{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:IGUHBM3ILRES72S7KXNTQ2X4OP","short_pith_number":"pith:IGUHBM3I","canonical_record":{"source":{"id":"2601.18842","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-01-26T11:33:40Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"561de28f1fcdf32719f360e548a8e38dbf8b37710edc60c70b9647ab6cacfffe","abstract_canon_sha256":"df4bc60a646a7ab43836927e2e0bb50eeb3cac54304fe1a6c87541239d00104a"},"schema_version":"1.0"},"canonical_sha256":"41a870b3685c492fea5f55db386afc73ecf0d8eb0bcfde704d8971a5b8d6b2aa","source":{"kind":"arxiv","id":"2601.18842","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.18842","created_at":"2026-05-18T03:09:24Z"},{"alias_kind":"arxiv_version","alias_value":"2601.18842v3","created_at":"2026-05-18T03:09:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.18842","created_at":"2026-05-18T03:09:24Z"},{"alias_kind":"pith_short_12","alias_value":"IGUHBM3ILRES","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"IGUHBM3ILRES72S7","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"IGUHBM3I","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:IGUHBM3ILRES72S7KXNTQ2X4OP","target":"record","payload":{"canonical_record":{"source":{"id":"2601.18842","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-01-26T11:33:40Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"561de28f1fcdf32719f360e548a8e38dbf8b37710edc60c70b9647ab6cacfffe","abstract_canon_sha256":"df4bc60a646a7ab43836927e2e0bb50eeb3cac54304fe1a6c87541239d00104a"},"schema_version":"1.0"},"canonical_sha256":"41a870b3685c492fea5f55db386afc73ecf0d8eb0bcfde704d8971a5b8d6b2aa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:24.270895Z","signature_b64":"kCKQrmOm17hxOU1gNyHtWL6EGwFQHYGu1vglc8o89z8EHldjEIC15L7vZs+Y7o+x/kwos3aWgEUW4fDyrdOpDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"41a870b3685c492fea5f55db386afc73ecf0d8eb0bcfde704d8971a5b8d6b2aa","last_reissued_at":"2026-05-18T03:09:24.270098Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:24.270098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2601.18842","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1zYPkkYYnO/J2mgUK3GSEQTS/83jNB0ySP47fMPhSVG9zv9VN27RKpzsZ9r5wsYQ/8Dc34XsvAvkcveULnpMAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T20:27:14.193129Z"},"content_sha256":"b3d4fd9162dfa7e5bce286c4d213e223a096e6c6584494f09ab79f68312a2caf","schema_version":"1.0","event_id":"sha256:b3d4fd9162dfa7e5bce286c4d213e223a096e6c6584494f09ab79f68312a2caf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:IGUHBM3ILRES72S7KXNTQ2X4OP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GUIGuard-Bench: Toward a General Evaluation for Privacy-Preserving GUI Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"GUIGuard-Bench shows current models detect private information in GUI screenshots but struggle with precise localization, category recognition, risk assessment, and judging task necessity.","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CR","authors_text":"Jie Zhang, Jiyan He, Qiannan Zhu, Shuxin Zheng, Weiming Zhang, Wenbo Zhou, Yanxi Wang, Yu Shi, Zhiling Zhang","submitted_at":"2026-01-26T11:33:40Z","abstract_excerpt":"As GUI agents increasingly rely on screenshots to perceive and operate digital environments, they may inadvertently expose sensitive information such as identities, accounts, locations, and behavioral traces. While existing benchmarks primarily focus on task completion, grounding, or defenses against third-party attacks, current visual privacy datasets remain largely restricted to static natural images, limiting their ability to capture the contextual dependence and task relevance of privacy risks in GUI task trajectories. To bridge this gap, we introduce \\textbf{GUIGuard-Bench}, a first-step "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our results show that current models can often detect whether a screenshot contains private information, but they struggle with fine-grained localization, category recognition, risk assessment, and task-necessity judgment. We also find that closed-source models, exemplified by Claude Sonnet 4.6, can maintain largely consistent planner semantics in Android environments after privacy protection is applied.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The human-provided region-level annotations for privacy bounding boxes, categories, risk levels, and task necessity are accurate and representative of real-world GUI agent privacy risks across the collected trajectories.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GUIGuard-Bench is a new benchmark with annotated GUI screenshots that measures privacy recognition, planning fidelity under protection, and utility impact for trajectory-based GUI agents.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"GUIGuard-Bench shows current models detect private information in GUI screenshots but struggle with precise localization, category recognition, risk assessment, and judging task necessity.