{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NP4IZKUYWJ2OF5ZXXR5S32MIKJ","short_pith_number":"pith:NP4IZKUY","schema_version":"1.0","canonical_sha256":"6bf88caa98b274e2f737bc7b2de9885255c27ad8ec8c44dad560babb187f0508","source":{"kind":"arxiv","id":"2606.19613","version":1},"attestation_state":"computed","paper":{"title":"StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Shuo Yang, Stefano Soatto, Vlad Sobal, Wei Xia, Yuting Zhang","submitted_at":"2026-06-17T21:36:09Z","abstract_excerpt":"We introduce StaminaBench, a benchmark that measures the stamina of coding agents: how many consecutive interaction turns (change requests) they can handle before failing. Unlike the prevailing fraction-of-tasks-solved metric, this matches real vibe-coding where sessions run dozens or hundreds of turns. In StaminaBench, agents implement a REST API server and modify it across a tunable number of procedurally generated follow-up change requests - 100 in our experiments, resulting in codebases of up to 6,000 lines. Tests are generated fully programmatically without LLM involvement, ensuring repro"},"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":"2606.19613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-17T21:36:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"de2cdc96eb966a9c6ec763f2f778b69ad41d48c1e11af27e5176c49918995c57","abstract_canon_sha256":"bafb25d41e092c73ba83da09e53c4d209e084068eab17522d44c574fa189c148"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:30.204146Z","signature_b64":"lXTp7RcOT8wXVlVs/1+9OLVKBGj2x+xKnvLzJv0zMKLbJPLwYtTotRJZv71BIsVAe8hN55rQRGEeJPSPjcCxCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6bf88caa98b274e2f737bc7b2de9885255c27ad8ec8c44dad560babb187f0508","last_reissued_at":"2026-06-19T16:12:30.203782Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:30.203782Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Shuo Yang, Stefano Soatto, Vlad Sobal, Wei Xia, Yuting Zhang","submitted_at":"2026-06-17T21:36:09Z","abstract_excerpt":"We introduce StaminaBench, a benchmark that measures the stamina of coding agents: how many consecutive interaction turns (change requests) they can handle before failing. Unlike the prevailing fraction-of-tasks-solved metric, this matches real vibe-coding where sessions run dozens or hundreds of turns. In StaminaBench, agents implement a REST API server and modify it across a tunable number of procedurally generated follow-up change requests - 100 in our experiments, resulting in codebases of up to 6,000 lines. Tests are generated fully programmatically without LLM involvement, ensuring repro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19613","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/2606.19613/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":"2606.19613","created_at":"2026-06-19T16:12:30.203844+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.19613v1","created_at":"2026-06-19T16:12:30.203844+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19613","created_at":"2026-06-19T16:12:30.203844+00:00"},{"alias_kind":"pith_short_12","alias_value":"NP4IZKUYWJ2O","created_at":"2026-06-19T16:12:30.203844+00:00"},{"alias_kind":"pith_short_16","alias_value":"NP4IZKUYWJ2OF5ZX","created_at":"2026-06-19T16:12:30.203844+00:00"},{"alias_kind":"pith_short_8","alias_value":"NP4IZKUY","created_at":"2026-06-19T16:12:30.203844+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/NP4IZKUYWJ2OF5ZXXR5S32MIKJ","json":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ.json","graph_json":"https://pith.science/api/pith-number/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/graph.json","events_json":"https://pith.science/api/pith-number/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/events.json","paper":"https://pith.science/paper/NP4IZKUY"},"agent_actions":{"view_html":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ","download_json":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ.json","view_paper":"https://pith.science/paper/NP4IZKUY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.19613&json=true","fetch_graph":"https://pith.science/api/pith-number/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/graph.json","fetch_events":"https://pith.science/api/pith-number/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/action/storage_attestation","attest_author":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/action/author_attestation","sign_citation":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/action/citation_signature","submit_replication":"https://pith.science/pith/NP4IZKUYWJ2OF5ZXXR5S32MIKJ/action/replication_record"}},"created_at":"2026-06-19T16:12:30.203844+00:00","updated_at":"2026-06-19T16:12:30.203844+00:00"}