{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:LZJMXACJ2QOP7HZXTFKINNIWCF","short_pith_number":"pith:LZJMXACJ","canonical_record":{"source":{"id":"2604.11088","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-13T07:10:01Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9e767301e3cdcde0e6a2b375adcccddcc3020265586dd650c38257304c39e7ca","abstract_canon_sha256":"9ddbe9ead40d1e75fbf7d5c1bd1d2688dc8c2b4a350c36f98988ecc1294a6228"},"schema_version":"1.0"},"canonical_sha256":"5e52cb8049d41cff9f37995486b516115c78fdf867d1b9c76f5964c7ca721b64","source":{"kind":"arxiv","id":"2604.11088","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.11088","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"arxiv_version","alias_value":"2604.11088v2","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.11088","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"pith_short_12","alias_value":"LZJMXACJ2QOP","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"pith_short_16","alias_value":"LZJMXACJ2QOP7HZX","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"pith_short_8","alias_value":"LZJMXACJ","created_at":"2026-05-29T01:05:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:LZJMXACJ2QOP7HZXTFKINNIWCF","target":"record","payload":{"canonical_record":{"source":{"id":"2604.11088","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-13T07:10:01Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"9e767301e3cdcde0e6a2b375adcccddcc3020265586dd650c38257304c39e7ca","abstract_canon_sha256":"9ddbe9ead40d1e75fbf7d5c1bd1d2688dc8c2b4a350c36f98988ecc1294a6228"},"schema_version":"1.0"},"canonical_sha256":"5e52cb8049d41cff9f37995486b516115c78fdf867d1b9c76f5964c7ca721b64","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:09.361213Z","signature_b64":"frFbqo4DjE5cPvigwi9VrOaBCuTF3ZA+9Mumiw1j7LjhRmjGjN+xbqtrOtfSY747pBGGJl/VnYaJpqrFCBhODw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5e52cb8049d41cff9f37995486b516115c78fdf867d1b9c76f5964c7ca721b64","last_reissued_at":"2026-05-29T01:05:09.360267Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:09.360267Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.11088","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-29T01:05:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YIcC9pLvGOyxclKiyJgnHqlCe2IzSBk4HZttExmT9zM4cPWktSTcEgJaOLjBmxIm9CHD0v1TKVuiKadiLtffBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T20:09:39.435043Z"},"content_sha256":"b0b39f4139c45382163e51a6e8574f6d79479d6f29445fa7366564afd21ef898","schema_version":"1.0","event_id":"sha256:b0b39f4139c45382163e51a6e8574f6d79479d6f29445fa7366564afd21ef898"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:LZJMXACJ2QOP7HZXTFKINNIWCF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Guardrails Beat Guidance: A Large-Scale Study of Rules, Skills, and Persistent Configuration for Coding Agents","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Negative constraints raise AI coding agent success rates while positive instructions lower them, and random rules match expert ones.","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Bing Zhu, Guanghui Wang, Peiyang He, Wei Qiu, Xing Zhang, Yanwei Cui, Ziyuan Li","submitted_at":"2026-04-13T07:10:01Z","abstract_excerpt":"Random rules improve a coding agent's task performance as much as expert-curated ones (both $+13.8$pp on a discriminative subset of SWE-bench Verified), and in our data every individually beneficial rule is a negative constraint (\"do not refactor unrelated code\"), while every individually harmful one is a positive directive (\"follow code style\"). We arrive at these findings through the first large-scale controlled study of agent rule files (\\texttt{CLAUDE.md}, \\texttt{.cursorrules}, and the broader family of agent skills, plugin manifests, and persona definitions): we scrape 679 rule files (25"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Rules improve performance by 7--14 percentage points, but random rules help as much as expert-curated ones -- suggesting rules work through context priming rather than specific instruction. Negative constraints are the only individually beneficial rule type, while positive directives actively hurt.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That differences in agent success rates are caused by the rule types themselves rather than unmeasured factors in how rules are inserted into prompts or variations across agent runs.