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f14d4c1d72c890f5f8b30b1fec81bdaaa9bfd75c5539ad0be5d226babc1483f0"},"source":{"id":"2601.18842","kind":"arxiv","version":3},"verdict":{"id":"3bebbdf8-7775-44f8-92ab-e0483d8e820e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T11:28:23.522972Z","strongest_claim":"Our results show that current models can often detect whether a screenshot contains private information, but they struggle with fine-grained localization, category recognition, risk assessment, and task-necessity judgment. We also find that closed-source models, exemplified by Claude Sonnet 4.6, can maintain largely consistent planner semantics in Android environments after privacy protection is applied.","one_line_summary":"GUIGuard-Bench is a new benchmark with annotated GUI screenshots that measures privacy recognition, planning fidelity under protection, and utility impact for trajectory-based GUI agents.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The human-provided region-level annotations for privacy bounding boxes, categories, risk levels, and task necessity are accurate and representative of real-world GUI agent privacy risks across the collected trajectories.","pith_extraction_headline":"GUIGuard-Bench shows current models detect private information in GUI screenshots but struggle with precise localization, category recognition, risk assessment, and judging task necessity."},"references":{"count":126,"sample":[{"doi":"","year":2025,"title":"https://blog","work_id":"315c2878-a94b-47e2-9449-4648d09db17b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Surfer 2: The next generation of cross- platform computer use agents","work_id":"f2aa74fc-5a55-4f6b-bf00-c477d3c9d375","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.18653/v1/2025","year":2025,"title":"Human Bias in the Face of AI: Examining Human Judgment Against Text Labeled as AI Generated","work_id":"e6d6ecae-3575-45e2-8c99-76359a926a1a","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Ghostei-bench: Do mobile agents resilience to environmental injection in dynamic on-device environments?","work_id":"b334713b-2584-421a-a1fe-3085ab1ca188","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Product webpage","work_id":"e6f957a9-2a2d-46de-8b04-d88e8a2ef26d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":126,"snapshot_sha256":"cf3f9b3cbfcfb24a971aa4795e27e33e4006432b05558778cf967b67468937ef","internal_anchors":5},"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":"3bebbdf8-7775-44f8-92ab-e0483d8e820e"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A+6Dr0T6o1GKIwgi0DYl11zFioaKblhTz2K2Qiofm58dmPNoQp8qEioKhScSrvNHLj3zCDYQ/Mkcxozovx7GDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T20:27:14.194256Z"},"content_sha256":"fbed1608699d3fe55c4754f0c5edb2d5411e4ee5cadf8e0920761153c8bb9f37","schema_version":"1.0","event_id":"sha256:fbed1608699d3fe55c4754f0c5edb2d5411e4ee5cadf8e0920761153c8bb9f37"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IGUHBM3ILRES72S7KXNTQ2X4OP/bundle.json","state_url":"https://pith.science/pith/IGUHBM3ILRES72S7KXNTQ2X4OP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IGUHBM3ILRES72S7KXNTQ2X4OP/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":"2026-05-30T20:27:14Z","links":{"resolver":"https://pith.science/pith/IGUHBM3ILRES72S7KXNTQ2X4OP","bundle":"https://pith.science/pith/IGUHBM3ILRES72S7KXNTQ2X4OP/bundle.json","state":"https://pith.science/pith/IGUHBM3ILRES72S7KXNTQ2X4OP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IGUHBM3ILRES72S7KXNTQ2X4OP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:IGUHBM3ILRES72S7KXNTQ2X4OP","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":"df4bc60a646a7ab43836927e2e0bb50eeb3cac54304fe1a6c87541239d00104a","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-01-26T11:33:40Z","title_canon_sha256":"561de28f1fcdf32719f360e548a8e38dbf8b37710edc60c70b9647ab6cacfffe"},"schema_version":"1.0","source":{"id":"2601.18842","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.18842","created_at":"2026-05-18T03:09:24Z"},{"alias_kind":"arxiv_version","alias_value":"2601.18842v3","created_at":"2026-05-18T03:09:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.18842","created_at":"2026-05-18T03:09:24Z"},{"alias_kind":"pith_short_12","alias_value":"IGUHBM3ILRES","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"IGUHBM3ILRES72S7","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"IGUHBM3I","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:fbed1608699d3fe55c4754f0c5edb2d5411e4ee5cadf8e0920761153c8bb9f37","target":"graph","created_at":"2026-05-18T03:09:24Z","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":"Our results show that current models can often detect whether a screenshot contains private information, but they struggle with fine-grained localization, category recognition, risk assessment, and task-necessity judgment. We also find that closed-source models, exemplified by Claude Sonnet 4.6, can maintain largely consistent planner semantics in Android environments after privacy protection is applied."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The human-provided region-level annotations for privacy bounding boxes, categories, risk levels, and task necessity are accurate and representative of real-world GUI agent privacy risks across the collected trajectories."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"GUIGuard-Bench is a new benchmark with annotated GUI screenshots that measures privacy recognition, planning fidelity under protection, and utility impact for trajectory-based GUI agents."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"GUIGuard-Bench shows current models detect private information in GUI screenshots but struggle with precise localization, category recognition, risk assessment, and judging task necessity."