{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TVCLVDSLUAFINYIQ3AQETNC3YR","short_pith_number":"pith:TVCLVDSL","canonical_record":{"source":{"id":"2605.14403","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T05:41:11Z","cross_cats_sorted":[],"title_canon_sha256":"d42a931dedcc7a55c3a6f1c75d7c51e5f1b2fbcc27c72312240bae7fded11678","abstract_canon_sha256":"ecbce4e7e338026f359f1072060c5209f7c4de3e364bbe662ecb7dbfffaf49cf"},"schema_version":"1.0"},"canonical_sha256":"9d44ba8e4ba00a86e110d82049b45bc46a6e97315478183141f90004ff2af2e4","source":{"kind":"arxiv","id":"2605.14403","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14403","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14403v1","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14403","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"TVCLVDSLUAFI","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"TVCLVDSLUAFINYIQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"TVCLVDSL","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TVCLVDSLUAFINYIQ3AQETNC3YR","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14403","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T05:41:11Z","cross_cats_sorted":[],"title_canon_sha256":"d42a931dedcc7a55c3a6f1c75d7c51e5f1b2fbcc27c72312240bae7fded11678","abstract_canon_sha256":"ecbce4e7e338026f359f1072060c5209f7c4de3e364bbe662ecb7dbfffaf49cf"},"schema_version":"1.0"},"canonical_sha256":"9d44ba8e4ba00a86e110d82049b45bc46a6e97315478183141f90004ff2af2e4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:07.460596Z","signature_b64":"xiIRWvuLMXgZhn+NaYNmN9WCMguxUSbLYYuZuf7MypxTdcpJgBs5NXAcG8cM7ANt1DyUOJ0MtyjfSgTQNE1bDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d44ba8e4ba00a86e110d82049b45bc46a6e97315478183141f90004ff2af2e4","last_reissued_at":"2026-05-17T23:39:07.459967Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:07.459967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14403","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7WdRVV8q5QNkKaSVJOffy+Pvt4BxFFgVG3ZSowh7M1rZAP1eXH98a/Lcd8uOqAxpi9RTUZTIBrHzVaXjumwTCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T15:20:42.484938Z"},"content_sha256":"9f8070ddc81f8e65c78be433db31a4de2017a8a944321e17fe2027769821a7fa","schema_version":"1.0","event_id":"sha256:9f8070ddc81f8e65c78be433db31a4de2017a8a944321e17fe2027769821a7fa"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TVCLVDSLUAFINYIQ3AQETNC3YR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DermAgent: A Self-Reflective Agentic System for Dermatological Image Analysis with Multi-Tool Reasoning and Traceable Decision-Making","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"DermAgent anchors each skin image prediction in retrieved cases and guidelines then self-corrects via critic gates to raise diagnostic accuracy above standard models.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feilong Tang, Lie Ju, Ming Hu, Siyuan Yan, Wei Feng, Xieji Li, Yize Liu, Zongyuan Ge","submitted_at":"2026-05-14T05:41:11Z","abstract_excerpt":"Dermatological diagnosis requires integrating fine-grained visual perception with expert clinical knowledge. Although Multimodal Large Language Models (MLLMs) facilitate interactive medical image analysis, their application in dermatology is hindered by insufficient domain-specific grounding and hallucinations. To address these issues, we propose DermAgent, a collaborative multi-tool agent that orchestrates seven specialized vision and language modules within a Plan-Execute-Reflect framework. DermAgent delivers stepwise, traceable diagnostic reasoning through three core components. First, it e"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"DermAgent consistently outperforms state-of-the-art MLLMs and medical agent baselines across zero-shot fine-grained disease diagnosis, concept annotation, and clinical captioning tasks, exceeding GPT-4o by 17.6% in skin disease diagnostic accuracy and 3.15% in captioning ROUGE-L.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The dual-modality retrieval from 413,210 diagnosed cases and 3,199 guideline chunks provides unbiased, comprehensive anchoring for every prediction, and the critic module's confidence-coverage-conflict gates reliably detect and correct hallucinations without introducing new errors or over-correction.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DermAgent orchestrates seven vision-language tools in a Plan-Execute-Reflect loop with dual-modality retrieval from 413k cases and a critic module to outperform GPT-4o by 17.6% in zero-shot dermatological diagnosis accuracy.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"DermAgent anchors each skin image prediction in retrieved cases and guidelines then self-corrects via critic gates to raise diagnostic accuracy above standard models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8acccfbf050f0deb092646722b2755d63fd1153ca71df69473a1cfa746866a10"},"source":{"id":"2605.14403","kind":"arxiv","version":1},"verdict":{"id":"393ca0aa-9def-41c7-b7ac-1b98ef12239e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:52:42.920869Z","strongest_claim":"DermAgent consistently outperforms state-of-the-art MLLMs and medical agent baselines across zero-shot fine-grained disease diagnosis, concept annotation, and clinical captioning tasks, exceeding GPT-4o by 17.6% in skin disease diagnostic accuracy and 3.15% in captioning ROUGE-L.","one_line_summary":"DermAgent orchestrates seven vision-language tools in a Plan-Execute-Reflect loop with dual-modality retrieval from 413k cases and a critic module to outperform GPT-4o by 17.6% in zero-shot dermatological diagnosis accuracy.