{"record": {"aliases": ["AgentSLABench: Evaluating and Benchmarking Agentic Systems Under Resource Constraints"], "artifacts": [{"id": "arxiv:2608.00805", "kind": "paper", "url": "https://arxiv.org/abs/2608.00805"}, {"id": "gh:meherbhaskar/agentslabench", "kind": "repo", "url": "https://github.com/MeherBhaskar/agentslabench"}], "categories": ["Agents & Tool Use", "Software & AI Compute", "cs.AI"], "description": {"en": "A resource-aware agent benchmark with 16 Dockerized task environments that profiles correctness alongside latency, cost, compute, memory, and network use under declared budgets."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.00805", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.00805", "title": "AgentSLABench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.00805", "modality": null, "name": "AgentSLABench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-09-25T04:40:22Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_agentslabench-evaluating-and-benchmarking-_4029ac01", "origin_source": "arxiv", "recognition_confidence": "0.55", "record_sha256": "3de3a0e610b156a8a25959c940b567a49e473ddd4cca6e34c917e64c9df651ad", "relation": "evaluates_only", "review_model": "claude-code-default", "review_state": "ai-reviewed", "reviewed_at": "2026-09-24T20:55:04.887078Z", "source_url": "https://arxiv.org/abs/2608.00805"}, "publisher": null, "released": "2026-08-01", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.00805", "source_url": "https://arxiv.org/abs/2608.00805"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-00805", "source": "claire_radar", "source_benchmark_id": "2608.00805", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["AgentSLABench: Evaluating and Benchmarking Agentic Systems Under Resource Constraints"], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Agents & Tool Use", "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Agent efficiency evaluation is an active topic, but this is a short new paper from an individual repository.", "score": 45}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-24T20:55:04.887078Z", "inputHash": "15f9d0ce643d308bacf53af8e9dcc9b774695204eb4a4a9713482b40a6c78ea7", "model": "claude-code-default"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "AgentSLABench: Evaluating and Benchmarking Agentic Systems Under Resource Constraints", "canonicalNameSource": "paper_title", "decisionReason": "Named benchmark with defined task environments, a concrete EASR scoring metric, sealed hashed test sets, and a public code link.", "evaluationEvidence": {"reuse": "Code, sealed test sets, and profiling artifacts available at: https://github.com/MeherBhaskar/agentslabench", "scoring": "we report the Efficiency-Adjusted Success Rate (EASR) - success weighted by resource consumption relative to declared budgets", "task": "AgentSLABench provides 16 task environments across 6 categories"}, "model": "claude-code-default", "reviewedAt": "2026-09-24T20:55:04.887078Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "A resource-aware agent benchmark with 16 Dockerized task environments that profiles correctness alongside latency, cost, compute, memory, and network use under declared budgets.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "evaluation protocol evidence", "public artifact URL", "named entity is a model, method, framework, or dataset used in evaluation"], "snippet": "We present AgentSLABench, a resource-aware evaluation framework for autonomous AI agents that measures correctness alongside latency, cost, compute, memory, and network usage under declared resource budgets."}, "familyId": "bmf_7c9fa77887fd", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_agentslabench-evaluating-and-benchmarking-_4029ac01", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/MeherBhaskar/agentslabench", "data": null, "paper": "https://arxiv.org/abs/2608.00805", "pdf": "https://arxiv.org/pdf/2608.00805", "project": null, "report": "https://arxiv.org/abs/2608.00805"}, "metrics": [], "motivation": "We present AgentSLABench, a resource-aware evaluation framework for autonomous AI agents that measures correctness alongside latency, cost, compute, memory, and network usage under declared resource budgets.", "name": "AgentSLABench", "oneLine": "A resource-aware agent benchmark with 16 Dockerized task environments that profiles correctness alongside latency, cost, compute, memory, and network use under declared budgets.", "paperTitle": "AgentSLABench: Evaluating and Benchmarking Agentic Systems Under Resource Constraints", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.00805", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "readiness": "Runnable", "recognitionConfidence": 0.55, "relation": "evaluates_only", "releasedAt": "2026-08-01", "reviewLifecycle": {"state": "pending"}, "source": {"authors": ["Meher Bhaskar Madiraju and Meher Sai Preetam Madiraju"], "categories": ["cs.AI"], "id": "2608.00805", "title": "AgentSLABench: Evaluating and Benchmarking Agentic Systems Under Resource Constraints", "type": "arxiv", "url": "https://arxiv.org/abs/2608.00805"}, "sourceUpdatedAt": "2026-08-01T18:17:47Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "It pairs task success with resource consumption through a single efficiency-adjusted metric and sealed test sets. This lets agents be compared on production viability, not accuracy alone."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
