 11}}, "confidence": "Medium", "coverage": 0.7, "rank": 209, "score": 33}, "method": "attention-ranking-v13", "rank": 209, "score": 33}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-03", "source": {"authors": ["Vernon Toh", "Navonil Majumder", "Zhengyuan Liu", "Nancy F. Chen", "Soujanya Poria"], "categories": ["cs.CL"], "id": "2608.02358", "title": "ScrambleToolBench: Agents Search Exhaustively Even When Their Own Map Points to the Next Step", "type": "arxiv", "url": "https://arxiv.org/abs/2608.02358"}, "sourceUpdatedAt": "2026-08-03T15:07:46Z", "topics": ["Reasoning"], "venueAttempts": [], "whyItMatters": "Isolates behavioral reasoning from prior knowledge, revealing gaps in agent adaptation and supporting comparison of agent architectures."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
