{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:HK3G3XLTLBPKUEROQLUGS5JYBR","short_pith_number":"pith:HK3G3XLT","canonical_record":{"source":{"id":"2606.11435","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T20:43:23Z","cross_cats_sorted":[],"title_canon_sha256":"3556efc6e9d983d3be2ac058c2645585ad6b91d40d5e818bdb4733ea7051ddab","abstract_canon_sha256":"a70dabcc6e7b0823795d4d8ea0089abbc860bcbb8c5cf81cdcd169bc907bb163"},"schema_version":"1.0"},"canonical_sha256":"3ab66ddd73585eaa122e82e86975380c730956d22d0c927ba0c00353b77bb655","source":{"kind":"arxiv","id":"2606.11435","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.11435","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"arxiv_version","alias_value":"2606.11435v1","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11435","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"pith_short_12","alias_value":"HK3G3XLTLBPK","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"pith_short_16","alias_value":"HK3G3XLTLBPKUERO","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"pith_short_8","alias_value":"HK3G3XLT","created_at":"2026-06-11T01:09:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:HK3G3XLTLBPKUEROQLUGS5JYBR","target":"record","payload":{"canonical_record":{"source":{"id":"2606.11435","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T20:43:23Z","cross_cats_sorted":[],"title_canon_sha256":"3556efc6e9d983d3be2ac058c2645585ad6b91d40d5e818bdb4733ea7051ddab","abstract_canon_sha256":"a70dabcc6e7b0823795d4d8ea0089abbc860bcbb8c5cf81cdcd169bc907bb163"},"schema_version":"1.0"},"canonical_sha256":"3ab66ddd73585eaa122e82e86975380c730956d22d0c927ba0c00353b77bb655","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:09:48.460658Z","signature_b64":"IgJODBpwXUf3dEYNWqCHWussTsoAYPm6CLWKvrDE+HlK59jf1Hdg6MQ2nfZ3CsN3ClCNGGOPcVQdHUsX2OpRDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ab66ddd73585eaa122e82e86975380c730956d22d0c927ba0c00353b77bb655","last_reissued_at":"2026-06-11T01:09:48.459773Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:09:48.459773Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.11435","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":"2026-06-11T01:09:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LjoNYxn3IGBDui+9AHZfr+O2lWitB5Rj+0HlNelXvYmHvzv311WDrt4J0zNbhB9K9zIowhf7kDJOO0apdmjSAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T07:15:02.337643Z"},"content_sha256":"ad959b17c52592c7f88d3f53c6c958dfcef866f3e517a6f1d4ff08baa56ef22e","schema_version":"1.0","event_id":"sha256:ad959b17c52592c7f88d3f53c6c958dfcef866f3e517a6f1d4ff08baa56ef22e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:HK3G3XLTLBPKUEROQLUGS5JYBR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Agent Skill Evaluation and Evolution: Frameworks and Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Can Jin, Dimitris N. Metaxas, Feng Tong, Kexin Ding, Mu Zhou, Yang Zhou","submitted_at":"2026-06-09T20:43:23Z","abstract_excerpt":"The growth of agent skills has transformed how agentic systems are built, evaluated, and deployed. As skill libraries continue to scale, rigorous evaluation becomes critical to ensuring their utility, quality, and safety in real-world applications. Consequently, the field is undergoing an emerging paradigm shift from isolated skill creation to automated, evaluation-driven skill evolution. In this survey, we systematically examine the landscape of skill evolution and evaluation beyond foundational skill creation. We categorize evolution into four distinct paradigms, spanning execution feedback,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11435","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.11435/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-11T01:09:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hfInd/HLnBdn5EhWTpMQW079+KTaxy9FdSscBFlLVsnRVaB8JoLl0LR3wnq1KrfBpeOW3JhP6dDrnbrDvVhDBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T07:15:02.338003Z"},"content_sha256":"5537c9cd9861a27f38f416527ea017f93b237e4a68973fef043c28f8e6d91bd7","schema_version":"1.0","event_id":"sha256:5537c9cd9861a27f38f416527ea017f93b237e4a68973fef043c28f8e6d91bd7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HK3G3XLTLBPKUEROQLUGS5JYBR/bundle.json","state_url":"https://pith.science