neutral_id	source	target	current_line	current_label	current_tuple_digest	current_line_digest	cohort	prior_disposition	decision_basis	authorization_status	under_review_context	primary_sort_digest	control_sort_digest
case-fa13dc1364da68	notes/abstract-an-experience-only-when-you-can-state-the-boundary.md	notes/first-principles-reasoning-selects-for-explanatory-reach-over.md	56	grounds	a510de678d966a0510d62dc268a1b5a65b81d44ea9f4218031fd7c6eda668340	e8518698a84ebf4f4e15d618b8b3b254f69d220b61fd4a8a05a9d41a72b99e02	primary	premised-on	grounds-baseline:P	candidate-new	the negative test (predict where it fails) is the boundary test this note applies to memory formation	0114817b08944bc4a0f8f53d427d44dc97efcc4d77a05537276cf8fc2e8590a5	c38c94f31443123b0187c8ceb58fdb38153278a3c9a29c4c627b783d03ff3331
case-15b46d34f5e07e	notes/directory-placement-is-total-frontmatter-classification-is-partial.md	notes/why-notes-have-types.md	52	grounds	5ae9dcbbfe50df45820a03657fae7c8f552efaa077489f1e477cca20c93e16ad	dd45935c22f23c80e42b5b87d59c3db137703b7bf2c20bebf3b07aa5cfbc0866	primary	premised-on	grounds-baseline:P	candidate-new	establishes that type is opt-in and free-form, the premise that makes type partial rather than total	01e31868541ba15361657e5f9108bf6197312fc4beaab62c568afb9f2103f0a9	66afdd6f8538df340f371289d7eb06490322a0f00fd79ad4a4456c845edc8ffa
case-93f1ff2073f163	notes/readable-artifact-loop-is-the-tractable-unit-for-continual-learning.md	notes/context-efficiency-is-the-central-design-concern-in-agent-systems.md	49	grounds	afb3d92625bd8e5fcb7bd2e24bfde18bf5ba69471d2dabdc158bda831009cf94	179fb81958900b27dee637ddca2db1f243d72514fff35bf12b1aa4e7150a06e1	primary	premised-on	grounds-baseline:P	candidate-new	the shared scarce resource that makes natural-language and symbolic artifacts a structurally coupled optimization target	0226b4eb58de13ffbd5709d784113741a06c57e0bd899f836cb2e24ae1ebe08f	b21e47de55c3a68e0f0807822cd4e291ab5effcca172f5a113030a2acefed2d5
case-1b6d3c4df444c1	notes/technical-constraints-make-kb-objective-choice-engineering.md	notes/kb-goals-in-always-loaded-context-guide-inclusion-decisions.md	73	grounds	f47656aa7bfc24bb195d417538d6abf027484f04f997df4a2744484d8061b432	65f80a8b374816b760ed152ca6255593e4d5d9df9c37b2921b0bbd3a273a68e1	primary	premised-on	grounds-baseline:P	candidate-new	the global goals are the always-loaded top-level loss the hierarchy specializes	02c45e247f20b1f09ec0b0a0de05ef0ac7e27cfc8a836989d6ab2175bfb77aee	52b4c50d7be279241c02276189f5dfa69809de9c66036501049a75e0c8fecc3c
case-1a80e03a9cb978	notes/prose-has-no-dereference-reinforce-facts-at-point-of-use.md	notes/agentic-systems-interpret-underspecified-instructions.md	39	grounds	3de390735a03207049bdd1c619eece985424e22abbee87ca5196967b782109a0	f85a5b63fb13f0c97b4f3b8df6b69e396e7a6add318c58c409f18aa36e288556	primary	premised-on	grounds-baseline:P	candidate-new	"that applying a declared fact is an interpretive, unpredictable act — even a deterministic model picks an interpretation you can't read off the spec — is the premise behind ""no reliable dereference"""	02dd8521650ca1adc9bad707ee4dce2cec5aeb71d29aff3797874bc7d288cb67	6159adf8a3536f6dc7378ba407d25434aca0e2f8d122020d7bde4c0adfb1f643
case-8a530b0c39ba8e	notes/llm-generation-confidence-tracks-typicality-not-soundness.md	notes/the-boundary-of-automation-is-the-boundary-of-verification.md	34	grounds	85d5511a5e8b2e0cd1378b66b7893c5a4c54378b11fa7caece1672ac94e549d8	c187514c2bf1e984053165430c50f327936bb757ef18381c4c841785d761968a	primary	premised-on	grounds-baseline:P	candidate-new	soundness needs a separate verifier precisely because confidence does not track it	047b61cffbfb4b9e89ee04e3e08ffaa16da784b717ea1cb88abe45049612235a	05c679fa3341c924816624dfabcfb1ea9ca701e38d17540f503b600d9e7b4a69
case-c34b3568d535fc	notes/goedel-machines-are-a-proof-governed-case-of-self-modification.md	notes/false-positive-generation-is-filtered-before-retention.md	67	grounds	53ddae210a497dbf6d27c7081338a730bc737d59befbebb063d5c2a1c368d86c	2531b80b04d74bc726651be706e9c64ad9cd1ca3d705d271e01496de04447558	primary	premised-on	grounds-baseline:P	candidate-new	why guaranteed retention concentrates the prompt-editing corner's risk entirely at the gate	08b2eb5d8469626dbe0e3a3114aa57c6b86a1069f6da591006e16405c850bd83	9f7c90f3b80960ccf52bcb813d8c649af105cf7d992b9d59d012dcf3ddddd814
case-65f16a15010b0b	notes/theory-mediated-learning-may-improve-sample-efficiency-under-shifts.md	notes/a-proposal-selection-loop-requires-search-evaluation-and-retention.md	107	grounds	a936fa152174b13f5569266757bf03cee678d44b342589e74414e6e4debfdba7	b8b0c31d492c642474bbf674609b0a13280d1c74eccfaa73ed82552f86858015	primary	premised-on	grounds-baseline:P	candidate-new	the search/evaluation split that keeps theory generation distinct from reach-assessment	09aebc48d9c19b2224814f45a09b401d002c7e9a6c30baa6f9b3ff45c55bccbe	54a763c8154127f03a82831ee9dff785f380a5aefe309a86ab3a53e4c36e92f7
case-ca2d686c8aa979	notes/an-accepted-edit-verifies-the-change-not-the-rule.md	notes/oracle-strength-spectrum.md	33	grounds	d075c9cb3222682c4b30b62fc3e7ed59cb3d23ab6dade5086a7795778a135fb1	341020dd7bef0fd8fede98b0ca80b5e923d36c85a33c2a19e104eb972faaa01c	primary	premised-on	grounds-baseline:P	candidate-new	the instance/rule asymmetry is a position change on the oracle-strength spectrum within one learning loop	0b215cc94eb1d44fda1a335029d899014b2261ac41346c807f6ccc5162581eda	036b6a0f07ab87fcbd54d20c65506f7e9f9127d92080437b2b4570314ddacb80
