{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.20950", "kind": "paper", "url": "https://arxiv.org/abs/2606.20950"}, {"id": "gh:trashchenkov/power-systems-agent-benchmark", "kind": "repo", "url": "https://github.com/trashchenkov/power-systems-agent-benchmark"}], "authors": ["Sergei Trashchenkov"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.20950", "source_url": "https://arxiv.org/abs/2606.20950"}, "categories": ["Agents", "General AI", "Language & Knowledge"], "description": {"en": "Power Systems Agent Benchmark is an executable benchmark for power-engineering agents. It includes 41 task families with deterministic evaluators that recompute engineering quantities and check constraints, plus held-out generation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.20950", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.20950", "title": "Power Systems Agent Benchmark"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.20950", "modality": null, "name": "Power Systems Agent Benchmark", "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_power-systems-agent-benchmark_68561f3f", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "1917fc0d1791e7327aa42c291d6fa706a18265cfb3dc20fa34dbd3131b2d2605", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.20950"}, "publisher": null, "released": "2026-06-18", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.20950", "source_url": "https://arxiv.org/abs/2606.20950"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-20950", "source": "claire_radar", "source_benchmark_id": "2606.20950", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/trashchenkov/power-systems-agent-benchmark", "githubScope": "benchmark_repo", "githubStars": 3, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2606.20950", "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": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Interactive Environment", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "b76a1ac6a6646826cbebd5ce468edd53af9f5493b023d78661b8b1792ffa1be2", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Power Systems Agent Benchmark is an executable benchmark for power-engineering agents. It includes 41 task families with deterministic evaluators that recompute engineering quantities and check constraints, plus held-out generation.", "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", "public artifact URL"], "snippet": "We introduce the Power Systems Agent Benchmark, an executable benchmark for power-engineering agents."}, "familyId": "bmf_9170ad13958e", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 1, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/trashchenkov/power-systems-agent-benchmark", "url": "https://github.com/trashchenkov/power-systems-agent-benchmark"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/trashchenkov/power-systems-agent-benchmark", "url": "https://huggingface.co/papers/2606.20950"}, {"basis": "linked-from-project", "linkedUrl": "https://github.com/trashchenkov/power-systems-agent-benchmark", "url": "https://doi.org/10.5281/zenodo.20753046"}, {"basis": "linked-from-project", "linkedUrl": "https://github.com/trashchenkov/power-systems-agent-benchmark/tree/v0.1-paper-artifacts", "url": "https://doi.org/10.5281/zenodo.20753046"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/trashchenkov/power-systems-agent-benchmark", "url": "https://arxiv.org/abs/2606.20950"}], "scope": "benchmark_repo", "url": "https://github.com/trashchenkov/power-systems-agent-benchmark", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_power-systems-agent-benchmark_68561f3f", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/trashchenkov/power-systems-agent-benchmark", "data": null, "hfPaper": "https://huggingface.co/papers/2606.20950", "paper": "https://arxiv.org/abs/2606.20950", "pdf": "https://arxiv.org/pdf/2606.20950", "project": "https://doi.org/10.5281/zenodo.20753046", "report": "https://arxiv.org/abs/2606.20950"}, "metrics": [], "motivation": "Executable evaluation -- checking the consequences of an agent's actions with a program rather than grading its prose -- has become a prominent way to assess tool-using AI agents in software settings.", "name": "Power Systems Agent Benchmark", "oneLine": "Power Systems Agent Benchmark is an executable benchmark for power-engineering agents. It includes 41 task families with deterministic evaluators that recompute engineering quantities and check constraints, plus held-out generation.", "paperTitle": "Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.20950", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Power Systems Agent Benchmark Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/trashchenkov/power-systems-agent-benchmark"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-18", "source": {"authors": ["Sergei Trashchenkov"], "categories": ["cs.AI", "cs.SY", "eess.SY"], "id": "2606.20950", "title": "Power Systems Agent Benchmark: Executable Evaluation of AI Agents in Electric Power Engineering", "type": "arxiv", "url": "https://arxiv.org/abs/2606.20950"}, "sourceUpdatedAt": "2026-07-02T14:10:13Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Executable evaluation in power engineering is missing. This benchmark provides a repeatable, contamination-resistant protocol for tool-using agents, with public and hidden splits."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