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The dual-modality retrieval from 413,210 diagnosed cases and 3,199 guideline chunks provides unbiased, comprehensive anchoring for every prediction, and the critic module's confidence-coverage-conflict gates reliably detect and correct hallucinations without introducing new errors or over-correction.","pith_extraction_headline":"DermAgent anchors each skin image prediction in retrieved cases and guidelines then self-corrects via critic gates to raise diagnostic accuracy above standard models."},"references":{"count":33,"sample":[{"doi":"","year":null,"title":"https://dermnetnz.org/","work_id":"c42c6104-02f6-4e95-88db-a5b846a88c91","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"https://www.mayoclinic.org/diseases-conditions","work_id":"930fa1dc-52ff-40c2-9863-f5d00b782a56","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.48550/arxiv.2511.21631","year":2025,"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","ref_index":3,"cited_arxiv_id":"2511.21631","is_internal_anchor":true},{"doi":"10.48550/arxiv.2406.19280","year":2024,"title":"Huatuogpt-vision, towards injecting medical visual knowledge into multimodal llms at scale","work_id":"dd32b8a1-4ad4-4155-b031-c317b565c6e7","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.48550/arxiv.2302.00785","year":2023,"title":"https://doi.org/10.48550/arXiv.2302.00785","work_id":"307faa4a-126f-4df4-a7c4-0e19d8db60d2","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":33,"snapshot_sha256":"f4c7ac85d904bc957c5366881b68a6a142f0de73c98b965bc82849200b6a9da8","internal_anchors":4},"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":"393ca0aa-9def-41c7-b7ac-1b98ef12239e"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HymR8+5LqeP/xDg8pUTfdwiXJXYSMHxybjVPTLy74tN3vIRZImFAaw5Gw0Heklz31+VHuMpuumn4ckhOKYStCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T15:20:42.485519Z"},"content_sha256":"a3ee1eb99d2d261203fe0b90fdec55b5a0d69454a29addbfdfde67e28a26aabf","schema_version":"1.0","event_id":"sha256:a3ee1eb99d2d261203fe0b90fdec55b5a0d69454a29addbfdfde67e28a26aabf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TVCLVDSLUAFINYIQ3AQETNC3YR/bundle.json","state_url":"https://pith.science/pith/TVCLVDSLUAFINYIQ3AQETNC3YR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TVCLVDSLUAFINYIQ3AQETNC3YR/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":"2026-05-27T15:20:42Z","links":{"resolver":"https://pith.science/pith/TVCLVDSLUAFINYIQ3AQETNC3YR","bundle":"https://pith.science/pith/TVCLVDSLUAFINYIQ3AQETNC3YR/bundle.json","state":"https://pith.science/pith/TVCLVDSLUAFINYIQ3AQETNC3YR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TVCLVDSLUAFINYIQ3AQETNC3YR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TVCLVDSLUAFINYIQ3AQETNC3YR","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":"ecbce4e7e338026f359f1072060c5209f7c4de3e364bbe662ecb7dbfffaf49cf","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T05:41:11Z","title_canon_sha256":"d42a931dedcc7a55c3a6f1c75d7c51e5f1b2fbcc27c72312240bae7fded11678"},"schema_version":"1.0","source":{"id":"2605.14403","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14403","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14403v1","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14403","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"TVCLVDSLUAFI","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"TVCLVDSLUAFINYIQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"TVCLVDSL","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:a3ee1eb99d2d261203fe0b90fdec55b5a0d69454a29addbfdfde67e28a26aabf","target":"graph","created_at":"2026-05-17T23:39:07Z","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":"DermAgent consistently outperforms state-of-the-art MLLMs and medical agent baselines across zero-shot fine-grained disease diagnosis, concept annotation, and clinical captioning tasks, exceeding GPT-4o by 17.6% in skin disease diagnostic accuracy and 3.15% in captioning ROUGE-L."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The dual-modality retrieval from 413,210 diagnosed cases and 3,199 guideline chunks provides unbiased, comprehensive anchoring for every prediction, and the critic module's confidence-coverage-conflict gates reliably detect and correct hallucinations without introducing new errors or over-correction."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"DermAgent orchestrates seven vision-language tools in a Plan-Execute-Reflect loop with dual-modality retrieval from 413k cases and a critic module to outperform GPT-4o by 17.6% in zero-shot dermatological diagnosis accuracy."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"DermAgent anchors each skin image prediction in retrieved cases and guidelines then self-corrects via critic gates to raise diagnostic accuracy above standard models."