case-9197caeafa3a8b	notes/design-for-the-first-time-human-except-on-access-cost.md	notes/context-efficiency-is-the-central-design-concern-in-agent-systems.md	24	grounds	816956609c5b32a6f4be7ab8d0e6fb20fd91f960ec05bf95fea081a937b9bda9	e685663647313aec32f6bbf8cf0bc64acf6c8c7d2a0364afebc63c9af53c0ef0	primary	premised-on	grounds-baseline:P	candidate-new	the agent's linear access cost is a context-window cost, and context is the binding scarce resource that makes the access-mode exception matter	0ce7aaf8127d550318b1812fa05410e78dbbb469b742a6e9734af863476b685a	837d4d379880524870478bf2f9a8e47d622fe0fef86550a4dc57001652be3e81
case-f88ac900b7b30e	notes/fix-what-the-executor-cant-determine-not-what-it-will.md	notes/solve-low-degree-of-freedom-subproblems-first-to-avoid-blocking.md	29	grounds	7a8afeecef7ade7e0ab3d3f5d3ac204649c27f3a7af0df697f2c1faaffce8e57	9a3b98b047e2751e8c642740857156bb12b7af707b2e737eb8a2c57b32086ae5	primary	premised-on	grounds-baseline:P	candidate-new	the ordering rule behind the second cost; the author's slot is earliest, so its commitments must be the least flexible ones	0d4138ae3f83873965204ea499d537c21945568280f06a4ccffa73bba0f78d74	79c80dd9020fcc09cf291655c0c00dc05c9923a3b1b6c749d5f598e865cb35c8
case-6c0a42a4b717c0	notes/semantic-review-catches-content-errors-that-structural-validation.md	notes/error-correction-works-above-chance-oracles-with-decorrelated-checks.md	50	grounds	ef1fc0829240d2cdca45b360364af53f1a115e98507f5fcdee2bf8d9d04d5063	6b5af4b9589d0bde6ca9c0e41ce2a81152e68cd50af755827309688ac528891d	primary	premised-on	grounds-baseline:P	candidate-new	the four semantic checks are decorrelated weak oracles; combining them amplifies discriminative power	0e1db59977fec3cfb1b830daf5da98c3ad08b1ef48be2e9e916fd5bc4cc931ed	7e65851450d03a246bb18217a03682a46f920b7bc1e83efce7d27fb78b4e8aa9
case-5996d10a688230	notes/an-enforced-tag-readme-is-a-moc-with-a-machine-checked-contract.md	notes/llm-recompute-cost-inverts-the-store-vs-recompute-default.md	50	grounds	bec1fa7d0bc0b43f0b2244801b4468034098becc85675dde56702cc698dacbbc	8b7ed13f6a6fcc45c5180121a68517a5a9159c1a6392cabfc1fa1315af2906c8	primary	premised-on	grounds-baseline:P	candidate-new	the economics under which the completeness check is cheap enough for the agent maintainer to supply what the human maintainer couldn't	0f35435127a49d99ffc5c4877b15b54569856f2b6b0cb0239dd12e447eafb333	961615cde0cf30594d6fc7479ad10b34bf3e19c79b8e6c288f9a0563468afe00
case-d78c572ba49b95	notes/short-composable-notes-maximize-combinatorial-discovery.md	notes/context-efficiency-is-the-central-design-concern-in-agent-systems.md	52	grounds	acb13cd739627b8b9c05aca91ef364a837b72272338b9e6d54e113d4b5ba4b35	cebf49b7824f930bc3b584e813ca1445791bbc2568c276f3b097adaac8dd86e1	primary	premised-on	grounds-baseline:P	candidate-new	bounded context makes co-loading capacity the scarce resource	10f52f954d1f5a37629c7ab7376fb1a458f68871c056d57035df175479fc1945	d7fc667edf445fd618e280a42d806c753a688f2df5f282eafa99f7cefd59d0fc
case-ad48aaedc0e182	notes/first-principles-are-inherited-constraints-not-design-choices.md	notes/short-composable-notes-maximize-combinatorial-discovery.md	59	grounds	46a8704ab0e0f7e224f88c776aaa0b1abb0bbff6023653844d1d6aa35ac2be68	565a9c68584701db01977a2a55e869b9dbc818bc5771e89d4326bd49bd3b6eb0	primary	premised-on	grounds-baseline:P	candidate-new	co-loading is why composability is inherited from how the consumer ingests artifacts	111fa8bc03547d1bfaf84e72165b447d4e2e5c9049594e8b2fd9973f72686951	c90efaf7789dc12902787cb3ada7cedab4817c3f00bceab545c20f1ed2a2d242
case-5fea6da992bdb6	notes/specification-level-separation-recovers-scoping-before-it-recovers.md	notes/llm-context-is-composed-without-scoping.md	33	grounds	869c324a95837a080d09b579975be89d7ecf0c2866144603af0c2ad94c18b5a5	fa127fcaf39b5bf811db0acb7bab8c7724acc9a37def656a24f48042ddf27434	primary	premised-on	grounds-baseline:P	candidate-new	the main gains OpenProse gets early are scoping gains, not hard reliability gains	12cc6630661c7ade8c0ce85e7c91ac643b5fb444cfdb0c590efa08710db7affe	54b6466cfe6a52242b81cdb3abdaf69c38fdb355bcf3d9309a10cdaf8667f657
case-a23b180adead23	notes/the-framework-is-often-larger-than-the-durable-contribution.md	notes/short-composable-notes-maximize-combinatorial-discovery.md	64	grounds	193488dd29023f1e1df611ecd86e7955edb797bdeaa37e1fb8d44fd7a639e95e	4e743cf3deea859ff54bcd57fba7f9c0f40ed53c8519412704053f8804601174	primary	premised-on	grounds-baseline:P	candidate-new	shared frameworks live as their own composable artifacts, so each note carries only its case's contribution	132a1c5f3800dab8e7ee011eaea4cbce3d3e6a846b8724ea0f52e2ac5367cc51	143b56a3c69a4ed7937a3f51cfa0f094ecfaf759dcdc5a2c1d920a843442f8e2
case-6152a54e6a18c8	notes/design-for-the-first-time-human-except-on-access-cost.md	notes/agentic-systems-interpret-underspecified-instructions.md	30	grounds	ed118e463bcacc1aec7e29ec53b654749b1d509c0f7f775e3b5fe9c85ea8f78d	c0dde090b3faa7558b0b9afd58d961f01adc32ba1888807d82b7c9730804f7ee	primary	premised-on	grounds-baseline:P	candidate-new	the closing confabulation exception is the underspecification case, where a human would ask for clarification but an agent may choose one interpretation and proceed	13ae8cf4bad6675d9a2e8d1683542504ee63bb1a692a9085a26f6117bbfbcce4	ee8eb4253971295bdd5a7116ce21ae7db084e56d3c30f61e163df36a2d6f3820
case-b1580658412f68	notes/structure-inference-needs-capture-at-the-decision-surface.md	notes/raw-accumulation-does-not-create-usable-memory.md	38	grounds	0c06832a638afafc6b98bf5c44bfeb90a70f100f1dab1379193dd3ff5d3952cf	5430762f4976b2ca9b0b4619a0bd6cc61ada43dfd3d747f2a8b7019cde19673f	primary	premised-on	grounds-baseline:P	candidate-new	state-without-why is exactly the ingress problem this note localizes to the capture point	142dfcec9d45cadcfb5053988523c60d0e5fb66a288a1c19f2e06b30b886501b	901d0b32df30991d7138577799362e6afc436a31abfab15497d6eb7cf57e3362
