[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71CC4T48VSPX2A9RVFFAP43":3,"skill-same:01M71CC4T48VSPX2A9RVFFAP43":79},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":57,"commitSha":58,"skillDirectory":59,"forksCount":60,"repositoryHtmlUrl":61,"repositoryHomepageUrl":6,"ownerHtmlUrl":62,"ownerBlogUrl":6,"ownerDescription":63,"updatedAt":64,"install":65,"resources":78},"01M71CC4T48VSPX2A9RVFFAP43","quantum-hypothesis-simulation",null,"Use when evidence arrives in an ORDER that matters, when hypotheses interfere\u002Fare non-commuting, when you need the Schmidt coupling criterion for the cross-reference engine, or when a classical Bayes model may be missing order-effects — quantum-hypothesis-tracker.mjs does real-valued Hilbert-space hypothesis tracking (superposition until measurement, order-aware sequential evidence, QQ-equality), gated to beat classical Bayes on order-sensitive data without spurious control wins. Tier 3 of YURI's 4-tier simulation arsenal.","nexuslinkproductions\u002Fyuri-os","nexuslinkproductions","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F273246263?v=4","01M718YX9NZRCW50B59B8ZVDKN","Marcel Spatz","01M718YXAQ7530F4B1VKN6SFQV",2,"MIT","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fquantum-hypothesis-simulation\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"quantum","hypothesis","quantum hypothesis simulation","simulation","missing","coupling","for","evidence","harness","measurement","data","qq","tracker","superposition","you","when","effects","decide","model","effect","2026-11-10 10:12:30",false,"---\nname: quantum-hypothesis-simulation\ndescription: Use when evidence arrives in an ORDER that matters, when hypotheses interfere\u002Fare non-commuting, when you need the Schmidt coupling criterion for the cross-reference engine, or when a classical Bayes model may be missing order-effects — quantum-hypothesis-tracker.mjs does real-valued Hilbert-space hypothesis tracking (superposition until measurement, order-aware sequential evidence, QQ-equality), gated to beat classical Bayes on order-sensitive data without spurious control wins. Tier 3 of YURI's 4-tier simulation arsenal.\ntriggers:\n  - \"\u002Fquantum-sim\"\n  - \"\u002Fqsim\"\n  - \"quantum sim\"\n  - \"quantum simulation\"\n  - \"order effect\"\n  - \"hypothesis superposition\"\n  - \"quantum vs bayes\"\n  - \"schmidt coupling\"\n  - \"non-commuting evidence\"\n  - \"simulation arsenal\"\n  - \"decide under uncertainty\"\nscope: harness\ninvocation: ability\n---\n\n# Quantum Hypothesis Simulation\n\nThe quantum-probability layer for YURI's claim\u002Fpulse machinery, and **tier 3 of the simulation arsenal**. It models hypotheses as a **superposition** in a real-valued Hilbert space (ℝ^N) that only **projects** (collapses) when evidence is applied — so the ORDER evidence arrives in changes the posterior. That is the one thing a classical order-blind Bayes update structurally cannot represent, and it is the whole point: when `P(H | A then B) ≠ P(H | B then A)`, this is the right instrument.\n\nEngine: `_SYSTEM\u002FScripts\u002Fquantum-hypothesis-tracker.mjs` (313 lines, 19 exports). Falsification gate: `_SYSTEM\u002FScripts\u002Fquantum-vs-bayes-benchmark.mjs`. **GATE re-verified PASS 2026-06-14** on current code (G1–G4 all true; tracker unit tests 5\u002F5; eval-processing 11\u002F11; validated at a 15M-eval scale). Wired as an instrument of `probabilistic-decision-core`. **Living system — actively being built**; the public export API is stable (method-map below intact) but internals + the surrounding arsenal keep moving, so re-read the source before relying on a fine detail.\n\n## The simulation arsenal (where this sits) — `faeb5b67` (arsenal) → `15cfc088` (learn-loop closed)\n\nA decision-under-uncertainty pipeline: **measure → robustify → commit → learn.