}],"snapshot_sha256":"8acccfbf050f0deb092646722b2755d63fd1153ca71df69473a1cfa746866a10"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Dermatological diagnosis requires integrating fine-grained visual perception with expert clinical knowledge. Although Multimodal Large Language Models (MLLMs) facilitate interactive medical image analysis, their application in dermatology is hindered by insufficient domain-specific grounding and hallucinations. To address these issues, we propose DermAgent, a collaborative multi-tool agent that orchestrates seven specialized vision and language modules within a Plan-Execute-Reflect framework. DermAgent delivers stepwise, traceable diagnostic reasoning through three core components. First, it e","authors_text":"Feilong Tang, Lie Ju, Ming Hu, Siyuan Yan, Wei Feng, Xieji Li, Yize Liu, Zongyuan Ge","cross_cats":[],"headline":"DermAgent anchors each skin image prediction in retrieved cases and guidelines then self-corrects via critic gates to raise diagnostic accuracy above standard models.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T05:41:11Z","title":"DermAgent: A Self-Reflective Agentic System for Dermatological Image Analysis with Multi-Tool Reasoning and Traceable Decision-Making"},"references":{"count":33,"internal_anchors":4,"resolved_work":33,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"https://dermnetnz.org/","work_id":"c42c6104-02f6-4e95-88db-a5b846a88c91","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"https://www.mayoclinic.org/diseases-conditions","work_id":"930fa1dc-52ff-40c2-9863-f5d00b782a56","year":null},{"cited_arxiv_id":"2511.21631","doi":"10.48550/arxiv.2511.21631","is_internal_anchor":true,"ref_index":3,"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","year":2025},{"cited_arxiv_id":"","doi":"10.48550/arxiv.2406.19280","is_internal_anchor":false,"ref_index":4,"title":"Huatuogpt-vision, towards injecting medical visual knowledge into multimodal llms at scale","work_id":"dd32b8a1-4ad4-4155-b031-c317b565c6e7","year":2024},{"cited_arxiv_id":"","doi":"10.48550/arxiv.2302.00785","is_internal_anchor":false,"ref_index":5,"title":"https://doi.org/10.48550/arXiv.2302.00785","work_id":"307faa4a-126f-4df4-a7c4-0e19d8db60d2","year":2023}],"snapshot_sha256":"f4c7ac85d904bc957c5366881b68a6a142f0de73c98b965bc82849200b6a9da8"},"source":{"id":"2605.14403","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T02:52:42.920869Z","id":"393ca0aa-9def-41c7-b7ac-1b98ef12239e","model_set":{"reader":"grok-4.3"},"one_line_summary":"DermAgent orchestrates seven vision-language tools in a Plan-Execute-Reflect loop with dual-modality retrieval from 413k cases and a critic module to outperform GPT-4o by 17.6% in zero-shot dermatological diagnosis accuracy.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"DermAgent anchors each skin image prediction in retrieved cases and guidelines then self-corrects via critic gates to raise diagnostic accuracy above standard models.","strongest_claim":"DermAgent consistently outperforms state-of-the-art MLLMs and medical agent baselines across zero-shot fine-grained disease diagnosis, concept annotation, and clinical captioning tasks, exceeding GPT-4o by 17.6% in skin disease diagnostic accuracy and 3.15% in captioning ROUGE-L.","weakest_assumption":"The dual-modality retrieval from 413,210 diagnosed cases and 3,199 guideline chunks provides unbiased, comprehensive anchoring for every prediction, and the critic module's confidence-coverage-conflict gates reliably detect and correct hallucinations without introducing new errors or over-correction."}},"verdict_id":"393ca0aa-9def-41c7-b7ac-1b98ef12239e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9f8070ddc81f8e65c78be433db31a4de2017a8a944321e17fe2027769821a7fa","target":"record","created_at":"2026-05-17T23:39:07Z","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":"ecbce4e7e338026f359f1072060c5209f7c4de3e364bbe662ecb7dbfffaf49cf","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T05:41:11Z","title_canon_sha256":"d42a931dedcc7a55c3a6f1c75d7c51e5f1b2fbcc27c72312240bae7fded11678"},"schema_version":"1.0","source":{"id":"2605.14403","kind":"arxiv","version":1}},"canonical_sha256":"9d44ba8e4ba00a86e110d82049b45bc46a6e97315478183141f90004ff2af2e4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9d44ba8e4ba00a86e110d82049b45bc46a6e97315478183141f90004ff2af2e4","first_computed_at":"2026-05-17T23:39:07.459967Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:07.459967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xiIRWvuLMXgZhn+NaYNmN9WCMguxUSbLYYuZuf7MypxTdcpJgBs5NXAcG8cM7ANt1DyUOJ0MtyjfSgTQNE1bDA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:07.460596Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14403","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9f8070ddc81f8e65c78be433db31a4de2017a8a944321e17fe2027769821a7fa","sha256:a3ee1eb99d2d261203fe0b90fdec55b5a0d69454a29addbfdfde67e28a26aabf"],"state_sha256":"780926aa38923593ff698961e29534d02cae74e042771f02b4fb49f3da0bc1d4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XKnnfTfjcddmNaAKs+ggd39AsilZ/K9GrVbtFiw73XUJRckFHSBrbE5V75TV8NMlYS7Bc9BMPd2WdrbGg7HNCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T15:20:42.489046Z","bundle_sha256":"aa1f01cbc78e6722fa834fbcda35aa375059d833cf646522d37af609cb6449ab"}}