case-a627c01cfeb3e0	notes/weakly-discriminated-qualities-tend-to-be-underselected.md	notes/the-boundary-of-automation-is-the-boundary-of-verification.md	78	grounds	8481e2ad82e32c24935dc15cf07f3c5b3722dab9b9e790ff6202acc8cc442ec7	00c07a1996ba280f10c47a5ba89288b631fc3827d1869ac468e3d2c0978d7de2	primary	premised-on	grounds-baseline:P	candidate-new	verification cost determines which qualities can support warranted automation	16122967d80bfb88a749ccb644bf4bf5bfa0518ef7d151543373d3f80dc4b27a	872a0c363cb5600ae187faf6484ed80e69069cad443b1f84f39d0c6689a0c293
case-1989c918c7946c	notes/compiling-coordination-preserves-primitive-not-aggregate-authority.md	notes/any-symbolic-program-with-llm-calls-is-a-select-call-program.md	59	grounds	5d7e419d3c5213d371f258cfe31ca67cb0db04fb02bc3485d0783250b9cf532f	133a8da87a74ee4b09bf384e1f4dfb9c1f16b502990619d16f56eccfc3e3919a	primary	premised-on	grounds-baseline:P	candidate-new	formalizes why a workflow script's JavaScript control flow and local variables instantiate the same select/call shape as the prompt-driven strategy	18205dcbe45d7660dc03ff3ae7d3f79eae06114858d669da82dbb020987de923	b649b958e861a3a3882e1498068be9e2f9a034563ea0a4e1eb005443bdad7220
case-ca121ccba67434	notes/agent-context-is-constrained-by-soft-degradation-not-hard-token-limits.md	notes/information-value-is-observer-relative.md	63	grounds	c87abcd5f1a39155fd434481cf12c26321b274cc119e50cec4f34b99783d03a6	59022c535d7f38c213b2ac7e2f75f78b5262d847f3e9cdc56e1f99bda5beaf9a	primary	premised-on	grounds-baseline:P	candidate-new	observer-relativity is what makes the soft bound task-dependent	1b9df0f1caa73f5c8501288ddb02cc1c04577f71f374b8c5696dc1fe36b02e42	0f6be89d824300481f304c3d7e30d42a5a810ec1d47054e191eaee6bf2d2ae4d
case-2632bc38d6c3e7	notes/definitions/reflective-system.md	notes/definitions/actionable-methodology.md	75	grounds	bb502542d295019c8ccbbe6a9fcc52e72c5deed1d59e7595244aaae741faaf12	13ebeb7e42e456e1d4613fa44cc3835802b99022c23e753a83d3a8d93c5d3ea3	primary	premised-on	grounds-baseline:P	candidate-new	an internal process may act through a methodology, but actionability alone does not establish reflection	1dfb8fe0aa0d2b0012c98f371ed61fed64902ffacee7e3a0db861350d45f6e4b	4d8bdd59a2b8ea2aad9ad9e4a481513dd1b81000208479ef6fef917d186b54cc
case-c3ffb8d4901349	notes/criteria-edits-invalidate-verdicts-process-edits-invalidate-artifacts.md	notes/history-has-one-chance-to-become-checkable.md	28	grounds	017a1f3c929d9285aa76f231c3d89a09e3f0a4fb3565e77f7210ccfdca46ecae	0153f26189d732d449dc90bda073bbb178dd803baa1d65001fc1ab7b645fbc2e	primary	premised-on	grounds-baseline:P	candidate-new	the state/history asymmetry this corollary rides on — why process deficiencies are invisible to review-time criteria	21e5c3e463ecdfe3d2c7583f961a4315100944c8547b1560fe1085c78a8cc93b	8958ca5240c3f8c94d8e8786157788aa05a5715e0607498e4a6aab65704d765b
case-8444045b653617	notes/formal-systems-assess-explanatory-reach-through-causal-and-proof.md	notes/warranted-autonomy-is-bounded-by-oracle-domain.md	50	grounds	729c03d1970afcf18ae6bab0632b626dccc653126bfdee4097d836abfbb021a9	7e62779fd5b127c102aa5b9ca57786777205fb531b8c5eacac64fc428939f162	primary	premised-on	grounds-baseline:P	candidate-new	formal guarantees remain bounded by the oracle or axiomatization that produces them	226baec13d4ec7158eec40ca15412e306172f1a61b2724993db54dcf525799a7	67a5e286241d4e7feb0aee5983cc62c2ff10c10b70c3c73d7dd2cfc421cd3b04
case-d88171537c6fb6	notes/design-for-the-first-time-human-except-on-access-cost.md	notes/llm-context-is-a-homoiconic-medium.md	29	grounds	6626621ff6554494b275c5fcb3b2d3175f05c9fd43426db17c6590a3438b5e50	958b2cd5c6eb7b8c59b58c7c0d72bf771ff78832c1111872998f3041e37e4e36	primary	premised-on	grounds-baseline:P	candidate-new	the closing text-as-instruction exception rests on there being no program/data boundary in the window, so any loaded text can act as instruction	237a3d8d2fdeef389b448aa73e339d21387864e448051df7fdcf3ec4891800f7	e25f9b6ecaae629b92838435a15bca05d8dbf6111c1cc814e59909c84b09bacf
case-fd065e5a4769ae	notes/measuring-autonomy-well-enough-to-see-it-improve-is-an-open-problem.md	notes/definitions/self-improving-system.md	31	grounds	61bc19c111c8cbc11ecb3d132334b5a86fb05df84ee53bdf3a9ba1d0fc5ed957	6fe0ec03811fc19a1ab5d84d346f3a955b65da2f5f6405c5a771c6c328bdee77	primary	premised-on	grounds-baseline:P	candidate-new	the declared-boundary relativity the allocation profile inherits from membership	25793e7c60299b9ac8e72f0f36659f19c92319746651f69771d649a65046ad05	28303acd69d7e39ae0cb29e1f992e57280f0b18872a720089af5c7372778fb29
case-c299eb39a9c550	notes/technical-constraints-make-kb-objective-choice-engineering.md	notes/a-knowledge-base-holds-theories-descriptions-and-prescriptions-with.md	75	grounds	3e4d3193ae5030ea99a54a806fb13d96783f2f160304d761e5a10db12a2bfcea	bb3f61a465cd2812a6412181b6bbf0e1466b0e66da81e8b55ce9ba5eab29f012	primary	premised-on	grounds-baseline:P	candidate-new	the three profiles and their quality goals this note treats as proven objective bundles	2628ecbfd2922d98f7696a2c9c0bba76a09b38c98d73fd9484b1bd27115f57c9	420ee7739a6fd70720db17ca576e23c6c0cc86f5a0e5b1efc544a07c43cddfcd
case-e46581928a37b0	notes/the-chat-history-model-trades-context-efficiency-for-implementation.md	notes/context-efficiency-is-the-central-design-concern-in-agent-systems.md	33	grounds	258e0cc662709189d64b90c7f3d3686e2932d096fa48f35a88a1fef7be403430	103839602115f8fbf50162ea00c65061a9ffa0d69064ed1a8ffb05b3c4aa7440	primary	premised-on	grounds-baseline:P	candidate-new	bounded context makes selective loading a first-class architectural concern	280454d3a7bdd02fa82a42ee8b994de5776c01c0bcb345b6c968849b255bc47c	8cbd222e935306eb433c20777ae6a7749add08116086cf24aca31f3d428645b8