**\n\n1. **`izanagi-simulator`** (`\u002Fizanagi`) — qualitative 3-branch commit (EV × reversibility × blast). The fast \"should I even branch\" front.\n2. **`decision-sim.mjs`** — robust optimization: `robustScore` (0.5·mean + 0.5·CVaR), `minimaxRegret`, `pgdWitness` (flip-rule), `crossEntropyOptimize`, `infoGapHorizon`, `multiverse`, + `halton`\u002F`sobolish` QMC. Reproducible (seeded `makeRng`). 7\u002F7 green.\n3. **`quantum-hypothesis-tracker.mjs`** ← *you are here* — order-aware evidence (below).\n4. **`izanagi-bridge.mjs`** (`@capability: izanagi-decision-bridge`) — turns measured\u002FMC option values into a robust ruling: CVaR + minimax-regret + PGD **with an AUTOMATIC corner-law vertex guard** (enumerates paramSpace vertices where an affine\u002Fmultilinear worst case hides from interior sampling) + per-axis flip thresholds. Demo-validated (`runDemo`), no unit test yet. This is izanagi made *computational*.\n   - Learn-loop edge: `izanagiRuling(…, {record:true})` writes the prediction to `prediction-ledger.mjs` (opt-in, off by default). **CLOSED 2026-06-13 (`15cfc088`):** `prediction-outcome-resolver.mjs` (`@capability`) re-runs propagation-scan on aged unresolved predictions, scores predicted-vs-observed, records outcomes + populates calibration — idempotent, execFileSync-safe, REPO_ROOT-anchored, live-verified (resolved=1, calibration n=2) + guard test. **Remaining edge:** a SCHEDULED age-sweep cadence (the resolver runs on-demand; a cron\u002Flaunchd beat to sweep aged predictions is owner-config).\n\nReach for **`izanagi-bridge`** for \"which option do I build \u002F commit under uncertainty\"; reach for **this quantum tier** when the *evidence order* itself carries information.\n\n## Use When (quantum tier specifically)\n\n- Evidence is **sequential and order-sensitive** — a later observation reframes an earlier one (question-order, framing, anchoring, path-dependent diagnosis).\n- Hypotheses **interfere** \u002F are non-commuting (the projectors don't commute), so a product-of-likelihoods Bayes update loses information.\n- You need the **Schmidt coupling test** — the cross-reference engine's mathematical criterion for whether two subsystems are genuinely coupled vs separable.\n- You suspect a classical model is silently missing an order-effect and want a falsifiable test of it.\n\n## Skip When\n\n- Evidence is order-independent and exchangeable → plain Bayes (`bayesSequential`) is correct and cheaper; don't reach for quantum to look sophisticated.\n- You want robust *optimization* over options, not order-effects → that's `decision-sim` \u002F `izanagi-bridge`.\n- The phases need to be complex (full ℂ^N interference) — this engine is **ℝ^N only**, phases 0 or π. Adequate for real-valued projectors; not a complex-interference model.\n\n## Method → question map (quantum tracker)\n\n| You want… | Function | What it gives |\n|---|---|---|\n| Posterior after an ORDERED evidence sequence | `hypothesisPosteriors(state, hypotheses, evidence)` | applies each projector in order (collapsing the state), then reads P(H) — order-dependent |\n| The raw sequential measurement | `measureSequential(projectors, psi)` | the order-effect core: project, renormalize, repeat |\n| The falsifiable order-effect signature | `qqEquality(state, P_A, P_B)` | QQ statistic `sAB − sBA`; the quantum model guarantees ≈ 0 — the Wang-Busemeyer test |\n| Are two subsystems genuinely COUPLED | `schmidtDecomposition(psi_AB, m, n)` | Schmidt spectrum (Jacobi SVD) — the cross-ref engine's coupling criterion (separable ⇒ uncoupled) |\n| The classical baseline you must BEAT | `bayesPosterior` \u002F `bayesSequential` | order-blind posterior — the control |\n| Run the whole falsification gate | `runBenchmark()` (benchmark module) | G1–G4 verdict |\n\nPlus the ℝ^N primitives (`dot`\u002F`norm`\u002F`normalize`\u002F`matVec`\u002F`matMul`\u002F`projector`\u002F`diagonalProjector`\u002F`measure`\u002F`stateVector`). Pure, dependency-free, owner-gate-free — run as a throwaway harness or wire into a sim test.