{"record": {"aliases": ["indian-agent-benchmark"], "artifacts": [{"id": "gh:imaginationai31-sys/indian-agent-benchmark", "kind": "repo", "url": "https://github.com/imaginationai31-sys/indian-agent-benchmark"}], "categories": ["Agents", "General AI", "Language & Knowledge"], "description": {"en": "A synthetic multilingual benchmark evaluating computer-use agents on browser-style e-commerce tasks across seven Indian languages, with expected action trajectories, 0-100 task success scoring, and automatic failure-mode classification."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://github.com/imaginationai31-sys/indian-agent-benchmark", "source": "claire_radar", "source_url": "https://github.com/imaginationai31-sys/indian-agent-benchmark", "title": "Indian Agent Benchmark"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:github:imaginationai31-sys/indian-agent-benchmark", "modality": null, "name": "Indian Agent 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_indian-agent-benchmark_f04be610", "origin_source": "github", "recognition_confidence": "0.4", "record_sha256": "70a3b086cd061ab39aee114bfa9fb62367fd085ea17b972b65ef085393a2bf08", "relation": "unclear", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-09-12T06:08:51.273940Z", "source_url": "https://github.com/imaginationai31-sys/indian-agent-benchmark"}, "publisher": null, "released": "2026-09-10", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:github:imaginationai31-sys/indian-agent-benchmark", "source_url": "https://github.com/imaginationai31-sys/indian-agent-benchmark"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-github-imaginationai31-sys-indian-agent-benchmark", "source": "claire_radar", "source_benchmark_id": "github:imaginationai31-sys/indian-agent-benchmark", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["indian-agent-benchmark"], "annotation": "Machine Generated", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/imaginationai31-sys/indian-agent-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": "Medium", "horizon": "7d", "reason": "A released multilingual Indian-language computer-use task set with runnable environments and tests addresses a comparatively uncovered agent setting, though the repository appears to be a small individual project.", "score": 44}, "capabilities": [], "confidence": "Medium", "construction": "Original Synthetic", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-13T09:43:44.761778Z", "inputHash": "0133f0ef447930ef30f2ab3079760ac8c79dd64b57edfc66930d90f2a4a86822", "model": "deepseek-v4-flash"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "# Indian Agent Benchmark A synthetic, multilingual benchmark for evaluating computer-use AI agents on realistic web tasks.", "canonicalNameSource": "official_readme", "decisionReason": "The repository describes 50 synthetic tasks and a 0–100 scoring range but does not include the actual task inputs, labels, or an evaluation script; it only provides a backend server and unit tests, making it impossible for others to score submissions.", "evaluationEvidence": {"reuse": "Run locally ```bash python backend/server.py ``` Then open `http://localhost:8000` in a browser.", "scoring": "Measure task success from 0–100", "task": "The benchmark now includes **50 runnable synthetic e-commerce tasks** covering English, Hindi, Hinglish, Tamil, Bengali, Telugu, and Marathi."}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-12T06:08:51.273940Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-candidate", "demo": false, "description": "A synthetic multilingual benchmark evaluating computer-use agents on browser-style e-commerce tasks across seven Indian languages, with expected action trajectories, 0-100 task success scoring, and automatic failure-mode classification.", "discoverySources": [{"id": "github:imaginationai31-sys/indian-agent-benchmark", "type": "github", "url": "https://github.com/imaginationai31-sys/indian-agent-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": "indian-agent-benchmark Indian multilingual computer-use agent benchmark with simulated web environments, task trajectories, evaluation, and training-ready datasets."}, "familyId": "bmf_58f4e76437ca", "firstSeenAt": "2026-09-11", "heat": null, "id": "bm_indian-agent-benchmark_f04be610", "indexedAt": "2026-09-11T06:20:17.513010Z", "industrySectors": [], "links": {"code": "https://github.com/imaginationai31-sys/indian-agent-benchmark", "project": "http://localhost:8000`"}, "metrics": [], "motivation": "indian-agent-benchmark Indian multilingual computer-use agent benchmark with simulated web environments, task trajectories, evaluation, and training-ready datasets.", "name": "Indian Agent Benchmark", "oneLine": "A synthetic multilingual benchmark evaluating computer-use agents on browser-style e-commerce tasks across seven Indian languages, with expected action trajectories, 0-100 task success scoring, and automatic failure-mode classification.", "paperTitle": "indian-agent-benchmark", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://github.com/imaginationai31-sys/indian-agent-benchmark", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-11T06:20:17.513010Z"}, "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": 187, "score": 23}, "method": "attention-ranking-v13", "rank": 187, "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": 451, "score": 23}, "method": "attention-ranking-v13", "rank": 451, "score": 23}}, "readiness": "Runnable", "recognitionConfidence": 0.4, "relation": "unclear", "releasedAt": "2026-09-10", "source": {"authors": ["imaginationai31-sys"], "categories": [], "id": "github:imaginationai31-sys/indian-agent-benchmark", "publicSignals": {"githubStars": 0}, "title": "indian-agent-benchmark", "type": "github", "url": "https://github.com/imaginationai31-sys/indian-agent-benchmark"}, "sourceUpdatedAt": "2026-09-10T23:52:57Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Computer-use agent evaluation is dominated by English and Western web environments, leaving multilingual and Indian-language constraint following largely unmeasured; the released tasks and scoring path let other teams inspect agent behavior on those settings."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