case-3d16fd3a85e16a	notes/methodological-and-computational-closure-track-different-changes.md	notes/a-methodology-governs-its-own-extension-only-as-far-as-it-settles.md	81	grounds	2f29f8174c38bbf05503befe8b333602a3cf92febd9bbeabe7fa24c8ce48e8c8	dba2178a331f17b837f540064e50c8b97d58049ec02f10dcdade7fc5f9139d90	primary	premised-on	grounds-baseline:P	candidate-new	supplies the methodological reading of closure	29eafc45a309a7113ac70037fc890288f60ff2a05c68cb8ae4d7b1cc7c34cc55	8d710ea64e719b0a0f49d1425b9ddd50b535d8c879e89deda06b66db001cccb3
case-10d4bb9800d9e7	notes/a-proximate-target-is-checked-for-achievement-not-for-warrant.md	notes/self-improvement-is-relative-to-a-declared-objective.md	67	grounds	2b1745355fecac8b0ce7790ec115b4cb92d35e26a1faff35d65628a082b573fe	1526a14eafa157194595b5f192cdea3a5c379139e9355b4cde3200cd5b68bdfc	primary	premised-on	grounds-baseline:P	candidate-new	the declared objective the target level stands in for, and the indexing this note adds a level below	2ad1caa30e25f4f80e425b5dddacdbaa1a828a0a608f83aace04e9c6b312548a	f29524bdcf858c5bec6d1e2496e5721d2f225d83d519c5a70fa554200297c537
case-013bbf561b7940	notes/methodological-and-computational-closure-track-different-changes.md	notes/only-explicit-retention-is-durable-writable-and-addressable.md	85	grounds	d72c3bacae698086133feb71b3794a06e5bfd870a72d62e91109018a1488739b	a845671748ef6c1a8af02234192c0a590c4fd996006b93254c6b2b7fa28db359	primary	premised-on	grounds-baseline:P	candidate-new	the shared reason tacit expertise is not retained methodology and externalization is allocation's transport	2bdad9078ff85e7546bdaddaa62b907c55e8db83098ee8c4ce884272a25e1c98	ed9963bb01d6b950c795967200b138c1668477e5b6d99e620d4fd4d892f8b815
case-47d9b2d46aa6b4	notes/theory-and-methodology-form-a-two-layer-execution-system.md	notes/first-principles-reasoning-selects-for-explanatory-reach-over.md	70	grounds	0898abaec7793314ac7cb7e99f6da74b8f383666ab42a030277847f9d621a2f2	848d0ff02c9ed9d301a0fff8b6be5dfda7dbef8c32675268a2b6dcfd9d6bfd59	primary	premised-on	grounds-baseline:P	candidate-new	explanatory-reach is what the theory layer contributes and the methodology gives up	32db5a7bc28c48fd547f5af0b3eda3cf8abe7899a04df4061d9811eb81083b40	53029ddbc499e33bbaa8236107a4de2bd0b1bad2f0263ff11a5b2cde7c553929
case-fe5439ef0e0cac	notes/theory-mediated-self-improvement-needs-interpretation-and-retention.md	notes/formal-systems-assess-explanatory-reach-through-causal-and-proof.md	62	grounds	1bba5af761be25f7f1ef61be66d0f146bf6e83c3fde0bb3b9bd319dd2ed0b839	ed452c31accf03a098e59e124175173378b507ad19127e522b920fb7695f62bc	primary	premised-on	grounds-baseline:P	candidate-new	what a supplied formal language buys and costs	37e2f6369a49990033d3dba7175ff07bd04f37889aa4d9bf6167978217e7d0f3	9f760ea1ede28fe57b127194152918961933927779598978c2bec35a696b66a8
case-93e833a0ae496b	notes/selective-revision-needs-a-faithful-rationale-not-just-a-legible-one.md	notes/warranted-autonomy-is-bounded-by-oracle-domain.md	50	grounds	977b23906c14737d3ed126bf8f4f15b36a2e326c9c9e6f90f1401024dd3836e9	92dcb123e86fc3ab42570ebefe23ebf987537338e11aa4049c2fc42e5867fe4c	primary	premised-on	grounds-baseline:P	candidate-new	the domain over which the intervention test can be run bounds where the advantage is real	38bad5cf31483ed48279fa6c7143e078ea06f3474c78497f9cd258c8bd774884	d958b10613c2d5b32642af699891155a7ed134b7cf90e50541e796aac11c1e89
case-9a724e4374e080	notes/title-as-claim-exposes-commitments-enabling-popperian-maintenance.md	notes/mechanistic-constraints-make-popperian-kb-recommendations-actionable.md	23	grounds	254e87270c4217de54d98e3f77dc681428212322bd40e75a6a65401837e48fc9	7cd8c19e3e53d6f572dd9cb1376695066ae002c66470a43e71c9cb22b4fa789d	primary	premised-on	grounds-baseline:P	candidate-new	the Popperian framing that recommendations must be falsifiable to be useful	39124c1a4959b63872b4be2c2bce94e05e93affc4b99393c2d88661e10f082d1	bab2226232f55f4fbcba813cb5bf2553a7384c0bf5aacac9e06e4b47c4954f8b
case-1d940291ad46a8	notes/skills-are-instructions-plus-routing-and-execution-policy.md	notes/instruction-specificity-should-match-loading-frequency.md	94	grounds	94a651eba94676d6a76899ac35b137550dafbfc24284ad3b501c20ccbe865ca9	cca69b508b510cb03ea2bd0cb92e605011fca259db7e48059f6d0a37a3af6158	primary	premised-on	grounds-baseline:P	candidate-new	the loading hierarchy both participate in	3b907e90de0268d68d0edc08771d01ac84ef4733480f7d345ce0132e6e6ff44b	9dc12f66d46c550731bbc1a214c726fb5318ddca7c2d07af3d76cc726695e26e
case-e1177dbf35d92b	notes/technical-constraints-make-kb-objective-choice-engineering.md	notes/first-principles-are-inherited-constraints-not-design-choices.md	76	grounds	e01063a2551958a19b5274d473caebe175b6c1a509e817cd8cb10d1f71cf485d	cdc36a95a638319ad90288ae3b580ab655b8ae7ff89be4b6f17a6503d71be871	primary	premised-on	grounds-baseline:P	candidate-new	the constraints prune because they are inherited from boundary commitments, so the dissolution is framework-relative	3bf65bb460641ca250865e0500f69c5a25bc928a1e537fc3186f695e3855bac9	839e12301e3391c664c890274d2db29eadbb9916fd926fa01a7d708062529ba3