\n\n## Eval-processing — turning a million-eval budget into a decision (`eval-processing.mjs`, NEW)\n\n**Reframe (the juicy part):** an eval count is a sampling *budget*, not a result. Needing millions usually means you allocated badly — good design resolves most decisions at **thousands**; millions are only for rare tails, fine grids, or hi-dim corners. \"Processing\" collapses N evals → one decision + flip rule + residual CI. The funnel (4 capabilities, **11\u002F11 green** as of `15cfc088`, all surface via capability-recall); `15cfc088` added residuals — `pairedDelta` `computeUnpaired:false` (skip the 2× baseline in hot loops) + `confidenceSequence` sub-Gaussian fallback (`range:null` for unbounded data, α-control validated on a Gaussian null):\n\n- `mkAggregator` (**streaming-aggregator**) — Welford moments + Vitter reservoir, O(1) memory, never store the rows.\n- `pairedDelta` (**crn-paired-delta**) — common-random-numbers → variance-reduced A\u002FB delta CI (the biggest free win; computes the unpaired CI too as proof).\n- `confidenceSequence` \u002F `sequentialDecide` (**sequential-stopping**) — empirical-Bernstein confidence sequence, peek-valid (α-control empirically validated, not just cited; distinct from quantum `measureSequential`).\n- `heldOutSplit` \u002F `kFold` \u002F `conformalQuantile` \u002F `inSampleVsHeldout` (**heldout-split**) — kills the in-sample ECE=0 optimism trap.\n\nThe CI width + flip location are what made the 16M-eval cap-sim honest — not the count.\n\n## The proof discipline (this is the precision, not decoration)\n\nA model that \"wins everywhere\" is overfitting, not an effect. Gates are domain-blind and **two-sided** — both halves must hold:\n\n- **G2 (earns its keep):** on real order-effect data (Gallup Clinton\u002FGore marginals: Clinton 50→57%, Gore 68→60%) quantum RMS must be **≥50% below** the best classical static model.\n- **G3 (honesty\u002Fcontrol):** on synthetic NO-order-effect data quantum must **NOT** beat classical by more than 0.005 RMS. A win on the control = laundering flexibility as signal → reject.\n- **G1** machinery recovery; **G4** QQ residual on the real joint \u003C 0.05.\n- **DATA FLAG:** the Clinton\u002FGore **JOINT** cells are literature-recalled (Wang & Busemeyer 2013) → `OWNER-VERIFY`; the **marginals** are robustly attested and carry G2 on their own.\n- **Corner law (automated downstream):** for an affine\u002Fmultilinear objective the worst case sits at a **VERTEX** an interior Dirichlet\u002Funiform sampler hits with probability 0 — `izanagi-bridge.cornerAwareReadout` enumerates vertices so a flip can't hide. See `[[feedback-affine-objective-enumerate-corners]]`.\n- **Promotion path:** must beat the classical Bayes baseline on order-sensitive *logged* sequences before touching any live organ. Until then, advisory.\n\n## Boundaries\n\n- Output is advisory until verified against live evidence — the falsification gate IS the verification; don't claim a quantum win without the two-sided G2+G3 pass.\n- ℝ^N real-valued only; don't claim complex-interference behavior it doesn't have.\n- The arsenal is **still being built** — value models \u002F bridge constants are hand-encoded (advisory, not measured truth); the learn loop is **closed on-demand** (`prediction-outcome-resolver` scores predictions→outcomes), with only a SCHEDULED sweep cadence still owner-config.\n\n## Pair with\n\n- **`izanagi-bridge.mjs`** (`@capability: izanagi-decision-bridge`) — the measure→robustify→commit step; corner-law-guarded ruling from option values.