{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.06624", "kind": "paper", "url": "https://arxiv.org/abs/2607.06624"}, {"id": "gh:agent-lens/agent-lens-bench", "kind": "repo", "url": "https://github.com/agent-lens/agent-lens-bench"}], "authors": ["Andrey Podivilov", "Vadim Lomshakov", "Sergey Savin", "Matvei Startsev", "Roman Pozharskiy", "Maksim Parshin", "Sergey Nikolenko"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.06624", "source_url": "https://arxiv.org/abs/2607.06624"}, "categories": ["Agents", "Agents & Tool Use", "General AI"], "description": {"en": "AgentLens evaluates interactive coding agents across entire trajectories, combining formal verification with LLM-written trajectory reviews and side-by-side comparisons to score dimensions such as instruction compliance, tool use, and interaction style."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.06624", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.06624", "title": "AgentLens"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2607.06624", "modality": null, "name": "AgentLens", "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_agentlens_07b9cd48", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "85fad6d1e0a14160c3cd3ce39d10e32d44c60dbe050750690db738c354b0503a", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.06624"}, "publisher": null, "released": "2026-07-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.06624", "source_url": "https://arxiv.org/abs/2607.06624"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-06624", "source": "claire_radar", "source_benchmark_id": "2607.06624", "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/agent-lens/agent-lens-bench", "githubScope": "benchmark_repo", "githubStars": 10, "hfDailySubmittedAt": "2026-07-09T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 10, "hfPaperUrl": "https://huggingface.co/papers/2607.06624", "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": [], "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": "8c31964b90f1f6e7fe19a4da453e193bd6736a323c42fc0e7a19f051516cc381", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "AgentLens evaluates interactive coding agents across entire trajectories, combining formal verification with LLM-written trajectory reviews and side-by-side comparisons to score dimensions such as instruction compliance, tool use, and interaction style.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We present AgentLens, a production-assessed benchmark for interactive code agents."}, "familyId": "bmf_95050afca81d", "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/agent-lens/agent-lens-bench", "url": "https://github.com/agent-lens/agent-lens-bench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/agent-lens/agent-lens-bench", "url": "https://huggingface.co/papers/2607.06624"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/agent-lens/agent-lens-bench", "url": "https://huggingface.co/api/papers/2607.06624"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/agent-lens/agent-lens-bench.", "url": "https://arxiv.org/abs/2607.06624"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/agent-lens/agent-lens-bench", "url": "https://arxiv.org/abs/2607.06624"}], "scope": "benchmark_repo", "url": "https://github.com/agent-lens/agent-lens-bench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_agentlens_07b9cd48", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/agent-lens/agent-lens-bench", "data": null, "hfPaper": "https://huggingface.co/papers/2607.06624", "paper": "https://arxiv.org/abs/2607.06624", "pdf": "https://arxiv.org/pdf/2607.06624", "project": null, "report": "https://arxiv.org/abs/2607.06624"}, "metrics": [], "motivation": "We present AgentLens, a production-assessed benchmark for interactive code agents.", "name": "AgentLens", "oneLine": "AgentLens evaluates interactive coding agents across entire trajectories, combining formal verification with LLM-written trajectory reviews and side-by-side comparisons to score dimensions such as instruction compliance, tool use, and interaction style.", "paperTitle": "AgentLens: Production-Assessed Trajectory Reviews for Coding Agent Evaluation", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.06624", "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": 0.32007591406276925, "value": 10}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.42369516881263364, "value": 10}}, "confidence": "Medium", "coverage": 0.7, "rank": 144, "score": 39}, "method": "attention-ranking-v13", "rank": 144, "score": 39}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-07", "source": {"authors": ["Andrey Podivilov", "Vadim Lomshakov", "Sergey Savin", "Matvei Startsev", "Roman Pozharskiy", "Maksim Parshin", "Sergey Nikolenko"], "categories": ["cs.AI", "cs.LG", "cs.SE"], "id": "2607.06624", "title": "AgentLens: Production-Assessed Trajectory Reviews for Coding Agent Evaluation", "type": "arxiv", "url": "https://arxiv.org/abs/2607.06624"}, "sourceUpdatedAt": "2026-07-14T09:13:17Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Most code-agent benchmarks reduce an episode to a single pass/fail bit, which is too coarse for production use. AgentLens provides a richer evaluation that captures real user experience and yields readable justifications for scores, aiding regression detection and model diagnosis."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
