{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GQR4WE3NNTL4QOFNNDPCXAFAIK","short_pith_number":"pith:GQR4WE3N","schema_version":"1.0","canonical_sha256":"3423cb136d6cd7c838ad68de2b80a042854b2b10af3c2c4568dd99f506640908","source":{"kind":"arxiv","id":"2606.17519","version":1},"attestation_state":"computed","paper":{"title":"Scaling Enterprise Agent Routing: Degradation, Diagnosis, and Recovery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kellen Gillespie, Robyn Perry","submitted_at":"2026-06-16T04:55:06Z","abstract_excerpt":"Production LLM assistants route user requests to growing libraries of specialized tools, but how does routing accuracy degrade as the catalog scales? We study single-step routing on a 110-agent, 584-tool catalog from a deployed enterprise productivity assistant, evaluating three frontier models from 10 to 110 agents. Routing F1 on under-specified requests drops 16--23 percentage points across models. An oracle analysis decomposes the degradation into a \\emph{retrieval} gap (the model cannot surface the right tool) and a \\emph{confusion} gap (even with perfect retrieval, the oracle ceiling drop"},"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":"2606.17519","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T04:55:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"52b486f108f575a3f9b70ea1a126d0eb6f7e8bb459bfaa3ec4f66b401196ebee","abstract_canon_sha256":"b0e9a2227da95c6a7006b702a4e1f5d50bd4bcc16e453ea8c14928856c804361"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:14.980205Z","signature_b64":"bBWHoAWZqQdlsaFGUN9NFa2dvKI+zBzRqB9h9dSvUU4/WToARbyMsrBZ+6eYuK0xDrgbjsbt1DOKpWQJWghDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3423cb136d6cd7c838ad68de2b80a042854b2b10af3c2c4568dd99f506640908","last_reissued_at":"2026-06-19T16:10:14.979851Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:14.979851Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Enterprise Agent Routing: Degradation, Diagnosis, and Recovery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kellen Gillespie, Robyn Perry","submitted_at":"2026-06-16T04:55:06Z","abstract_excerpt":"Production LLM assistants route user requests to growing libraries of specialized tools, but how does routing accuracy degrade as the catalog scales? We study single-step routing on a 110-agent, 584-tool catalog from a deployed enterprise productivity assistant, evaluating three frontier models from 10 to 110 agents. Routing F1 on under-specified requests drops 16--23 percentage points across models. An oracle analysis decomposes the degradation into a \\emph{retrieval} gap (the model cannot surface the right tool) and a \\emph{confusion} gap (even with perfect retrieval, the oracle ceiling drop"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.17519","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/2606.17519/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.17519","created_at":"2026-06-19T16:10:14.979912+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.17519v1","created_at":"2026-06-19T16:10:14.979912+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.17519","created_at":"2026-06-19T16:10:14.979912+00:00"},{"alias_kind":"pith_short_12","alias_value":"GQR4WE3NNTL4","created_at":"2026-06-19T16:10:14.979912+00:00"},{"alias_kind":"pith_short_16","alias_value":"GQR4WE3NNTL4QOFN","created_at":"2026-06-19T16:10:14.979912+00:00"},{"alias_kind":"pith_short_8","alias_value":"GQR4WE3N","created_at":"2026-06-19T16:10:14.979912+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/GQR4WE3NNTL4QOFNNDPCXAFAIK","json":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK.json","graph_json":"https://pith.science/api/pith-number/GQR4WE3NNTL4QOFNNDPCXAFAIK/graph.json","events_json":"https://pith.science/api/pith-number/GQR4WE3NNTL4QOFNNDPCXAFAIK/events.json","paper":"https://pith.science/paper/GQR4WE3N"},"agent_actions":{"view_html":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK","download_json":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK.json","view_paper":"https://pith.science/paper/GQR4WE3N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.17519&json=true","fetch_graph":"https://pith.science/api/pith-number/GQR4WE3NNTL4QOFNNDPCXAFAIK/graph.json","fetch_events":"https://pith.science/api/pith-number/GQR4WE3NNTL4QOFNNDPCXAFAIK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK/action/storage_attestation","attest_author":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK/action/author_attestation","sign_citation":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK/action/citation_signature","submit_replication":"https://pith.science/pith/GQR4WE3NNTL4QOFNNDPCXAFAIK/action/replication_record"}},"created_at":"2026-06-19T16:10:14.979912+00:00","updated_at":"2026-06-19T16:10:14.979912+00:00"}