\n- **`decision-sim.mjs`** — robust optimization tier (CVaR \u002F regret \u002F flip-rule \u002F info-gap \u002F multiverse).\n- **`eval-processing.mjs`** — the funnel that turns a quantum\u002FMC sim's evals into an honest decision+CI.\n- **`prediction-outcome-resolver.mjs`** (`@capability`) — closes the learn loop: scores recorded predictions against observed outcomes (re-runs propagation-scan), populates calibration. Pairs with `izanagi-bridge`'s `{record:true}` ledger writes.\n- **`probabilistic-decision-core`** (`\u002Fpdc`) — the EV\u002Fcalibration discipline the tracker sharpens.\n- **`cross-reference-navigation`** (`\u002Fxref`) — the Schmidt coupling test is the cross-ref engine's coupling criterion; use together when judging whether two mechanisms are truly linked.\n- **`izanagi-simulator`** (`\u002Fizanagi`) — the qualitative 3-branch front.\n- Recall surface: memory `[[ref-simulation-arsenal]]` (the 4-tier map + roadmap) · `[[ref-capability-first-wiring]]`.\n\n## Session Notes\n\n### 2026-06-14\n- session: 101m | peak ctx: 0% | compacts: 0\n- tools: Bash×455, Read×142, Edit×74, Write×44, WebSearch×31, WebFetch×19, StructuredOutput×17, ToolSearch×8, TodoWrite×5, Workflow×3, AskUserQuestion×2, Skill×2, TaskStop×1\n- corrections: Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fcross-reference-navigation\n\n# Cross-Reference Navigation (XREF)\n\nThe GROUND step of the work loop, made reflexive. One question asked a | Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fquantum-hypothesis-simulation\n\n# Quantum Hypothesis Simulation\n\nThe quantum-probability layer for YURI's claim\u002Fpulse machinery. It mode | Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fcross-reference-navigation\n\n# Cross-Reference Navigation (XREF)\n\nThe GROUND step of the work loop, made reflexive. One question asked a\n- errors: none\n","# Quantum Hypothesis Simulation\n\nThe quantum-probability layer for YURI's claim\u002Fpulse machinery, and **tier 3 of the simulation arsenal**. It models hypotheses as a **superposition** in a real-valued Hilbert space (ℝ^N) that only **projects** (collapses) when evidence is applied — so the ORDER evidence arrives in changes the posterior. That is the one thing a classical order-blind Bayes update structurally cannot represent, and it is the whole point: when `P(H | A then B) ≠ P(H | B then A)`, this is the right instrument.\n\nEngine: `_SYSTEM\u002FScripts\u002Fquantum-hypothesis-tracker.mjs` (313 lines, 19 exports). Falsification gate: `_SYSTEM\u002FScripts\u002Fquantum-vs-bayes-benchmark.mjs`. **GATE re-verified PASS 2026-06-14** on current code (G1–G4 all true; tracker unit tests 5\u002F5; eval-processing 11\u002F11; validated at a 15M-eval scale). Wired as an instrument of `probabilistic-decision-core`. **Living system — actively being built**; the public export API is stable (method-map below intact) but internals + the surrounding arsenal keep moving, so re-read the source before relying on a fine detail.\n\n## The simulation arsenal (where this sits) — `faeb5b67` (arsenal) → `15cfc088` (learn-loop closed)\n\nA decision-under-uncertainty pipeline: **measure → robustify → commit → learn.**\n\n1. **`izanagi-simulator`** (`\u002Fizanagi`) — qualitative 3-branch commit (EV × reversibility × blast). The fast \"should I even branch\" front.\n2. **`decision-sim.mjs`** — robust optimization: `robustScore` (0.5·mean + 0.5·CVaR), `minimaxRegret`, `pgdWitness` (flip-rule), `crossEntropyOptimize`, `infoGapHorizon`, `multiverse`, + `halton`\u002F`sobolish` QMC. Reproducible (seeded `makeRng`). 7\u002F7 green.\n3. **`quantum-hypothesis-tracker.mjs`** ← *you are here* — order-aware evidence (below).