case-72ed741ccbd80f	notes/trace-extracted-memory-earns-authority-per-operation-not-at-capture.md	notes/agent-memory-requirements/use-trace-extraction-as-meta-learning.md	61	grounds	255f0803b866c8171d99e8582e1dddcc8ae030a19bbd2b4d484c4c0fc7439da2	f9fac6a72cd2bbb93ba4d1b91f109c35c7b423b79b074202eee370bbb8f43caf	primary	premised-on	grounds-baseline:P	candidate-new	the fail rung's signal-quality distinctions and candidate-status fields keep early-rung artifacts from being read as mature rules	3c146be2d87195f7e2875774696ee1d60211a6c92906d1b9c38c0e6c051b119e	8fc1a414002ea1b69da24030d47de20b8ecc49a3fee234508b13a4a4615bf6ca
case-6c50ade3801134	notes/agent-runtimes-decompose-into-scheduler-context-engine-and-execution.md	notes/inspectable-artifact-not-supervision-defeats-the-blackbox-problem.md	111	grounds	251708e6dde9271fe436c2fee04d71e5a58bef314a733e2d6131e0c5b043c026	eb162ba8a6810180eb289310985cfcf9612029988d35c5c865792c364ff00259	primary	premised-on	grounds-baseline:P	candidate-new	why repo artifacts and exact state belong to a separate substrate component	3e4de68339d592917f71a2896c2aa8e9d3c92f60cd539654b7c5380b73cc2e43	926d480ce53a59174f7a866a36400cab69a66310cae9abbd7a9bd36d3e813812
case-de38abeca27643	notes/the-practical-scheduler-is-the-host-language.md	notes/any-symbolic-program-with-llm-calls-is-a-select-call-program.md	67	grounds	03ced4d47e9e61e88272701b9c721a90ba7e67cffef9711612165c8cf51c9095	ffbb51d5bf5e65a11780cb23ed5b87f152d24a4a4f32f6d6a8bb30391b6508a5	primary	premised-on	grounds-baseline:P	candidate-new	why the host program already is the symbolic scheduler	400e1ec1f40c4f9cc75c078d0ddfa2929573a158886771a580d671b92b6347df	a36e64a14ce0104e4a4df3f7970c08dba33dd77537a4e048d5c6ed1146a878f8
case-f2af1a1b9045ca	notes/increasing-computational-autonomy-relocates-human-effort.md	notes/methodological-and-computational-closure-track-different-changes.md	54	grounds	e27776745a7d6a398c460423b132470a01b5be18f44c1887b13288f9191dd5af	a636df4c640b61cb85a019e35fe79f156b0d2483345c9edcee4ac46737a0cd10	primary	premised-on	grounds-baseline:P	candidate-new	identifies computational closure as an actor-allocation endpoint rather than methodological settlement	410edda52a89b855d8f756361b75527b73f4c0ed126306ab5a02c396ab51432a	ff3489c40eb4304ce6f8ef73708634e778f28fbf142862992f74bb8806a516cf
case-b940441aeb021c	notes/definitions/self-improving-system.md	notes/definitions/reflective-system.md	65	grounds	10d7780c0d1d426b77ac9c3f781df03da264fc15d2f3b20d5a012c8b1a9782c9	14843cc5bd4ea3c6c76108f1fef7d0345f96078fa521094c8909c2ff9ea26a09	primary	premised-on	grounds-baseline:P	candidate-new	the causally connected self-representation, and the intercession capability, that reflective self-improvement routes through	44ccfc28c8f3b925841cf9be17a2ff028ffd66f32ea3a7470542ea13a7391650	13b4f0f559f46ff9fb6a67cd5f876cf1af36f81df02baec50535bb70ccbc5b96
case-792c56f9b2a2f7	notes/first-principles-are-inherited-constraints-not-design-choices.md	notes/agent-context-is-constrained-by-soft-degradation-not-hard-token-limits.md	58	grounds	6c56748144768dda7f5a73c0c9256d733f4ab05b3d0e1281637a692a98f5ad5e	89aa6a9d40c19814675352f3dd1d6f966d62b63c2fa56858929d42ae7a1d20fd	primary	premised-on	grounds-baseline:P	candidate-new	the binding form of the context bound the economy answers to	4538e9353c18df8843976fbb98b152e6d51f552a91e1f7e561af607a0a08ed50	66100fe2403a02b9026c76a30609fba91cb42dcefafd95e55931efa1a2823401
case-4d03ee8de3d7c3	notes/fix-what-the-executor-cant-determine-not-what-it-will.md	notes/design-for-the-first-time-human-except-on-access-cost.md	26	grounds	2a564d64dc5a466f703bd4d9757fd12d50b5b9c6ebc46a584b0eca70acb2096a	155ea941b79b9feb36a8927eb4d0cdf7ad045ff360dcde38d4dce84bc303cb99	primary	premised-on	grounds-baseline:P	candidate-new	the competent-executor premise this claim rests on	46411037881006df28e391f04c5d4bcde0152ef4186f7052301c56f83a4150de	d8d9faf588a5176c2e35cb903de3406b8c03e8f1a5f9b68e0fc36bdcf6e79c5d
case-b5510a490b71bf	notes/a-derived-copy-of-recomputable-truth-must-be-checked-or-absent.md	notes/llm-recompute-cost-inverts-the-store-vs-recompute-default.md	72	grounds	bb79c38f3ad353c54ddd31330f76c7606a69d25f71b8c54b6cba9ce075079e20	ac9d89316633aa22812fc40342b4a185aab4e319fe5e13a69cd2686a8c1e1061	primary	premised-on	grounds-baseline:P	candidate-new	the value half — recompute being dear to a model is what makes a checked derived copy worth keeping in the first place	4720164d81c38c800ef86f2cadff78298ab92e2c7b650ed8722d5b16c3c4e1bf	0ba82417f316d5eab31b5630a05d8d877b47dedddc5840d6255713b33628fa74
case-d45ce3e679afe0	notes/commitment-not-derivation-creates-new-ground-truth.md	notes/agentic-systems-interpret-underspecified-instructions.md	88	grounds	47445cc55f1c0b508da07073250b8c9888a1f23e219ed919a2f0036073226cfb	7d44b64eb3f67bd091c2be601c3ef8fb8c07f48bca54ba772f91f3d351f956e8	primary	premised-on	grounds-baseline:P	candidate-new	the projection-versus-compilation distinction behind the generated-code case, including why temperature-zero determinism does not entail the additions	4826fd8214d61b8456bc88281c1961be47601d18ca5f3686235f087a29345ad1	5085a89fb9ef2ab828d9f3a811c771594b9da38411845ec067bb14402b93f0c9
case-a628538d6d8332	notes/definitions/reach-assessment.md	notes/first-principles-reasoning-selects-for-explanatory-reach-over.md	65	grounds	998da3f024180211e11fe05ed74ef565dc0fbd75191f623e3b5c2be1eb6fca79	4e7a3e4a65139f0da74a8f45e3214863d800abf911f5b996e5f0ad1a71ffb908	primary	premised-on	grounds-baseline:P	candidate-new	the four-part negative test this note asks a system's own evaluator to apply, not only a human author	49ca49f2b9cd699aa5820818ae95fb0f0d765dc4d49510ca19b1524eb458a353	3d73e5a3b1e0bb4beb147092bc3610832bdfaab1e8044cc0b212797612e2d4ff
