{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KPVMV4X4LIOJKWALS7BR4VQCIW","short_pith_number":"pith:KPVMV4X4","canonical_record":{"source":{"id":"2604.18543","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-04-20T17:36:49Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"929be60f014861936cc2403c6b2b93219a7d1af4dfe39e50deaf07901b31ec8d","abstract_canon_sha256":"3dac45e93250f62ab2a8158f719a4ef41f09d3347377cfb1829867614ae70ebe"},"schema_version":"1.0"},"canonical_sha256":"53eacaf2fc5a1c95580b97c31e56024599f6d7d071549694b9c2639da5e724c1","source":{"kind":"arxiv","id":"2604.18543","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.18543","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"arxiv_version","alias_value":"2604.18543v4","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.18543","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"pith_short_12","alias_value":"KPVMV4X4LIOJ","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"pith_short_16","alias_value":"KPVMV4X4LIOJKWAL","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"pith_short_8","alias_value":"KPVMV4X4","created_at":"2026-06-11T01:09:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KPVMV4X4LIOJKWALS7BR4VQCIW","target":"record","payload":{"canonical_record":{"source":{"id":"2604.18543","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-04-20T17:36:49Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"929be60f014861936cc2403c6b2b93219a7d1af4dfe39e50deaf07901b31ec8d","abstract_canon_sha256":"3dac45e93250f62ab2a8158f719a4ef41f09d3347377cfb1829867614ae70ebe"},"schema_version":"1.0"},"canonical_sha256":"53eacaf2fc5a1c95580b97c31e56024599f6d7d071549694b9c2639da5e724c1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:09:35.723754Z","signature_b64":"rtHm12Gkz7gR7lpKntOK82joNHPUVXbV8uHNHQVyP0cHjZXxKQe91/I4VQsfRRY8pM6EfmoqK4TQLjh4DB7xBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53eacaf2fc5a1c95580b97c31e56024599f6d7d071549694b9c2639da5e724c1","last_reissued_at":"2026-06-11T01:09:35.722689Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:09:35.722689Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.18543","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-11T01:09:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6QOoKKyjcAfK20nMIDe3nOsUW9l64reTkDa0hC1OwA8qOEyztWMu5RkjhrNQoN27iengQxQh3oXrm1EbEjrJAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T09:47:56.424071Z"},"content_sha256":"a6beb62436b7b62ab6b724fa67375256060f55b11de8af1eb61ace06730363cf","schema_version":"1.0","event_id":"sha256:a6beb62436b7b62ab6b724fa67375256060f55b11de8af1eb61ace06730363cf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KPVMV4X4LIOJKWALS7BR4VQCIW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ClawEnvKit: Automatic Environment Generation for Claw-Like Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language descriptions.","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Cho-Jui Hsieh, Ion Stoica, Ming Li, Tianyi Zhou, Xirui Li","submitted_at":"2026-04-20T17:36:49Z","abstract_excerpt":"Constructing environments for training and evaluating claw-like agents remains a manual, human-intensive process that does not scale. We argue that what is needed is not just a dataset, but an automated pipeline capable of generating diverse, verified environments on demand. To this end, we introduce ClawEnvKit, an autonomous generation pipeline that instantiates this formalism from natural language descriptions. The pipeline comprises three modules: (1) a parser that extracts structured generation parameters from natural language input; (2) a generator that produces the task specification, to"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Using ClawEnvKit, we construct Auto-ClawEval, the first large-scale benchmark for claw-like agents, comprising 1,040 environments across 24 categories. Empirically, Auto-ClawEval matches or exceeds human-curated environments on coherence and clarity at 13,800x lower cost.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The validator can reliably enforce feasibility, diversity, structural validity, and internal consistency for environments generated from arbitrary natural language descriptions without missing critical edge cases or introducing systematic biases.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language, producing the Auto-ClawEval benchmark of 1,040 environments that matches human-curated quality at 13,800x lower cost.