\n4. **`izanagi-bridge.mjs`** (`@capability: izanagi-decision-bridge`) — turns measured\u002FMC option values into a robust ruling: CVaR + minimax-regret + PGD **with an AUTOMATIC corner-law vertex guard** (enumerates paramSpace vertices where an affine\u002Fmultilinear worst case hides from interior sampling) + per-axis flip thresholds. Demo-validated (`runDemo`), no unit test yet. This is izanagi made *computational*.\n   - Learn-loop edge: `izanagiRuling(…, {record:true})` writes the prediction to `prediction-ledger.mjs` (opt-in, off by default). **CLOSED 2026-06-13 (`15cfc088`):** `prediction-outcome-resolver.mjs` (`@capability`) re-runs propagation-scan on aged unresolved predictions, scores predicted-vs-observed, records outcomes + populates calibration — idempotent, execFileSync-safe, REPO_ROOT-anchored, live-verified (resolved=1, calibration n=2) + guard test. **Remaining edge:** a SCHEDULED age-sweep cadence (the resolver runs on-demand; a cron\u002Flaunchd beat to sweep aged predictions is owner-config).\n\nReach for **`izanagi-bridge`** for \"which option do I build \u002F commit under uncertainty\"; reach for **this quantum tier** when the *evidence order* itself carries information.\n\n## Use When (quantum tier specifically)\n\n- Evidence is **sequential and order-sensitive** — a later observation reframes an earlier one (question-order, framing, anchoring, path-dependent diagnosis).\n- Hypotheses **interfere** \u002F are non-commuting (the projectors don't commute), so a product-of-likelihoods Bayes update loses information.\n- You need the **Schmidt coupling test** — the cross-reference engine's mathematical criterion for whether two subsystems are genuinely coupled vs separable.\n- You suspect a classical model is silently missing an order-effect and want a falsifiable test of it.\n\n## Skip When\n\n- Evidence is order-independent and exchangeable → plain Bayes (`bayesSequential`) is correct and cheaper; don't reach for quantum to look sophisticated.\n- You want robust *optimization* over options, not order-effects → that's `decision-sim` \u002F `izanagi-bridge`.\n- The phases need to be complex (full ℂ^N interference) — this engine is **ℝ^N only**, phases 0 or π. Adequate for real-valued projectors; not a complex-interference model.\n\n## Method → question map (quantum tracker)\n\n| You want… | Function | What it gives |\n|---|---|---|\n| Posterior after an ORDERED evidence sequence | `hypothesisPosteriors(state, hypotheses, evidence)` | applies each projector in order (collapsing the state), then reads P(H) — order-dependent |\n| The raw sequential measurement | `measureSequential(projectors, psi)` | the order-effect core: project, renormalize, repeat |\n| The falsifiable order-effect signature | `qqEquality(state, P_A, P_B)` | QQ statistic `sAB − sBA`; the quantum model guarantees ≈ 0 — the Wang-Busemeyer test |\n| Are two subsystems genuinely COUPLED | `schmidtDecomposition(psi_AB, m, n)` | Schmidt spectrum (Jacobi SVD) — the cross-ref engine's coupling criterion (separable ⇒ uncoupled) |\n| The classical baseline you must BEAT | `bayesPosterior` \u002F `bayesSequential` | order-blind posterior — the control |\n| Run the whole falsification gate | `runBenchmark()` (benchmark module) | G1–G4 verdict |\n\nPlus the ℝ^N primitives (`dot`\u002F`norm`\u002F`normalize`\u002F`matVec`\u002F`matMul`\u002F`projector`\u002F`diagonalProjector`\u002F`measure`\u002F`stateVector`). Pure, dependency-free, owner-gate-free — run as a throwaway harness or wire into a sim test.