case-04324ff043d41a	notes/methodological-and-computational-closure-track-different-changes.md	notes/a-proposal-selection-loop-requires-search-evaluation-and-retention.md	79	grounds	4a65f829b6b7d31e7b874ed4766712bfa5b818c287e391b7fdb46dd41166b422	8eae3fecdbc216628978a0e9270132307ce89fc72918ffc7b60169515c24b756	primary	premised-on	grounds-baseline:P	candidate-new	supplies the pathway functions over which both properties are reported	49eb1c947e9601c2a1374aa783a308d172de11461d709374b02dda8e7237869d	e3d303e1d74e58523057b856b4f48627d3278291500b964b51bf65124fa563a5
case-92aa92d533a0f1	notes/linking-theory.md	notes/agents-navigate-by-deciding-what-to-read-next.md	88	grounds	dbc7b352b95fa48e56ee229544bcc578825526add94b052e2e48f15bcb8890c7	551caec2cad8a3f1996320c9654bcb94f97718802438935f139a1c33ea7f1ac1	boundary-correction	premised-on	grounds-boundary	candidate-new	the navigation-decision model	e5c8c54f317267ad332b99482d39b11202bd8d9137945fb7b87a6dbea5ac9f69	ff3d6463c03b26223f77cbebdcc90e96b2c692ee961cba89f647eb419c6b88c0
case-26b9c135a0d7e2	notes/linking-theory.md	notes/title-as-claim-enables-traversal-as-reasoning.md	89	grounds	23cbd8f9c6bcaa282fc385926903c5777f92d503c95bbe83d3021e55d21ca7bd	f4c3dc2ef940711c86a77955adcde0ee2587c5b7648d95aca31149f89c00c309	boundary-correction	premised-on	grounds-boundary	candidate-new	claim titles as argument structure	dc50db4e96172d9af6a1fe8a24e7dcca49f0b854f257c55e15bd6ca339449918	11441c783ea5abcbc4e1c95f438720067e439463e39240204cbab2b3d88fa558
case-e42bba438fb9cb	notes/linking-theory.md	notes/title-as-claim-exposes-commitments-enabling-popperian-maintenance.md	90	grounds	11e37edcf5f7394b086fd11b5a3690808fefe2831b59fc3dd573efecf35b3f36	d4e3f722f272f934edf4613388f9c4f19b059bb92f771364ec36312a7cc73fa0	boundary-correction	premised-on	grounds-boundary	candidate-new	claim titles as maintenance surface	fe83b42ffb4ced4169e583d5fe6f039af26f3c610214de940ff7956b8d22435e	11c9774083f7af9b89f0c327a85001bdf7aa89e668ebd59fc121aad503f30afb
case-4f00efdb446dff	notes/linking-theory.md	notes/title-as-claim-makes-overlap-between-notes-visible.md	91	grounds	f4d0c6da58e1eab64a19b021dca3700d781d16eb7bbf1a8b9122deb70a6fdbf8	615319f901298007cb645bdcd681200249ac2ed5790e1ec02962834c1bcffa46	boundary-correction	premised-on	grounds-boundary	candidate-new	claim titles as redundancy detector	1f1033655a840bc4ec7612c70904477e0233a533bc5411bb89981a138924c4df	2608856bac47b19d16bf42d860b2b04dfea841b8fc3ff1cdf8c9ced499a20ef6
case-1220e2f6e7f005	notes/pointer-design-tradeoffs-in-progressive-disclosure.md	notes/agents-navigate-by-deciding-what-to-read-next.md	77	grounds	27613b99a005f75d3051536a9bacb40c4997b446af283e3f202cbfe7b58c6bb8	5be7115a04bbb41cb55446dd09b56d618d588ae3a80b3d9a9a46ecf65b81ce62	boundary-correction	premised-on	grounds-boundary	candidate-new	the navigation decision is what pointers optimize for; link phrases are the most context-specific pointer type for that decision	665ff3ef085065589637c92303498380d74c4975b5e5b519689a5c1251bc5682	c05d96045e67d3d0cdf510c75bc29048ae5af1eee27c8802740ef999cb994ca3
case-3cb813691b28b6	notes/retaining-the-episode-keeps-a-distilled-rule-re-derivable.md	notes/artifacts-produced-from-sources-need-lineage-recorded-at-the-source.md	46	grounds	b42d5352c661026c4d60aa8ba6a163002352dfe4134c4bd8837bf71ef374cf4c	9bebbeb088aba5a3f5886dfe77e4b13cee03263fa4b9f429138b1641d8dedbfc	boundary-correction	premised-on	grounds-boundary	candidate-new	the dependency record that keeps the pair a pair	e33393c0807a4b9fa047a35cbe6d2c5c501fa2b37943554887e45f2f600f948d	89b0e8dd652128717c8ee0b3d110afbfbe040ec17ef6e82e0452009d6cb001fc
case-394ae9372bce28	notes/short-composable-notes-maximize-combinatorial-discovery.md	notes/conjecture-is-seeing-the-particular-as-an-instance-of-the-general.md	53	grounds	94d7dcd2c9ac3cd33cf757ba3df023b505c5b4443e2f0378cb9c712e0c4cc070	4f068fcfc562a95a35839adb7d0fe70032943778eb84f6f0d260a50a0fec8d0c	boundary-correction	premised-on	grounds-boundary	candidate-new	discovery requires co-presence of multiple particulars	a94e8ba6c83d495a58b89e1df8e1629e2a9206cdd1ca844fdb90b4dfa4de9fd7	7ce5517115fa25238da0cdd0bb1498510a9e5217ba71c15f125b947b120934e8
case-e5c79d4a63f396	notes/title-as-claim-enables-traversal-as-reasoning.md	notes/agents-navigate-by-deciding-what-to-read-next.md	73	grounds	0ce528041ec90c1a4448e57d44ffd03575d6d8f0745b896c2715a1d5815e5518	6b9146e27b0e97b93f7e3fcde86127f25b82cf0154c75fa51dc23fee7e49853b	boundary-correction	premised-on	grounds-boundary	candidate-new	claim titles make the navigation decision cheap by carrying the argument in the pointer itself	4c96e5990ba888bdf5251a1a0fce9461903c15980626a14da43b5c4de229027a	a80595a7f139611bf575fe323a75f5f63cd3016b687b02e9f42a4a281bbedeac
case-5ca86f39e5b984	notes/constraining-and-extraction-both-trade-generality-for-reliability.md	notes/exact-implementation-does-not-validate-a-requirement.md	40	grounds	c1d9d020a9c8aefaaccd497045a63ffa82027bfcc836ec4ee4c97568fc6729c5	e23ea7884914b9154c48fc79cfd759457803b1a8e1698cdbe258dbfc202f1580	prior-drift	premised-on	grounds-drift:G009	candidate-new	distinguishes cases where narrowing can harden confidently from cases where relaxing may be needed	ba1e0ebb8a706d246322cb5eb1ad850aac2bb711dd671465aec4522e37da3ab6	a0601c8dfd81053b16c884b7405ff3e7548ec873865aa894f1a2d454d17ccd4d
case-30111db7b8e1ad	notes/machinery-persists-by-warrant-not-position-in-a-reflective-loop.md	notes/scheduler-llm-separation-exploits-an-error-correction-asymmetry.md	49	grounds	daecf73abdfd0f3064695a4eb6b0c72f5d3f6e1fcc98fbfcee48d73fc58de061	688a587e10bf1868fdb670a84ab7d7abb8bc9a21bb736610d773ab88553a1eb3	prior-drift	premised-on	grounds-drift:G001	candidate-new	the symbolic layer as learning target, with codification as the write path	795f46a4c59c0c554f1ca793cc7c2e8ea6beeb4f433664666b93c9c759416fbb	5294625f57a22fd3a7f3fae1f75033e61e76031af35089aa4e53f795a68a0a2c