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language descriptions.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"adf68b1ee4c23dccbcddcdb62ab2a2eadadfb2a1516255beb2e55aa05370ec4e"},"source":{"id":"2604.18543","kind":"arxiv","version":4},"verdict":{"id":"1a13e02d-2297-4f42-9a00-bb946fa0ef44","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T04:21:06.430100Z","strongest_claim":"Using ClawEnvKit, we construct Auto-ClawEval, the first large-scale benchmark for claw-like agents, comprising 1,040 environments across 24 categories. Empirically, Auto-ClawEval matches or exceeds human-curated environments on coherence and clarity at 13,800x lower cost.","one_line_summary":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language, producing the Auto-ClawEval benchmark of 1,040 environments that matches human-curated quality at 13,800x lower cost.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The validator can reliably enforce feasibility, diversity, structural validity, and internal consistency for environments generated from arbitrary natural language descriptions without missing critical edge cases or introducing systematic biases.","pith_extraction_headline":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language descriptions."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.18543/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-20T03:53:19.084606Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"38cf7a8fe292d49355c79e3014b02f53ed2a82651e4c384aed4fd72db95ab375"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"1a13e02d-2297-4f42-9a00-bb946fa0ef44"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-11T01:09:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H5mDO3lqklcdOXX4hPHVcp9fyNAPpD4NiqYU9bERYWBTWLTIDPGvx8XUKJEt5/DMDchiJatk2boLuR7lOdDrCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T09:47:56.424549Z"},"content_sha256":"e5606c84fe1270e2eec241d5d1ee501e215453e43c59e61008e47297b0e0ed7a","schema_version":"1.0","event_id":"sha256:e5606c84fe1270e2eec241d5d1ee501e215453e43c59e61008e47297b0e0ed7a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KPVMV4X4LIOJKWALS7BR4VQCIW/bundle.json","state_url":"https://pith.science/pith/KPVMV4X4LIOJKWALS7BR4VQCIW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KPVMV4X4LIOJKWALS7BR4VQCIW/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-12T09:47:56Z","links":{"resolver":"https://pith.science/pith/KPVMV4X4LIOJKWALS7BR4VQCIW","bundle":"https://pith.science/pith/KPVMV4X4LIOJKWALS7BR4VQCIW/bundle.json","state":"https://pith.science/pith/KPVMV4X4LIOJKWALS7BR4VQCIW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KPVMV4X4LIOJKWALS7BR4VQCIW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KPVMV4X4LIOJKWALS7BR4VQCIW","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"3dac45e93250f62ab2a8158f719a4ef41f09d3347377cfb1829867614ae70ebe","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-04-20T17:36:49Z","title_canon_sha256":"929be60f014861936cc2403c6b2b93219a7d1af4dfe39e50deaf07901b31ec8d"},"schema_version":"1.0","source":{"id":"2604.18543","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.18543","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"arxiv_version","alias_value":"2604.18543v4","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.18543","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"pith_short_12","alias_value":"KPVMV4X4LIOJ","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"pith_short_16","alias_value":"KPVMV4X4LIOJKWAL","created_at":"2026-06-11T01:09:35Z"},{"alias_kind":"pith_short_8","alias_value":"KPVMV4X4","created_at":"2026-06-11T01:09:35Z"}],"graph_snapshots":[{"event_id":"sha256:e5606c84fe1270e2eec241d5d1ee501e215453e43c59e61008e47297b0e0ed7a","target":"graph","created_at":"2026-06-11T01:09:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Using ClawEnvKit, we construct Auto-ClawEval, the first large-scale benchmark for claw-like agents, comprising 1,040 environments across 24 categories. Empirically, Auto-ClawEval matches or exceeds human-curated environments on coherence and clarity at 13,800x lower cost."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The validator can reliably enforce feasibility, diversity, structural validity, and internal consistency for environments generated from arbitrary natural language descriptions without missing critical edge cases or introducing systematic biases."