}],"snapshot_sha256":"f14d4c1d72c890f5f8b30b1fec81bdaaa9bfd75c5539ad0be5d226babc1483f0"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"As GUI agents increasingly rely on screenshots to perceive and operate digital environments, they may inadvertently expose sensitive information such as identities, accounts, locations, and behavioral traces. While existing benchmarks primarily focus on task completion, grounding, or defenses against third-party attacks, current visual privacy datasets remain largely restricted to static natural images, limiting their ability to capture the contextual dependence and task relevance of privacy risks in GUI task trajectories. To bridge this gap, we introduce \\textbf{GUIGuard-Bench}, a first-step ","authors_text":"Jie Zhang, Jiyan He, Qiannan Zhu, Shuxin Zheng, Weiming Zhang, Wenbo Zhou, Yanxi Wang, Yu Shi, Zhiling Zhang","cross_cats":["cs.AI","cs.CV"],"headline":"GUIGuard-Bench shows current models detect private information in GUI screenshots but struggle with precise localization, category recognition, risk assessment, and judging task necessity.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-01-26T11:33:40Z","title":"GUIGuard-Bench: Toward a General Evaluation for Privacy-Preserving GUI Agents"},"references":{"count":126,"internal_anchors":5,"resolved_work":126,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"https://blog","work_id":"315c2878-a94b-47e2-9449-4648d09db17b","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Surfer 2: The next generation of cross- platform computer use agents","work_id":"f2aa74fc-5a55-4f6b-bf00-c477d3c9d375","year":2025},{"cited_arxiv_id":"","doi":"10.18653/v1/2025","is_internal_anchor":false,"ref_index":3,"title":"Human Bias in the Face of AI: Examining Human Judgment Against Text Labeled as AI Generated","work_id":"e6d6ecae-3575-45e2-8c99-76359a926a1a","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Ghostei-bench: Do mobile agents resilience to environmental injection in dynamic on-device environments?","work_id":"b334713b-2584-421a-a1fe-3085ab1ca188","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Product webpage","work_id":"e6f957a9-2a2d-46de-8b04-d88e8a2ef26d","year":2025}],"snapshot_sha256":"cf3f9b3cbfcfb24a971aa4795e27e33e4006432b05558778cf967b67468937ef"},"source":{"id":"2601.18842","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-16T11:28:23.522972Z","id":"3bebbdf8-7775-44f8-92ab-e0483d8e820e","model_set":{"reader":"grok-4.3"},"one_line_summary":"GUIGuard-Bench is a new benchmark with annotated GUI screenshots that measures privacy recognition, planning fidelity under protection, and utility impact for trajectory-based GUI agents.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"GUIGuard-Bench shows current models detect private information in GUI screenshots but struggle with precise localization, category recognition, risk assessment, and judging task necessity.","strongest_claim":"Our results show that current models can often detect whether a screenshot contains private information, but they struggle with fine-grained localization, category recognition, risk assessment, and task-necessity judgment. We also find that closed-source models, exemplified by Claude Sonnet 4.6, can maintain largely consistent planner semantics in Android environments after privacy protection is applied.","weakest_assumption":"The human-provided region-level annotations for privacy bounding boxes, categories, risk levels, and task necessity are accurate and representative of real-world GUI agent privacy risks across the collected trajectories."}},"verdict_id":"3bebbdf8-7775-44f8-92ab-e0483d8e820e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b3d4fd9162dfa7e5bce286c4d213e223a096e6c6584494f09ab79f68312a2caf","target":"record","created_at":"2026-05-18T03:09:24Z","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":"df4bc60a646a7ab43836927e2e0bb50eeb3cac54304fe1a6c87541239d00104a","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-01-26T11:33:40Z","title_canon_sha256":"561de28f1fcdf32719f360e548a8e38dbf8b37710edc60c70b9647ab6cacfffe"},"schema_version":"1.0","source":{"id":"2601.18842","kind":"arxiv","version":3}},"canonical_sha256":"41a870b3685c492fea5f55db386afc73ecf0d8eb0bcfde704d8971a5b8d6b2aa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"41a870b3685c492fea5f55db386afc73ecf0d8eb0bcfde704d8971a5b8d6b2aa","first_computed_at":"2026-05-18T03:09:24.270098Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:24.270098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kCKQrmOm17hxOU1gNyHtWL6EGwFQHYGu1vglc8o89z8EHldjEIC15L7vZs+Y7o+x/kwos3aWgEUW4fDyrdOpDA==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:24.270895Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.18842","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b3d4fd9162dfa7e5bce286c4d213e223a096e6c6584494f09ab79f68312a2caf","sha256:fbed1608699d3fe55c4754f0c5edb2d5411e4ee5cadf8e0920761153c8bb9f37"],"state_sha256":"b1a924c78c26b72bb3f1071a92171bf1669975359f0db0aee63d5208ee5aebfc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZuUJZdBDpFy4R6aum0KKSIV94Ro0d2umr/VIl+Vfwi5gX8Y3r96oRUDNZ0CJ5k5iRNVw2XTOyzsKd+5nRMf7Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T20:27:14.199841Z","bundle_sha256":"087b26387f90066b0cbe9c9b22775951f32447952b2750073b4a2fa6a8a9839b"}}