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Negative constraints in agent rules improve coding performance via context priming while positive directives degrade it, with collective rules remaining helpful up to 50 rules.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Negative constraints raise AI coding agent success rates while positive instructions lower them, and random rules match expert ones.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e7dd00151feaf1be5a651358c4b6a755fc576a6d4ca45b0f75e2f20f231d11e7"},"source":{"id":"2604.11088","kind":"arxiv","version":2},"verdict":{"id":"a2e26d41-360e-4016-8608-abd8319657b3","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T15:33:36.613027Z","strongest_claim":"Rules improve performance by 7--14 percentage points, but random rules help as much as expert-curated ones -- suggesting rules work through context priming rather than specific instruction. Negative constraints are the only individually beneficial rule type, while positive directives actively hurt.","one_line_summary":"Negative constraints in agent rules improve coding performance via context priming while positive directives degrade it, with collective rules remaining helpful up to 50 rules.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That differences in agent success rates are caused by the rule types themselves rather than unmeasured factors in how rules are inserted into prompts or variations across agent runs.","pith_extraction_headline":"Negative constraints raise AI coding agent success rates while positive instructions lower them, and random rules match expert ones."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.11088/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":"a2e26d41-360e-4016-8608-abd8319657b3"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-29T01:05:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H5q7ZSTs2smZtYRdvONVx4hfQrZG6uMD0Sdcep2/vkvUjx/QgsnJkPTX0CsGHzEEmHyN8eF5T9PvhLmpd9LPDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T20:09:39.435531Z"},"content_sha256":"4bf7b7e6594fc0c46bb6863b8017d9b75246654b421e28afea684fcbe861f0e6","schema_version":"1.0","event_id":"sha256:4bf7b7e6594fc0c46bb6863b8017d9b75246654b421e28afea684fcbe861f0e6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LZJMXACJ2QOP7HZXTFKINNIWCF/bundle.json","state_url":"https://pith.science/pith/LZJMXACJ2QOP7HZXTFKINNIWCF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LZJMXACJ2QOP7HZXTFKINNIWCF/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":"2026-06-01T20:09:39Z","links":{"resolver":"https://pith.science/pith/LZJMXACJ2QOP7HZXTFKINNIWCF","bundle":"https://pith.science/pith/LZJMXACJ2QOP7HZXTFKINNIWCF/bundle.json","state":"https://pith.science/pith/LZJMXACJ2QOP7HZXTFKINNIWCF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LZJMXACJ2QOP7HZXTFKINNIWCF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:LZJMXACJ2QOP7HZXTFKINNIWCF","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":"9ddbe9ead40d1e75fbf7d5c1bd1d2688dc8c2b4a350c36f98988ecc1294a6228","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-13T07:10:01Z","title_canon_sha256":"9e767301e3cdcde0e6a2b375adcccddcc3020265586dd650c38257304c39e7ca"},"schema_version":"1.0","source":{"id":"2604.11088","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.11088","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"arxiv_version","alias_value":"2604.11088v2","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.11088","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"pith_short_12","alias_value":"LZJMXACJ2QOP","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"pith_short_16","alias_value":"LZJMXACJ2QOP7HZX","created_at":"2026-05-29T01:05:09Z"},{"alias_kind":"pith_short_8","alias_value":"LZJMXACJ","created_at":"2026-05-29T01:05:09Z"}],"graph_snapshots":[{"event_id":"sha256:4bf7b7e6594fc0c46bb6863b8017d9b75246654b421e28afea684fcbe861f0e6","target":"graph","created_at":"2026-05-29T01:05:09Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Rules improve performance by 7--14 percentage points, but random rules help as much as expert-curated ones -- suggesting rules work through context priming rather than specific instruction. Negative constraints are the only individually beneficial rule type, while positive directives actively hurt."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That differences in agent success rates are caused by the rule types themselves rather than unmeasured factors in how rules are inserted into prompts or variations across agent runs."