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language, producing the Auto-ClawEval benchmark of 1,040 environments that matches human-curated quality at 13,800x lower cost."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language descriptions."}],"snapshot_sha256":"adf68b1ee4c23dccbcddcdb62ab2a2eadadfb2a1516255beb2e55aa05370ec4e"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-20T03:53:19.084606Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.18543/integrity.json","findings":[],"snapshot_sha256":"38cf7a8fe292d49355c79e3014b02f53ed2a82651e4c384aed4fd72db95ab375","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Constructing environments for training and evaluating claw-like agents remains a manual, human-intensive process that does not scale. We argue that what is needed is not just a dataset, but an automated pipeline capable of generating diverse, verified environments on demand. To this end, we introduce ClawEnvKit, an autonomous generation pipeline that instantiates this formalism from natural language descriptions. The pipeline comprises three modules: (1) a parser that extracts structured generation parameters from natural language input; (2) a generator that produces the task specification, to","authors_text":"Cho-Jui Hsieh, Ion Stoica, Ming Li, Tianyi Zhou, Xirui Li","cross_cats":["cs.CL"],"headline":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language descriptions.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-04-20T17:36:49Z","title":"ClawEnvKit: Automatic Environment Generation for Claw-Like Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.18543","kind":"arxiv","version":4},"verdict":{"created_at":"2026-05-10T04:21:06.430100Z","id":"1a13e02d-2297-4f42-9a00-bb946fa0ef44","model_set":{"reader":"grok-4.3"},"one_line_summary":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language, producing the Auto-ClawEval benchmark of 1,040 environments that matches human-curated quality at 13,800x lower cost.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"ClawEnvKit automates generation of diverse verified environments for claw-like agents from natural language descriptions.","strongest_claim":"Using ClawEnvKit, we construct Auto-ClawEval, the first large-scale benchmark for claw-like agents, comprising 1,040 environments across 24 categories. Empirically, Auto-ClawEval matches or exceeds human-curated environments on coherence and clarity at 13,800x lower cost.","weakest_assumption":"The validator can reliably enforce feasibility, diversity, structural validity, and internal consistency for environments generated from arbitrary natural language descriptions without missing critical edge cases or introducing systematic biases."}},"verdict_id":"1a13e02d-2297-4f42-9a00-bb946fa0ef44"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a6beb62436b7b62ab6b724fa67375256060f55b11de8af1eb61ace06730363cf","target":"record","created_at":"2026-06-11T01:09:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"3dac45e93250f62ab2a8158f719a4ef41f09d3347377cfb1829867614ae70ebe","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-04-20T17:36:49Z","title_canon_sha256":"929be60f014861936cc2403c6b2b93219a7d1af4dfe39e50deaf07901b31ec8d"},"schema_version":"1.0","source":{"id":"2604.18543","kind":"arxiv","version":4}},"canonical_sha256":"53eacaf2fc5a1c95580b97c31e56024599f6d7d071549694b9c2639da5e724c1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"53eacaf2fc5a1c95580b97c31e56024599f6d7d071549694b9c2639da5e724c1","first_computed_at":"2026-06-11T01:09:35.722689Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-11T01:09:35.722689Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rtHm12Gkz7gR7lpKntOK82joNHPUVXbV8uHNHQVyP0cHjZXxKQe91/I4VQsfRRY8pM6EfmoqK4TQLjh4DB7xBQ==","signature_status":"signed_v1","signed_at":"2026-06-11T01:09:35.723754Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.18543","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a6beb62436b7b62ab6b724fa67375256060f55b11de8af1eb61ace06730363cf","sha256:e5606c84fe1270e2eec241d5d1ee501e215453e43c59e61008e47297b0e0ed7a"],"state_sha256":"bf7d8e08852bb6ca2c8803cd94532449114ce3e89ae018921f5a139201982399"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SBRQoZ1vrO54IJu/x7qyrrgLp62pcJYdsbAyRNfcwjYyZbBV04jQxqABWFNnFSgbtyIlwmNc+y2Cg9FnqrtmCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-12T09:47:56.426891Z","bundle_sha256":"51f53f14618dd2d330dc300640bc19836fc0553508233cbf21c90fbc95b12ec9"}}