case-f885ae7745ebc0	notes/machinery-persists-by-warrant-not-position-in-a-reflective-loop.md	notes/self-improvement-is-relative-to-a-declared-objective.md	50	grounds	4637b503f7eb74cf84af4906a3684e8a6360a2f44a680282f63ac54a39725736	ed967da5c1754b9e1c07f272414f976dac5bdd584852f7a6ef59c255dfff911c	prior-drift	premised-on	grounds-drift:G002	candidate-new	the function class that stays external by category	d381c8560c730abdf13fd6d25c914bf5ca9fb9777646fd9ea20500dd3135a542	47a6e0c4b710c23a1304f228c6c90b114b08cea0a16974c727000a85f2559ec3
case-0f41b0d385292b	notes/machinery-persists-by-warrant-not-position-in-a-reflective-loop.md	notes/warranted-autonomy-is-bounded-by-oracle-domain.md	51	grounds	7921cfe70f21b086586728c9de9b8fa2ad06839616578811082f0b9ef8349af4	bb6013a1298404268af87458032e72b85bbb23f93bdd50a1b98b098fd3a0a34f	prior-drift	premised-on	grounds-drift:G014	candidate-new	the governance that stands in for the fixed point	96e09ef983758b38da19f1b8a6b3c0a11a1a1e9ece8c4c47cf17d80810139f8e	ec3fae4f6aa63a7aaabc334664c40db6e694c5039db71c023872da385607fbf7
case-c44bc99dfa7da4	notes/scheduler-llm-separation-exploits-an-error-correction-asymmetry.md	notes/exact-implementation-does-not-validate-a-requirement.md	82	grounds	0e1fccf9c73b0fbed37fbeebeebf6883b4ba3a70cc25c456409fd56c2e800745	2c1773b0bfd4709bafd78165493f3084e77c3bcc812dc34a58b458243f51ae24	prior-drift	premised-on	grounds-drift:G016	candidate-new	the substrate executes its implemented function exactly; whether that function is the right one is the requirement-chain question this note delegates	826b88878c882dd4dbaf4fffd5b1390c7ad44ab4b32011cf25e946881a388e2a	d42d78eef5e100a7df258e1ab3ccac514f349cbf74f486da064fa6dcaeb8a62e
case-41428517ec8382	notes/agent-memory-needs-discoverable-composable-trusted-knowledge-under.md	notes/claw-learning-loops-must-improve-action-capacity-not-just-retrieval.md	36	grounds	bcbf0eda937bc478f7015b1c5d943d8c7299340242d819f779d625dac3f2e3c4	21a1058ca8c6a1cbaf28cfe1fda4bde1c24a94d14cf2c333df690db334ead97a	negative-control	explained-by	mechanism-core:F002	candidate-new	contextual competence is the success criterion for learned material, not retrieval accuracy	a11daee03e176260611a478c4c5f1c4545f2fa049fcca67ac6917fd62a5ab36d	045e76c2ce7d26240809800fb318c3c1761fc4ff7ce717232cb645b1d06a91d0
case-3496ee2db136b1	notes/brainstorming-maintainability-oracles-for-agentic-development.md	notes/weakly-discriminated-qualities-tend-to-be-underselected.md	188	grounds	c816592692ee5e513bbf4a7d76e0e0f8a6b89eac7f112151ac3808faad6a8e3e	86beed60bd0bce69cdb42ce069d2a957629a5b90a6d98b73db4aedf2b2437db9	negative-control	explained-by	mechanism-core:F114	candidate-new	the selection mechanism that motivates building a stronger maintainability oracle	2d7e22c2d3c135f98390be3a753765d8d4291a4befe37dcc324a8163f8d44d2d	0b79db94e94953afa4cd9fd41c03e1cc3d6e1cc02364c85d574f81025444f991
case-575b06979464eb	notes/the-bitter-lesson-selects-production-methods-not-representational.md	notes/scaling-absorbs-scaffolding-at-fixed-difficulty-not-at-the-frontier.md	71	grounds	d803db7e8d780d5f2d49b6d94ab7f473a8d44a255b176b5d941d718e23de5536	46b58e56db16e8faf58109870925e78c4f6efc5e01907e61d5de6645082a22ca	negative-control	explained-by	grounds-drift:G017	candidate-new	"the answer to ""the harness is temporary"""	14dfde637105f089c6e031294982bed793aa1c2eb13b11545d4dce2abd7a5edf	134daa0fa6cc4df197dadf9f793c1b8c43db8739430ef12801c198f2a8911ab2
case-082d18281753b9	notes/the-bitter-lesson-selects-production-methods-not-representational.md	notes/bitter-lesson-selects-against-unearned-reach-not-against-structure.md	68	grounds	3bac13649fe3126969999cabb582596d8ea31cd13671ce298153544f123eaffb	d7615ad19bccdbb3b7adf388374ab7064a921a5410095dee79863a1809f9d1ae	negative-control	explained-by	grounds-drift:G007	candidate-new	what the selection actually removes, which is what frees it from the form axis	697dc24c21b2baacd05c26aa052d462aa9aec767173f34f69bf293d33f70e67a	143da468d16e772fcd7f5b3df27034e24166899af71ccc6eb3fecb88885428b9
case-debb81f590eb94	notes/orchestration-strategies-and-run-state-have-opposite-persistence.md	notes/bounded-context-orchestration-model.md	46	grounds	eb93150e0575b335a15320a10454891ff5e6e1ea8da939086b5d9286d256e074	3591220b04f71416fff309922c111024e17dde902aff6e6344bd14e604390b80	negative-control	operates-through	mechanism-core:F129	candidate-new	the `select`/`K` decomposition whose two halves this note assigns opposite persistence economics	59a15a9c42ba03ed035ddc28a9f87be711825bc22a9dec1e3c9eba637f422173	05524e6beed784955a089b20d6a2ddab03ce7c0a0565d193626d0329f1f37bf8
case-5ea996e957b487	notes/llm-context-is-a-homoiconic-medium.md	notes/methodology-enforcement-is-constraining.md	45	grounds	4c97de8214b947f94730a7ef4ec98f8475a4c94fe139f3bd087b4096371340fe	693b1f4861b454740393f70a1795a729812ac2659c358c16e7725a13a8bfa741	negative-control	operates-through	mechanism-core:F072	candidate-new	the constraining gradient from instruction to script imposes structural boundaries that the homoiconic medium itself cannot provide	8eaa4cb53d4585c61db8aab328a0eb55cd900bd0cf0d023530fb5504fce3fd9d	10550cfddf83598d3d41ade7bc63a7ba2260ae65e8d3ac9c1c8bca1f1d1bacd6
case-cacf9c28c10bff	notes/evaluation-automation-is-phase-gated-by-comprehension.md	notes/spec-mining-as-codification.md	50	grounds	cd1a3b8918e49d2818116ad7f265af913c60aa4ff1f9d23585475c370ca6b5e8	6c11fc2f96b753a40d601186b34e2dc5787b5244c921f89f444f263ed3dd0944	negative-control	operates-through	mechanism-core:F108	candidate-new	converting observed failures into reusable evaluators is spec mining	3402f2e7bad94e4a44996ce4a06520d50a3fcdfc5c31b10596ed73aed2fc6e74	2038c37b5f34d16f317fec45ede15a85fc317aa452c259894141721bc029ed05
