{"record": {"aliases": ["agentic-netdevops-benchmark"], "artifacts": [{"id": "gh:andersballegaard/agentic-netdevops-benchmark", "kind": "repo", "url": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark"}, {"kind": "website", "url": "https://raw.githubusercontent.com/AndersBallegaard/agentic-netdevops-benchmark/refs/heads/main/LLM.txt"}], "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "A multi-stage environment with a Linux management host and five VyOS routers, requiring agents to complete netdevops tasks from information gathering to evaluation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://github.com/AndersBallegaard/agentic-netdevops-benchmark", "source": "claire_radar", "source_url": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark", "title": "Agentic NetDevOps Benchmark"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:github:andersballegaard/agentic-netdevops-benchmark", "modality": null, "name": "Agentic NetDevOps Benchmark", "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_agentic-netdevops-benchmark_23670ccd", "origin_source": "github", "recognition_confidence": "0.4", "record_sha256": "f86767848c2863694148efc805f1ee25f169e96edb8029c92c729d0bd571f06c", "relation": "unclear", "review_model": "deepseek-v4-pro", "review_state": "source-reviewed-defer", "reviewed_at": "2026-09-08T16:57:34.057731Z", "source_url": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark"}, "publisher": null, "released": "2026-09-08", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:github:andersballegaard/agentic-netdevops-benchmark", "source_url": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-github-andersballegaard-agentic-netdevops-benchmark", "source": "claire_radar", "source_benchmark_id": "github:andersballegaard/agentic-netdevops-benchmark", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["agentic-netdevops-benchmark"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/AndersBallegaard/agentic-netdevops-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 niche netdevops focus and lack of automated scoring are likely to limit early engagement.", "score": 15}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-09T06:27:57.312318Z", "inputHash": "50f3b439f0659d6a048ee108e1ad8c8a5903de061abcfc9b2325acc65b3a19ea", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "# Agentic NetDevOps Benchmark", "canonicalNameSource": "official_readme", "decisionReason": "Primary-source review: public task stages and run artifacts are available, but the cited scoring evidence describes artifacts rather than a defined scoring rubric or pass/fail contract. Hold until a comparable evaluation rule is verified.", "evaluationEvidence": {"reuse": "Please fetch this URL and perform the task within https://raw.githubusercontent.com/AndersBallegaard/agentic-netdevops-benchmark/refs/heads/main/LLM.txt", "scoring": "The benchmark does not produce a traditional numeric score. Instead, each run creates artifacts: a branch, commit history, ADRs, evaluation notes, templates, and state information. These materials provide most of the meaningful signal.", "task": "The benchmark presents the agent with the following stages: - Stage 1 - Information gathering - Stage 2 - Basic setup - Stage 3 - Update devices - Stage 4 - Configure underlay VPN - Stage 5 - Setup overlay - Stage 6 - Setup L2VPN - Stage 7 - Internet VRF - Stage 8 - High availability - Stage 9 - Centralized zone-based firewalling - Stage 10 - PWHT - Stage 11 - iPOE authentication - Stage 12 - Evaluation"}, "method": "primary-source-review-v1", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-08T16:57:34.057731Z", "state": "source-reviewed-defer"}, "dataStatus": "primary-source-candidate", "demo": false, "description": "A multi-stage environment with a Linux management host and five VyOS routers, requiring agents to complete netdevops tasks from information gathering to evaluation.", "discoverySources": [{"id": "github:andersballegaard/agentic-netdevops-benchmark", "type": "github", "url": "https://github.com/AndersBallegaard/agentic-netdevops-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": "agentic-netdevops-benchmark A benchmark testing the ability for AI agents to perform netdevops tasks # Agentic NetDevOps Benchmark A benchmark designed to evaluate how AI agents perform on a deliberately atypical infrastructure."}, "familyId": "bmf_dccce876e396", "firstSeenAt": "2026-09-08", "heat": null, "id": "bm_agentic-netdevops-benchmark_23670ccd", "indexedAt": "2026-09-08T16:33:20.670873Z", "industrySectors": [], "links": {"code": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark", "project": "https://raw.githubusercontent.com/AndersBallegaard/agentic-netdevops-benchmark/refs/heads/main/LLM.txt"}, "metrics": [], "motivation": "agentic-netdevops-benchmark A benchmark testing the ability for AI agents to perform netdevops tasks # Agentic NetDevOps Benchmark A benchmark designed to evaluate how AI agents perform on a deliberately atypical infrastructure.", "name": "Agentic NetDevOps Benchmark", "oneLine": "A multi-stage environment with a Linux management host and five VyOS routers, requiring agents to complete netdevops tasks from information gathering to evaluation.", "paperTitle": "agentic-netdevops-benchmark", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-08T16:33:20.670873Z"}, "publications": [], "publishers": [{"name": "Anders Ballegaard", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark"}], "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": 195, "score": 23}, "method": "attention-ranking-v13", "rank": 195, "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": 459, "score": 23}, "method": "attention-ranking-v13", "rank": 459, "score": 23}}, "readiness": "Runnable", "recognitionConfidence": 0.4, "relation": "unclear", "releasedAt": "2026-09-08", "source": {"authors": ["AndersBallegaard"], "categories": [], "id": "github:andersballegaard/agentic-netdevops-benchmark", "publicSignals": {"githubStars": 0}, "title": "agentic-netdevops-benchmark", "type": "github", "url": "https://github.com/AndersBallegaard/agentic-netdevops-benchmark"}, "sourceUpdatedAt": "2026-09-08T15:16:41Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "It targets AI agent performance on atypical infrastructure, where no numeric score is produced and qualitative artifacts provide signal."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