\n\n## Eval-processing — turning a million-eval budget into a decision (`eval-processing.mjs`, NEW)\n\n**Reframe (the juicy part):** an eval count is a sampling *budget*, not a result. Needing millions usually means you allocated badly — good design resolves most decisions at **thousands**; millions are only for rare tails, fine grids, or hi-dim corners. \"Processing\" collapses N evals → one decision + flip rule + residual CI. The funnel (4 capabilities, **11\u002F11 green** as of `15cfc088`, all surface via capability-recall); `15cfc088` added residuals — `pairedDelta` `computeUnpaired:false` (skip the 2× baseline in hot loops) + `confidenceSequence` sub-Gaussian fallback (`range:null` for unbounded data, α-control validated on a Gaussian null):\n\n- `mkAggregator` (**streaming-aggregator**) — Welford moments + Vitter reservoir, O(1) memory, never store the rows.\n- `pairedDelta` (**crn-paired-delta**) — common-random-numbers → variance-reduced A\u002FB delta CI (the biggest free win; computes the unpaired CI too as proof).\n- `confidenceSequence` \u002F `sequentialDecide` (**sequential-stopping**) — empirical-Bernstein confidence sequence, peek-valid (α-control empirically validated, not just cited; distinct from quantum `measureSequential`).\n- `heldOutSplit` \u002F `kFold` \u002F `conformalQuantile` \u002F `inSampleVsHeldout` (**heldout-split**) — kills the in-sample ECE=0 optimism trap.\n\nThe CI width + flip location are what made the 16M-eval cap-sim honest — not the count.\n\n## The proof discipline (this is the precision, not decoration)\n\nA model that \"wins everywhere\" is overfitting, not an effect. Gates are domain-blind and **two-sided** — both halves must hold:\n\n- **G2 (earns its keep):** on real order-effect data (Gallup Clinton\u002FGore marginals: Clinton 50→57%, Gore 68→60%) quantum RMS must be **≥50% below** the best classical static model.\n- **G3 (honesty\u002Fcontrol):** on synthetic NO-order-effect data quantum must **NOT** beat classical by more than 0.005 RMS. A win on the control = laundering flexibility as signal → reject.\n- **G1** machinery recovery; **G4** QQ residual on the real joint \u003C 0.05.\n- **DATA FLAG:** the Clinton\u002FGore **JOINT** cells are literature-recalled (Wang & Busemeyer 2013) → `OWNER-VERIFY`; the **marginals** are robustly attested and carry G2 on their own.\n- **Corner law (automated downstream):** for an affine\u002Fmultilinear objective the worst case sits at a **VERTEX** an interior Dirichlet\u002Funiform sampler hits with probability 0 — `izanagi-bridge.cornerAwareReadout` enumerates vertices so a flip can't hide. See `[[feedback-affine-objective-enumerate-corners]]`.\n- **Promotion path:** must beat the classical Bayes baseline on order-sensitive *logged* sequences before touching any live organ. Until then, advisory.\n\n## Boundaries\n\n- Output is advisory until verified against live evidence — the falsification gate IS the verification; don't claim a quantum win without the two-sided G2+G3 pass.\n- ℝ^N real-valued only; don't claim complex-interference behavior it doesn't have.\n- The arsenal is **still being built** — value models \u002F bridge constants are hand-encoded (advisory, not measured truth); the learn loop is **closed on-demand** (`prediction-outcome-resolver` scores predictions→outcomes), with only a SCHEDULED sweep cadence still owner-config.\n\n## Pair with\n\n- **`izanagi-bridge.mjs`** (`@capability: izanagi-decision-bridge`) — the measure→robustify→commit step; corner-law-guarded ruling from option values.\n- **`decision-sim.mjs`** — robust optimization tier (CVaR \u002F regret \u002F flip-rule \u002F info-gap \u002F multiverse).\n- **`eval-processing.mjs`** — the funnel that turns a quantum\u002FMC sim's evals into an honest decision+CI.\n- **`prediction-outcome-resolver.mjs`** (`@capability`) — closes the learn loop: scores recorded predictions against observed outcomes (re-runs propagation-scan), populates calibration. Pairs with `izanagi-bridge`'s `{record:true}` ledger writes.