/pith/HK3G3XLTLBPKUEROQLUGS5JYBR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HK3G3XLTLBPKUEROQLUGS5JYBR/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":"2026-07-05T07:15:02Z","links":{"resolver":"https://pith.science/pith/HK3G3XLTLBPKUEROQLUGS5JYBR","bundle":"https://pith.science/pith/HK3G3XLTLBPKUEROQLUGS5JYBR/bundle.json","state":"https://pith.science/pith/HK3G3XLTLBPKUEROQLUGS5JYBR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HK3G3XLTLBPKUEROQLUGS5JYBR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HK3G3XLTLBPKUEROQLUGS5JYBR","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":"a70dabcc6e7b0823795d4d8ea0089abbc860bcbb8c5cf81cdcd169bc907bb163","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T20:43:23Z","title_canon_sha256":"3556efc6e9d983d3be2ac058c2645585ad6b91d40d5e818bdb4733ea7051ddab"},"schema_version":"1.0","source":{"id":"2606.11435","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.11435","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"arxiv_version","alias_value":"2606.11435v1","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11435","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"pith_short_12","alias_value":"HK3G3XLTLBPK","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"pith_short_16","alias_value":"HK3G3XLTLBPKUERO","created_at":"2026-06-11T01:09:48Z"},{"alias_kind":"pith_short_8","alias_value":"HK3G3XLT","created_at":"2026-06-11T01:09:48Z"}],"graph_snapshots":[{"event_id":"sha256:5537c9cd9861a27f38f416527ea017f93b237e4a68973fef043c28f8e6d91bd7","target":"graph","created_at":"2026-06-11T01:09:48Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.11435/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The growth of agent skills has transformed how agentic systems are built, evaluated, and deployed. As skill libraries continue to scale, rigorous evaluation becomes critical to ensuring their utility, quality, and safety in real-world applications. Consequently, the field is undergoing an emerging paradigm shift from isolated skill creation to automated, evaluation-driven skill evolution. In this survey, we systematically examine the landscape of skill evolution and evaluation beyond foundational skill creation. We categorize evolution into four distinct paradigms, spanning execution feedback,","authors_text":"Can Jin, Dimitris N. Metaxas, Feng Tong, Kexin Ding, Mu Zhou, Yang Zhou","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T20:43:23Z","title":"Agent Skill Evaluation and Evolution: Frameworks and Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11435","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ad959b17c52592c7f88d3f53c6c958dfcef866f3e517a6f1d4ff08baa56ef22e","target":"record","created_at":"2026-06-11T01:09:48Z","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":"a70dabcc6e7b0823795d4d8ea0089abbc860bcbb8c5cf81cdcd169bc907bb163","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T20:43:23Z","title_canon_sha256":"3556efc6e9d983d3be2ac058c2645585ad6b91d40d5e818bdb4733ea7051ddab"},"schema_version":"1.0","source":{"id":"2606.11435","kind":"arxiv","version":1}},"canonical_sha256":"3ab66ddd73585eaa122e82e86975380c730956d22d0c927ba0c00353b77bb655","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3ab66ddd73585eaa122e82e86975380c730956d22d0c927ba0c00353b77bb655","first_computed_at":"2026-06-11T01:09:48.459773Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-11T01:09:48.459773Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IgJODBpwXUf3dEYNWqCHWussTsoAYPm6CLWKvrDE+HlK59jf1Hdg6MQ2nfZ3CsN3ClCNGGOPcVQdHUsX2OpRDQ==","signature_status":"signed_v1","signed_at":"2026-06-11T01:09:48.460658Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.11435","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ad959b17c52592c7f88d3f53c6c958dfcef866f3e517a6f1d4ff08baa56ef22e","sha256:5537c9cd9861a27f38f416527ea017f93b237e4a68973fef043c28f8e6d91bd7"],"state_sha256":"8685e10ae60b7cd0f00ad33af6cae8a6f737c13f26f27b41eae5d375fd5ca0a8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+xrzAkrX3tJOGfa9reqnI0Jz0sDpUhcce4x30NZXNG6a4jc9j9plapgZbXzzSASErbSo3wtXCsfaY2JlVNJLDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-05T07:15:02.339825Z","bundle_sha256":"eeb29383ba21276f586c1492050aaec4f2cebf703f314a38e62a4a87b308d49c"}}