{"record": {"aliases": [], "artifacts": [{"id": "gh:bluebunnyanon/agent-benchmark-2d-maze", "kind": "repo", "url": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze"}], "categories": ["Agents", "Agents & Tool Use", "Multimodal", "Software & AI Compute"], "description": {"en": "Evaluates interactive multimodal agents in structured 2D maze environments requiring long-horizon planning, mechanism interaction, and error recovery, scored by mechanism-aware progress."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://github.com/bluebunnyanon/agent-benchmark-2d-maze", "source": "claire_radar", "source_url": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze", "title": "Agentic Evaluation in 2D Mazes"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:github:bluebunnyanon/agent-benchmark-2d-maze", "modality": null, "name": "Agentic Evaluation in 2D Mazes", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-candidate", "display_eligible": "false", "origin_record_id": "bm_agent-benchmark-2d-maze_4d511c00", "origin_source": "github", "recognition_confidence": "0.4", "record_sha256": "6e7d193f7bc0f6ce5303dd9eb341de54e31b64afe11a60ab46a61bbbbee31ff9", "relation": "unclear", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-08-25T23:28:56.873215Z", "source_url": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze"}, "publisher": null, "released": "2026-08-22", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:github:bluebunnyanon/agent-benchmark-2d-maze", "source_url": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-github-bluebunnyanon-agent-benchmark-2d-maze", "source": "claire_radar", "source_benchmark_id": "github:bluebunnyanon/agent-benchmark-2d-maze", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Agents & Tool Use", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": null, "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": {"state": "not_applicable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "The benchmark addresses a growing area in agent evaluation, but its anonymous nature and niche 2D maze focus may limit early visibility.", "score": 45}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-25T23:28:56.873215Z", "inputHash": "0ded56072cac6d4ce03a7c2e9308e8a11650b04b56b4c9c513b6347e59ac0e37", "model": "deepseek-v4-pro"}, "curation": {"decisionReason": "The formal benchmark name is source-grounded, but the public release does not yet meet the independent evidence or adoption threshold.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-25T23:28:56.873215Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-candidate", "demo": false, "description": "Evaluates interactive multimodal agents in structured 2D maze environments requiring long-horizon planning, mechanism interaction, and error recovery, scored by mechanism-aware progress.", "discoverySources": [{"id": "github:bluebunnyanon/agent-benchmark-2d-maze", "type": "github", "url": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze"}], "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["discovered via github", "benchmark term in abstract", "evaluation protocol evidence", "no explicit benchmark release evidence"], "snippet": "agent-benchmark-2d-maze # Agentic Evaluation in 2D Mazes This repository contains the code, benchmark instances, evaluation harness, and reproducibility artifacts accompanying an anonymous submission on evaluating interactive multimodal agents in structured 2D maze environments."}, "familyId": "bmf_d82de6e84a4d", "firstSeenAt": "2026-08-23", "heat": null, "id": "bm_agent-benchmark-2d-maze_4d511c00", "indexedAt": "2026-08-23T05:46:57.857309Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze"}, "metrics": [], "motivation": "agent-benchmark-2d-maze # Agentic Evaluation in 2D Mazes This repository contains the code, benchmark instances, evaluation harness, and reproducibility artifacts accompanying an anonymous submission on evaluating interactive multimodal agents in structured 2D maze environments.", "name": "Agentic Evaluation in 2D Mazes", "oneLine": "Evaluates interactive multimodal agents in structured 2D maze environments requiring long-horizon planning, mechanism interaction, and error recovery, scored by mechanism-aware progress.", "paperTitle": "agent-benchmark-2d-maze", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-23T05:46:57.857309Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.55, "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": 0.0, "value": 0}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 506, "score": 23}, "method": "attention-ranking-v13", "rank": 506, "score": 23}}, "readiness": "Runnable", "recognitionConfidence": 0.4, "relation": "unclear", "releasedAt": "2026-08-22", "source": {"authors": ["bluebunnyanon"], "categories": [], "id": "github:bluebunnyanon/agent-benchmark-2d-maze", "title": "agent-benchmark-2d-maze", "type": "github", "url": "https://github.com/bluebunnyanon/agent-benchmark-2d-maze"}, "sourceUpdatedAt": "2026-08-22T20:14:45Z", "topics": ["Agents", "Multimodal"], "venueAttempts": [], "whyItMatters": "Provides a controlled testbed for agent capabilities beyond single-step tasks, with a human-playable interface and modular harness for model comparison."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