\n- **`probabilistic-decision-core`** (`\u002Fpdc`) — the EV\u002Fcalibration discipline the tracker sharpens.\n- **`cross-reference-navigation`** (`\u002Fxref`) — the Schmidt coupling test is the cross-ref engine's coupling criterion; use together when judging whether two mechanisms are truly linked.\n- **`izanagi-simulator`** (`\u002Fizanagi`) — the qualitative 3-branch front.\n- Recall surface: memory `[[ref-simulation-arsenal]]` (the 4-tier map + roadmap) · `[[ref-capability-first-wiring]]`.\n\n## Session Notes\n\n### 2026-06-14\n- session: 101m | peak ctx: 0% | compacts: 0\n- tools: Bash×455, Read×142, Edit×74, Write×44, WebSearch×31, WebFetch×19, StructuredOutput×17, ToolSearch×8, TodoWrite×5, Workflow×3, AskUserQuestion×2, Skill×2, TaskStop×1\n- corrections: Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fcross-reference-navigation\n\n# Cross-Reference Navigation (XREF)\n\nThe GROUND step of the work loop, made reflexive. One question asked a | Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fquantum-hypothesis-simulation\n\n# Quantum Hypothesis Simulation\n\nThe quantum-probability layer for YURI's claim\u002Fpulse machinery. It mode | Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fcross-reference-navigation\n\n# Cross-Reference Navigation (XREF)\n\nThe GROUND step of the work loop, made reflexive. One question asked a\n- errors: none\n",{"name":5,"scope":27,"triggers":44,"invocation":56,"description":7},[45,46,47,48,49,50,51,52,53,54,55],"\u002Fquantum-sim","\u002Fqsim","quantum sim","quantum simulation","order effect","hypothesis superposition","quantum vs bayes","schmidt coupling","non-commuting evidence","simulation arsenal","decide under uncertainty","ability","main","4c6d8d65f3d6096e2fc25b0845e481dda73c2794","skills\u002Fquantum-hypothesis-simulation",0,"https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions","Building deterministic AI governance and continuity systems. Workflow architect + visual storyteller","2026-11-11 16:42:59",{"source":8,"skillName":5,"commands":66,"prompt":76,"zipAvailable":77},[67,70,73],{"tool":68,"command":69},"npx","npx skills add nexuslinkproductions\u002Fyuri-os --skill quantum-hypothesis-simulation",{"tool":71,"command":72},"bunx","bunx skills add nexuslinkproductions\u002Fyuri-os --skill quantum-hypothesis-simulation",{"tool":74,"command":75},"pnpm","pnpm dlx skills add nexuslinkproductions\u002Fyuri-os --skill quantum-hypothesis-simulation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fquantum-hypothesis-simulation\u002FSKILL.md",true,[],[80,100,123,153,182,201],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":39},"01M71CG1AME5WC0A79AE65WT0X","zoom-out","Tell the agent to zoom out and give broader context or a higher-level perspective. Use when you're unfamiliar with a section of code or need to understand how it fits into the bigger picture.","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fzoom-out\u002FSKILL.md",[],[87,88,89,27,90,91,92,93,94,95,96,34,33,97,98,99],"zoom out","zoom","with","how","section","re","true","understand","use","perspective","code","it","context",{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":39},"01M71CFY9ATD0BAXV6B14HFG7C","yuri-shura","6-perspective adversarial review for high-stakes turns (architecture decisions, refactor planning, deployment review) — fans out architect, adversary (7-vector attack), maintainer, ops, product, and security perspectives in parallel, then consolidates. Use when user says 'review this decision', '\u002Fshura', 'adversarial review', 'what could go wrong', 'architecture review', or 'deployment review'.","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-shura\u002FSKILL.md",[],[107,108,27,109,110,25,111,112,113,114,115,95,116,34,117,118,119,120,121,122],"yuri shura","parallel","architecture