{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TAVMBJ67FGXUIJCE3RAVBUETFQ","short_pith_number":"pith:TAVMBJ67","schema_version":"1.0","canonical_sha256":"982ac0a7df29af442444dc4150d0932c1c57d891023b804bd98d20dd5708539e","source":{"kind":"arxiv","id":"2605.09817","version":2},"attestation_state":"computed","paper":{"title":"Evaluating Tool Cloning in Agentic-AI Ecosystems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Tool cloning creates widespread hidden duplication across public agent-tool repositories.","cross_cats":["cs.CR"],"primary_cat":"cs.SE","authors_text":"David Jiang, Neil Gong, Taein Kim, Yuepeng Hu, Yuqi Jia","submitted_at":"2026-05-10T23:39:44Z","abstract_excerpt":"Agent tools are becoming a core interface through which LLM agents access external data, services, and execution environments. As these tools are distributed through public marketplaces, raw tool counts may substantially overstate ecosystem diversity if many repositories are cloned, lightly modified, or derived from shared templates. Such hidden duplication can contaminate benchmark splits, propagate vulnerable implementations, bias measurements of tool-use generalization, and raise provenance, attribution, and intellectual-property concerns. We present, to our knowledge, the first large-scale"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.09817","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-05-10T23:39:44Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"8d3347556766190236283151325702a0688fb7f608757e33a5b673e84b264f02","abstract_canon_sha256":"43eccb701e698e900205c314128c552cd4184a0035606db90dcd108ebbf4a6be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:16.636356Z","signature_b64":"sXgcS47Racm2pSsmKQoPaqxTdt/y06bes5d5nDxCe5SbaBvFrJt5aMX38Jk1z0yvifvF9TdEECevgWeh/Up5Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"982ac0a7df29af442444dc4150d0932c1c57d891023b804bd98d20dd5708539e","last_reissued_at":"2026-05-20T00:03:16.635297Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:16.635297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating Tool Cloning in Agentic-AI Ecosystems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Tool cloning creates widespread hidden duplication across public agent-tool repositories.","cross_cats":["cs.CR"],"primary_cat":"cs.SE","authors_text":"David Jiang, Neil Gong, Taein Kim, Yuepeng Hu, Yuqi Jia","submitted_at":"2026-05-10T23:39:44Z","abstract_excerpt":"Agent tools are becoming a core interface through which LLM agents access external data, services, and execution environments. As these tools are distributed through public marketplaces, raw tool counts may substantially overstate ecosystem diversity if many repositories are cloned, lightly modified, or derived from shared templates. Such hidden duplication can contaminate benchmark splits, propagate vulnerable implementations, bias measurements of tool-use generalization, and raise provenance, attribution, and intellectual-property concerns. We present, to our knowledge, the first large-scale"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"These results indicate that tool cloning is a pervasive and severe source of hidden duplication in agent-tool ecosystems. They further suggest that agent-tool datasets and benchmarks should account for repository provenance and implementation similarity when measuring tool diversity or constructing evaluation splits.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That lexical and fuzzy-structural similarity metrics, calibrated via manual verification of 100 sampled pairs per ecosystem, reliably distinguish true cloning from independent but coincidentally similar implementations without significant false positives.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Tool cloning is pervasive in agentic AI ecosystems, with 60% of high-Jaccard and 85% of high-ssdeep similar pairs verified as true clones in a study of over 8,800 repositories.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Tool cloning creates widespread hidden duplication across public agent-tool repositories.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"0d20769054a67e8b91091980aa7f631e3f8b97e81404aad6d74e0e6f27b764f2"},"source":{"id":"2605.09817","kind":"arxiv","version":2},"verdict":{"id":"a7bb4d6b-e5ea-4193-ae35-d3af1c04d0a3","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T01:57:26.145080Z","strongest_claim":"These results indicate that tool cloning is a pervasive and severe source of hidden duplication in agent-tool ecosystems. They further suggest that agent-tool datasets and benchmarks should account for repository provenance and implementation similarity when measuring tool diversity or constructing evaluation splits.","one_line_summary":"Tool cloning is pervasive in agentic AI ecosystems, with 60% of high-Jaccard and 85% of high-ssdeep similar pairs verified as true clones in a study of over 8,800 repositories.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That lexical and fuzzy-structural similarity metrics, calibrated via manual verification of 100 sampled pairs per ecosystem, reliably distinguish true cloning from independent but coincidentally similar implementations without significant false positives.","pith_extraction_headline":"Tool cloning creates widespread hidden duplication across public agent-tool repositories."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.09817/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T16:35:19.804887Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T12:31:17.683305Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T09:54:36.294667Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"d1ac579df76977fd5353d5ab6f831fcc1c0ec76c565d5fc8e740b8e1a04ed605"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.09817","created_at":"2026-05-20T00:03:16.635447+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.09817v2","created_at":"2026-05-20T00:03:16.635447+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09817","created_at":"2026-05-20T00:03:16.635447+00:00"},{"alias_kind":"pith_short_12","alias_value":"TAVMBJ67FGXU","created_at":"2026-05-20T00:03:16.635447+00:00"},{"alias_kind":"pith_short_16","alias_value":"TAVMBJ67FGXUIJCE","created_at":"2026-05-20T00:03:16.635447+00:00"},{"alias_kind":"pith_short_8","alias_value":"TAVMBJ67","created_at":"2026-05-20T00:03:16.635447+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ","json":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ.json","graph_json":"https://pith.science/api/pith-number/TAVMBJ67FGXUIJCE3RAVBUETFQ/graph.json","events_json":"https://pith.science/api/pith-number/TAVMBJ67FGXUIJCE3RAVBUETFQ/events.json","paper":"https://pith.science/paper/TAVMBJ67"},"agent_actions":{"view_html":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ","download_json":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ.json","view_paper":"https://pith.science/paper/TAVMBJ67","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.09817&json=true","fetch_graph":"https://pith.science/api/pith-number/TAVMBJ67FGXUIJCE3RAVBUETFQ/graph.json","fetch_events":"https://pith.science/api/pith-number/TAVMBJ67FGXUIJCE3RAVBUETFQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ/action/storage_attestation","attest_author":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ/action/author_attestation","sign_citation":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ/action/citation_signature","submit_replication":"https://pith.science/pith/TAVMBJ67FGXUIJCE3RAVBUETFQ/action/replication_record"}},"created_at":"2026-05-20T00:03:16.635447+00:00","updated_at":"2026-05-20T00:03:16.635447+00:00"}