case-33b17d160ff4f4	notes/ad-hoc-prompts-extend-the-system-without-schema-changes.md	notes/llm-context-is-composed-without-scoping.md	64	grounds	ed2813fd3a89dacb63894abbb007600fdeeaa6cb6ca2277e7d1038abee83c54a	2a94386b8eb496c82bb11068d56fa819ca072c15ad883932d2bd29133ca54d26	negative-control	operates-through	mechanism-core:F038	candidate-new	ad hoc instructions notes are effective sub-agent interfaces because they provide lexically scoped frames — the sub-agent sees only what the caller explicitly passed	e544be2403e207e349c30f63846a740f0a7a38a24e979479f8dfb80e8af255b8	29285e4948862ac84465709acf8b0b99c5a712c87c98eccf8f46a22e9e170a55
case-bac2e954ce6f0b	notes/retaining-the-episode-keeps-a-distilled-rule-re-derivable.md	notes/reflective-coverage-is-graded-across-representational-forms.md	51	grounds	b13969e163db66d01d459208be6d25ce2ef267101e90f563afbbdb5d2fb65a63	e1eeb79a24a510397b8159163337869f7b7adfd825a884576dc8c0d603df9175	negative-control	extends	grounds-boundary	authorized	the profile vocabulary that states episode replay's selection-grade dependency on the reader model	23f3eed2f8fd8827dc8cd2c9985e8da8ba2827dce25b4f194e5821548077f68f	0cd9005d8617236f1929d7ba6acd49727b55097e6fc13033d1e5b3d782b43494
case-cbbd1d7667e81e	notes/only-derivation-and-inheritance-warrant-a-scope-claim-use-earns-it.md	notes/bitter-lesson-selects-against-unearned-reach-not-against-structure.md	66	grounds	8e180f6fd91258dfed59762f18cec794f13177ef9ac9b452ce817416ce09bbe1	253c2aedbff45a0f76949bff2bf13aee3608c6308557e39d042d0c3420654445	negative-control	extends	grounds-drift:G012	authorized	identifies asserted-versus-tested scope as the property on which scale selects, and supplies the search loop through which free-born decompositions must earn scope extensionally	250a7ed78621c38fffe339983e89e76aefa28162e53a64c0f5199553fc4626b1	128c7f23a024f333c2486194f52fd74b3e87669961cc9121afbf2089b8a70371
case-68854084829173	notes/commitment-not-derivation-creates-new-ground-truth.md	notes/artifacts-produced-from-sources-need-lineage-recorded-at-the-source.md	94	grounds	65173fb26db057d384c688b5e36ada20c0dbb0779c1f17381dfcb0fc38a14f83	536e338dd654b82b174a4bef8334b71a4063cd80a6947964bc7377bb95ceb3f7	negative-control	extends	grounds-boundary	authorized	the managed-staleness machinery that governs derivation's judgment-checked zone	53ef46d6cf203fb38530b7e17c690ea224eecb4505f8d6d642ec58a5b9e2cf38	12c99174ff71d82ec0af9d75db464de3f0a429dd5956f52895c535635ca0a314
case-3f238a6bfb20f0	notes/false-positive-generation-is-filtered-before-retention.md	notes/a-proposal-selection-loop-requires-search-evaluation-and-retention.md	67	grounds	72f5d70d28a91388ed4e5a1835d403b020b1c3a47d7427a47a112e53b0c03eb6	500bb06a3f01760ccc5a973432a2980c6f680f8e46c21682c7bc3ed693d5249f	negative-control	extends	grounds-boundary	authorized	the loop whose functions this claim distinguishes, and the search-range-versus-oracle-strength point it extends	4df81d99bba0a3c0889a5816054714d7101ae09c77da4b7c160d1773fb46f0f1	1429f889ace1678385f4f3afca8fd48c7113f16c8cd2f429f4d2aebef9244237
case-a0e84e8b184274	notes/the-bitter-lesson-selects-production-methods-not-representational.md	notes/the-meta-harness-ablation-bounds-summarization-not-theory-formation.md	75	grounds	c4cc6dfbf9604338f65ea2bf7e9ccb6816b62a3a87eeb2949d7788d4df57d006	5c04eecbd086877571aaf58a872236aa545855e7a90b88a3b6355afb70eb2344	negative-control	evidenced-by	grounds-drift:G021	authorized	the precise reading of the strongest fourth-quadrant fragment	49949880d3fefdc152886b19200aae34b16e71cf0869660d9981ce5666fb46b1	0366851c8c4ed7ebdaf708ffde8ffc25e3ec61a5a20138b9e7445e147b01e924
case-7434c2dafbccda	notes/compiling-coordination-preserves-primitive-not-aggregate-authority.md	notes/rlm-has-the-model-write-ephemeral-orchestrators-over-sub-agents.md	61	grounds	d2ac722c38cab2a8739613a9fb9edf065dd72d279a841f493ad2b806995d0889	c0ea6f6344acac5a1314bf890a32ca9702ebde9f1905d45c5e262634798c5cb7	negative-control	evidenced-by	mechanism-boundary:F095	authorized	supplies the model-writes-the-orchestrator distinction that dynamic workflows reuse in a sandboxed workflow surface	e92351dc4638234e2609c367cf4ae818f432ffa60945a1807cdea3de00591f71	4a73ef6bd883a719db37fa12a48ad7d082a691bb4113453e80aaab4dec23b2ac
case-b43b0ab7700ffe	notes/only-derivation-and-inheritance-warrant-a-scope-claim-use-earns-it.md	notes/definitions/reflective-system.md	70	grounds	218c626cab1776435382bfb17f15f4c5e1a683529b1865b21bd0b95ea23c3ad1	fa860f7818f0adf6e48465f9ee4ad874d71501c39abb1c614f34033472b11b57	negative-control	evidenced-by	grounds-drift:G019	authorized	provides a worked instance of transferred warrant, in which the inherited criteria and local extension are separated in an auditable provenance record	800963352cfa38fbfe81f9708021f27a5b62134375da07d47f897aab496dc4ba	94dd5abe6cbe75128d43594d5d2b3071fcf9804bd2af2f404fdbb2f6bd9df31a
case-10fea94cddfcc6	notes/self-improvement-is-relative-to-a-declared-objective.md	notes/real-self-improving-systems-occupy-combinations-no-rung-captures.md	62	grounds	e31f8eb80ffff0cbdbc354c6aa89fa22ff5c40cc0dc9496292b36fd5814dc147	9c4b1d7501a2583e598a8d4f64e0e37c11838ca0a0277aaf83301179495f194d	negative-control	evidenced-by	grounds-boundary	authorized	the combinations that actually occur, which is why no ordering follows from the profile	4fb6346aabcbfe012a640a45f13987203e71beaa426a80f909822073cdb16dea	a0c4f5cb4ca344846557d78835f446ae3f78f4412190ea8c6c0e4fcfbe78c1c5