review","planning","deployment","vector","strategic review","decision","strategic","ops","workflow","decisions","architecture","review","perspectives","adversarial",{"skillNo":124,"name":125,"nameZh":6,"description":126,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":134,"publishedAt":39},"01M71CFTTYW1HA2MNX38EV1GJ5","yuri-sales-intelligence","Unified sales and outreach skill generating cold outreach openers, social selling sequences, objection rebuttals, and ad creative for any paid platform — multi-persona aware, adapting tone and offer frame to buyer type. Use when user says 'write a cold email', 'outreach sequence', 'objection handling', 'ad copy', or 'sales pitch'.","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-sales-intelligence\u002FSKILL.md",[129],{"categoryNo":130,"categoryName":131,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000005","邮件与沟通","SCENARIO","使用场景",[135,136,137,138,139,140,141,142,143,34,144,145,146,147,148,149,150,25,151,152],"intelligence","sales intelligence","yuri sales intelligence","sales","cold","paid","social","persona","pitch","cold email","ad creative","write","sequences","sequence","platform","email","cold outreach","frame",{"skillNo":154,"name":155,"nameZh":6,"description":156,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":157,"categories":158,"keywords":167,"publishedAt":39},"01M71CFQMBFTFADJNGP0Z2WCYK","yuri-code-intelligence","Unified code quality intelligence layer that diagnoses code smells (Long Method, Feature Envy, design pattern issues), prescribes targeted Fowler refactoring techniques, and reviews code for security, performance, and clarity. Use when user says 'review this code', 'refactor this', 'what's wrong with this method', 'code smells', or 'clean this up'.","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-code-intelligence\u002FSKILL.md",[159,162],{"categoryNo":160,"categoryName":161,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":163,"categoryName":164,"taxonomyType":165,"taxonomyTypeDesc":166},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[168,97,169,135,27,170,171,172,25,173,174,175,176,177,178,179,180,181,95,34],"yuri code intelligence","code intelligence","reviews","code quality","feature","performance","code review","long","up","issues","clean","techniques","improve","design",{"skillNo":183,"name":184,"nameZh":6,"description":185,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":186,"categories":187,"keywords":188,"publishedAt":39},"01M71CFP32RT6GD3GKNHCJP8VP","yuri-adw","Invoke when user asks to run the ADW loop \u002F agentic engineering loop \u002F software factory workflow.","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-adw\u002FSKILL.md",[],[189,190,191,192,193,194,195,196,197,34,198,117,199,200],"yuri adw","run","engineering loop","software factory","software","agentic engineering","factory","loop","invoke","engineering","agentic","user",{"skillNo":202,"name":203,"nameZh":6,"description":204,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":205,"categories":206,"keywords":219,"publishedAt":39},"01M71CFK4VPEKY6A73237P7RSN","writing-skills","Use when creating new skills, editing existing skills, or verifying skills work before deployment","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-skills\u002FSKILL.md",[207,210,213,216],{"categoryNo":208,"categoryName":209,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":211,"categoryName":212,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":214,"categoryName":215,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":217,"categoryName":218,"taxonomyType":165,"taxonomyTypeDesc":166},"01JBSEEDCAT000000000000036","内容创作者",[220,221,222,34,117,223,27,224,111,95,225],"writing","writing skills","new","work","editing","Existing"]