{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QFCLTIUX77VTP3BDYHYKE6Q76K","short_pith_number":"pith:QFCLTIUX","canonical_record":{"source":{"id":"2608.06301","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-06T17:21:05Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"1120a839c5d26efaf20b7f9d6e40b5edb1f399fbf8873087805ccb918d8daad3","abstract_canon_sha256":"1e521610acc37fa404b34b509fc2d46a4e828948ce70d0316bd7e334943d2029"},"schema_version":"1.0"},"canonical_sha256":"8144b9a297ffeb37ec23c1f0a27a1ff2867f417e170018a008d3c16f4cd7abe8","source":{"kind":"arxiv","id":"2608.06301","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.06301","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"arxiv_version","alias_value":"2608.06301v1","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.06301","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"pith_short_12","alias_value":"QFCLTIUX77VT","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"pith_short_16","alias_value":"QFCLTIUX77VTP3BD","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"pith_short_8","alias_value":"QFCLTIUX","created_at":"2026-08-07T01:41:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QFCLTIUX77VTP3BDYHYKE6Q76K","target":"record","payload":{"canonical_record":{"source":{"id":"2608.06301","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-06T17:21:05Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"1120a839c5d26efaf20b7f9d6e40b5edb1f399fbf8873087805ccb918d8daad3","abstract_canon_sha256":"1e521610acc37fa404b34b509fc2d46a4e828948ce70d0316bd7e334943d2029"},"schema_version":"1.0"},"canonical_sha256":"8144b9a297ffeb37ec23c1f0a27a1ff2867f417e170018a008d3c16f4cd7abe8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-07T01:41:19.623670Z","signature_b64":"4l98QFbdoy61CBz56Tw4KmGsgX4adl9fF7z2Y9uQBuYN83L/4murx4S+D2FoTrr+sH5GkLXpZEu2YB2CSqGTCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8144b9a297ffeb37ec23c1f0a27a1ff2867f417e170018a008d3c16f4cd7abe8","last_reissued_at":"2026-08-07T01:41:19.622012Z","signature_status":"signed_v1","first_computed_at":"2026-08-07T01:41:19.622012Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.06301","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":"2026-08-07T01:41:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XslRvzeR8hwIBYCtcT9M4aCFMHPI46D8ZzUMUhv9Jm5hU8PG1YaNCES2tDl37ilXdWsY4RcC55RR9GnGcWUHAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:00:43.989784Z"},"content_sha256":"b6974b3732cd70df6fdb7ce2c6838ad33f0323e45147fbbdde240d8018ddf6c7","schema_version":"1.0","event_id":"sha256:b6974b3732cd70df6fdb7ce2c6838ad33f0323e45147fbbdde240d8018ddf6c7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QFCLTIUX77VTP3BDYHYKE6Q76K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HarnessOpt-Bench: Evaluating LLMs at Harness Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Apaar Shanker, Shehab Yasser, Varun Ursekar, Veronica Chatrath, Vijay S. Kalmath, Yash Maurya, Yuan Xue","submitted_at":"2026-08-06T17:21:05Z","abstract_excerpt":"As LLMs are increasingly deployed within agentic systems, their capabilities depend not only on the model weights but also on the harness: the prompts, tools, control flow, memory, and orchestration code surrounding them. This makes automated harness optimization -- the iterative and evaluation-guided improvement of a harness by an AI system -- both an important route to improving AI systems and a demanding capability for AI systems themselves. Yet the community lacks a common protocol for measuring how well frontier LLMs perform at this task. We introduce HarnessOpt-Bench, a benchmark for end"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.06301","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/2608.06301/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":"2026-08-07T01:41:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mV/IgD7ZhsID0w4AndElvKGNgfDCFuB6oZF+g5XlSGcDWOxMY32Mf6Dk3Y9jYaXtd3bKVmyQREgpDBu4WN+lDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:00:43.990311Z"},"content_sha256":"6130ee7e7a52a8e14c8875923ec8eb23f4fb5eff8b5fe158a818e7a6bae65aa3","schema_version":"1.0","event_id":"sha256:6130ee7e7a52a8e14c8875923ec8eb23f4fb5eff8b5fe158a818e7a6bae65aa3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QFCLTIUX77VTP3BDYHYKE6Q76K/bundle.json","state_url":"https