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Negative constraints in agent rules improve coding performance via context priming while positive directives degrade it, with collective rules remaining helpful up to 50 rules."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Negative constraints raise AI coding agent success rates while positive instructions lower them, and random rules match expert ones."}],"snapshot_sha256":"e7dd00151feaf1be5a651358c4b6a755fc576a6d4ca45b0f75e2f20f231d11e7"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.11088/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Random rules improve a coding agent's task performance as much as expert-curated ones (both $+13.8$pp on a discriminative subset of SWE-bench Verified), and in our data every individually beneficial rule is a negative constraint (\"do not refactor unrelated code\"), while every individually harmful one is a positive directive (\"follow code style\"). We arrive at these findings through the first large-scale controlled study of agent rule files (\\texttt{CLAUDE.md}, \\texttt{.cursorrules}, and the broader family of agent skills, plugin manifests, and persona definitions): we scrape 679 rule files (25","authors_text":"Bing Zhu, Guanghui Wang, Peiyang He, Wei Qiu, Xing Zhang, Yanwei Cui, Ziyuan Li","cross_cats":["cs.CL"],"headline":"Negative constraints raise AI coding agent success rates while positive instructions lower them, and random rules match expert ones.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-13T07:10:01Z","title":"Guardrails Beat Guidance: A Large-Scale Study of Rules, Skills, and Persistent Configuration for Coding Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.11088","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T15:33:36.613027Z","id":"a2e26d41-360e-4016-8608-abd8319657b3","model_set":{"reader":"grok-4.3"},"one_line_summary":"Negative constraints in agent rules improve coding performance via context priming while positive directives degrade it, with collective rules remaining helpful up to 50 rules.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Negative constraints raise AI coding agent success rates while positive instructions lower them, and random rules match expert ones.","strongest_claim":"Rules improve performance by 7--14 percentage points, but random rules help as much as expert-curated ones -- suggesting rules work through context priming rather than specific instruction. Negative constraints are the only individually beneficial rule type, while positive directives actively hurt.","weakest_assumption":"That differences in agent success rates are caused by the rule types themselves rather than unmeasured factors in how rules are inserted into prompts or variations across agent runs."}},"verdict_id":"a2e26d41-360e-4016-8608-abd8319657b3"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b0b39f4139c45382163e51a6e8574f6d79479d6f29445fa7366564afd21ef898","target":"record","created_at":"2026-05-29T01:05:09Z","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":"9ddbe9ead40d1e75fbf7d5c1bd1d2688dc8c2b4a350c36f98988ecc1294a6228","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-13T07:10:01Z","title_canon_sha256":"9e767301e3cdcde0e6a2b375adcccddcc3020265586dd650c38257304c39e7ca"},"schema_version":"1.0","source":{"id":"2604.11088","kind":"arxiv","version":2}},"canonical_sha256":"5e52cb8049d41cff9f37995486b516115c78fdf867d1b9c76f5964c7ca721b64","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5e52cb8049d41cff9f37995486b516115c78fdf867d1b9c76f5964c7ca721b64","first_computed_at":"2026-05-29T01:05:09.360267Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T01:05:09.360267Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"frFbqo4DjE5cPvigwi9VrOaBCuTF3ZA+9Mumiw1j7LjhRmjGjN+xbqtrOtfSY747pBGGJl/VnYaJpqrFCBhODw==","signature_status":"signed_v1","signed_at":"2026-05-29T01:05:09.361213Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.11088","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b0b39f4139c45382163e51a6e8574f6d79479d6f29445fa7366564afd21ef898","sha256:4bf7b7e6594fc0c46bb6863b8017d9b75246654b421e28afea684fcbe861f0e6"],"state_sha256":"48e7fdb43d54c90f437e3816219adce8aa5ebadbe062317d67d1c592ed20bf14"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mOguO9w28nf0wssD5SaLQGdwpvofrODGJiEDsbeW+v85aE0Nq+PJu584hD27BG/LTYQfv0xkeqKshaG/0pFZAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T20:09:39.437849Z","bundle_sha256":"483af8db84ccdbf51206acb2bb2839d6c6e8d59841c8348c06f67fbd781e86db"}}