{"record": {"aliases": ["agent-systems-benchmark"], "artifacts": [{"id": "gh:martin-beck/agent-systems-benchmark", "kind": "repo", "url": "https://github.com/martin-beck/agent-systems-benchmark"}], "categories": ["Agents", "Language & Knowledge", "Software & AI Compute"], "description": {"en": "A Linux terminal framework for measuring how AI coding agents scale under concurrent sessions while maintaining task quality, latency, and resource consumption within declared bounds."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://github.com/martin-beck/agent-systems-benchmark", "source": "claire_radar", "source_url": "https://github.com/martin-beck/agent-systems-benchmark", "title": "Agent Systems Benchmark (ASB)"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:github:martin-beck/agent-systems-benchmark", "modality": null, "name": "Agent Systems Benchmark (ASB)", "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": "2026-10-14T09:44:51Z", "data_status": "primary-source-candidate", "display_eligible": "false", "origin_record_id": "bm_agent-systems-benchmark_6fb06572", "origin_source": "github", "recognition_confidence": "0.4", "record_sha256": "a5abf54cfc083b479bb2feead7a3690a8c8fa5e3c3fe3bb30174fec128ab391f", "relation": "unclear", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-09-09T01:08:06.842497Z", "source_url": "https://github.com/martin-beck/agent-systems-benchmark"}, "publisher": null, "released": "2026-09-06", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:github:martin-beck/agent-systems-benchmark", "source_url": "https://github.com/martin-beck/agent-systems-benchmark"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-github-martin-beck-agent-systems-benchmark", "source": "claire_radar", "source_benchmark_id": "github:martin-beck/agent-systems-benchmark", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["agent-systems-benchmark"], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/martin-beck/agent-systems-benchmark", "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 is in early development with narrow focus on agent scalability in Linux environments and limited initial agent support, reducing immediate attention.", "score": 5}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-09T01:08:06.842497Z", "inputHash": "53abc9898d297ab3c847df6f9b46d144494ee1250d01d21cae278fa6b93f7995", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "# Agent Systems Benchmark (ASB)", "canonicalNameSource": "official_readme", "decisionReason": "The repository is under active development, with key provider recording/replay and live-provider selection not implemented, and lacks a clear stable scoring contract or demonstrated public reuse path.", "evaluationEvidence": {"reuse": "The CLI implements `doctor`, `plan`, `run`, `sweep`, `compare`, `report` and the local `serve` control endpoint. `run` and `sweep` execute the original in-tree workloads through a digest-pinned `batch-stdio-v1` executable, then persist bounded run evidence for reporting and comparison.", "scoring": "```text asb doctor asb plan EXPERIMENT.toml asb run EXPERIMENT.toml asb sweep EXPERIMENT.toml asb compare RUN... asb report RUN... asb serve CONTROL.toml ``` `plan` validates without launching. `run` executes one configured capacity point; `sweep` executes the bounded range in the plan. `compare` requires at least two persisted run directories and `report` requires at least one.", "task": "ASB is a Linux terminal framework for measuring how AI coding agents scale: how many concurrent sessions a system can sustain while task quality, latency and resource consumption remain within declared bounds."}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-09T01:08:06.842497Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-candidate", "demo": false, "description": "A Linux terminal framework for measuring how AI coding agents scale under concurrent sessions while maintaining task quality, latency, and resource consumption within declared bounds.", "discoverySources": [{"id": "github:martin-beck/agent-systems-benchmark", "type": "github", "url": "https://github.com/martin-beck/agent-systems-benchmark"}], "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["discovered via github", "benchmark term in abstract", "evaluation protocol evidence", "public artifact URL", "no explicit benchmark release evidence"], "snippet": "agent-systems-benchmark Linux systems benchmark for AI coding agents ai-agents benchmarking linux # Agent Systems Benchmark (ASB) ASB is a Linux terminal framework for measuring how AI coding agents scale: how many concurrent sessions a system can sustain while task quality, latency and resource consumption remain within declared bounds."}, "familyId": "bmf_23eeec5f1b4b", "firstSeenAt": "2026-09-07", "heat": null, "id": "bm_agent-systems-benchmark_6fb06572", "indexedAt": "2026-09-07T06:22:04.880155Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/martin-beck/agent-systems-benchmark"}, "metrics": [], "motivation": "agent-systems-benchmark Linux systems benchmark for AI coding agents ai-agents benchmarking linux # Agent Systems Benchmark (ASB) ASB is a Linux terminal framework for measuring how AI coding agents scale: how many concurrent sessions a system can sustain while task quality, latency and resource consumption remain within declared bounds.", "name": "Agent Systems Benchmark (ASB)", "oneLine": "A Linux terminal framework for measuring how AI coding agents scale under concurrent sessions while maintaining task quality, latency, and resource consumption within declared bounds.", "paperTitle": "agent-systems-benchmark", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://github.com/martin-beck/agent-systems-benchmark", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-07T06:22:04.880155Z"}, "publications": [], "ranking": {"30d": {"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": 30, "historyDate": "2026-09-05", "rank": null, "score": null, "windowDays": 30}, "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": 205, "score": 23}, "method": "attention-ranking-v13", "rank": 205, "score": 23}, "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": 469, "score": 23}, "method": "attention-ranking-v13", "rank": 469, "score": 23}}, "readiness": "Runnable", "recognitionConfidence": 0.4, "relation": "unclear", "releasedAt": "2026-09-06", "source": {"authors": ["martin-beck"], "categories": [], "id": "github:martin-beck/agent-systems-benchmark", "publicSignals": {"githubStars": 0}, "title": "agent-systems-benchmark", "type": "github", "url": "https://github.com/martin-beck/agent-systems-benchmark"}, "sourceUpdatedAt": "2026-09-07T04:11:15Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "It addresses the scarcity of reproducible benchmarks for evaluating the operational scalability of multi-agent coding systems, which is critical for production deployment decisions."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