://pith.science/pith/QFCLTIUX77VTP3BDYHYKE6Q76K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QFCLTIUX77VTP3BDYHYKE6Q76K/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":"2026-08-09T14:00:43Z","links":{"resolver":"https://pith.science/pith/QFCLTIUX77VTP3BDYHYKE6Q76K","bundle":"https://pith.science/pith/QFCLTIUX77VTP3BDYHYKE6Q76K/bundle.json","state":"https://pith.science/pith/QFCLTIUX77VTP3BDYHYKE6Q76K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QFCLTIUX77VTP3BDYHYKE6Q76K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QFCLTIUX77VTP3BDYHYKE6Q76K","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":"1e521610acc37fa404b34b509fc2d46a4e828948ce70d0316bd7e334943d2029","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-06T17:21:05Z","title_canon_sha256":"1120a839c5d26efaf20b7f9d6e40b5edb1f399fbf8873087805ccb918d8daad3"},"schema_version":"1.0","source":{"id":"2608.06301","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.06301","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"arxiv_version","alias_value":"2608.06301v1","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.06301","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"pith_short_12","alias_value":"QFCLTIUX77VT","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"pith_short_16","alias_value":"QFCLTIUX77VTP3BD","created_at":"2026-08-07T01:41:19Z"},{"alias_kind":"pith_short_8","alias_value":"QFCLTIUX","created_at":"2026-08-07T01:41:19Z"}],"graph_snapshots":[{"event_id":"sha256:6130ee7e7a52a8e14c8875923ec8eb23f4fb5eff8b5fe158a818e7a6bae65aa3","target":"graph","created_at":"2026-08-07T01:41:19Z","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/2608.06301/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As LLMs are increasingly deployed within agentic systems, their capabilities depend not only on the model weights but also on the harness: the prompts, tools, control flow, memory, and orchestration code surrounding them. This makes automated harness optimization -- the iterative and evaluation-guided improvement of a harness by an AI system -- both an important route to improving AI systems and a demanding capability for AI systems themselves. Yet the community lacks a common protocol for measuring how well frontier LLMs perform at this task. We introduce HarnessOpt-Bench, a benchmark for end","authors_text":"Apaar Shanker, Shehab Yasser, Varun Ursekar, Veronica Chatrath, Vijay S. Kalmath, Yash Maurya, Yuan Xue","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-06T17:21:05Z","title":"HarnessOpt-Bench: Evaluating LLMs at Harness Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.06301","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:b6974b3732cd70df6fdb7ce2c6838ad33f0323e45147fbbdde240d8018ddf6c7","target":"record","created_at":"2026-08-07T01:41:19Z","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":"1e521610acc37fa404b34b509fc2d46a4e828948ce70d0316bd7e334943d2029","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-06T17:21:05Z","title_canon_sha256":"1120a839c5d26efaf20b7f9d6e40b5edb1f399fbf8873087805ccb918d8daad3"},"schema_version":"1.0","source":{"id":"2608.06301","kind":"arxiv","version":1}},"canonical_sha256":"8144b9a297ffeb37ec23c1f0a27a1ff2867f417e170018a008d3c16f4cd7abe8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8144b9a297ffeb37ec23c1f0a27a1ff2867f417e170018a008d3c16f4cd7abe8","first_computed_at":"2026-08-07T01:41:19.622012Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-07T01:41:19.622012Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4l98QFbdoy61CBz56Tw4KmGsgX4adl9fF7z2Y9uQBuYN83L/4murx4S+D2FoTrr+sH5GkLXpZEu2YB2CSqGTCg==","signature_status":"signed_v1","signed_at":"2026-08-07T01:41:19.623670Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.06301","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b6974b3732cd70df6fdb7ce2c6838ad33f0323e45147fbbdde240d8018ddf6c7","sha256:6130ee7e7a52a8e14c8875923ec8eb23f4fb5eff8b5fe158a818e7a6bae65aa3"],"state_sha256":"db36bb9e4039a81dcf3e0a3e5bcf8df92350203ffbf1682807d4b411e62851ee"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BM/Gr3Q4YarH2N+Sh8m3jj6LU+7p8+jKOz0tq3ikpgSLVzmGfU0he862NWRcgC8VhbTMNxdS/BQlY/Df4v4fAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T14:00:43.995498Z","bundle_sha256":"c72747e2df12700e7e8a6f5a916682d7a0d77170632dbd5d6bff89edd955e9cb"}}