[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71BVVWD6TRXS5YMXECV6XT5":3,"skill-same:01M71BVVWD6TRXS5YMXECV6XT5":77},{"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":56,"commitSha":57,"skillDirectory":58,"forksCount":59,"repositoryHtmlUrl":60,"repositoryHomepageUrl":6,"ownerHtmlUrl":61,"ownerBlogUrl":6,"ownerDescription":62,"updatedAt":63,"install":64,"resources":76},"01M71BVVWD6TRXS5YMXECV6XT5","probabilistic-decision-core",null,"Operational probability, calibration, and expected-value discipline for YURI OS decisions under uncertainty. Use when making or supporting a decision involving task priority, route selection, risk acceptance, opportunity ranking, forecast-backed planning, resource allocation, go\u002Fno-go judgment, escalation threshold, or postmortem calibration — e.g. '\u002Fyuri probability', '\u002Fprobability', or '\u002Fpdc'.","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\u002F.claude\u002Fskills\u002Fprobabilistic-decision-core\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"core","probabilistic decision core","decision","probability","go","when","calibration","allocation","resource allocation","ranking","infinity","os","loop","judgment","active","opportunity","route","selection","discipline","acceptance","2026-11-10 10:12:30",false,"---\nname: probabilistic-decision-core\ndescription: \"Operational probability, calibration, and expected-value discipline for YURI OS decisions under uncertainty. Use when making or supporting a decision involving task priority, route selection, risk acceptance, opportunity ranking, forecast-backed planning, resource allocation, go\u002Fno-go judgment, escalation threshold, or postmortem calibration — e.g. '\u002Fyuri probability', '\u002Fprobability', or '\u002Fpdc'.\"\nversion: 1.0.0\nstatus: active\nenterprise_ready: true\nnon_destructive_default: true\ntriggers:\n  - \"\u002Fyuri probability\"\n  - \"\u002Fprobability\"\n  - \"\u002Fpdc\"\nrequires:\n  - execution-domain-core\n  - non-destructive-infinity-guard\n  - failure-evolution-loop\nscope: harness\ninvocation: ability\n---\n\n# Probabilistic Decision Core Skill\n\n## When to use\n\nUse this skill when Yuri OS \u002F YURI must make or support an operational decision under uncertainty:\n\n- task priority\n- route selection\n- risk acceptance\n- opportunity ranking\n- forecast-backed planning\n- resource allocation\n- go \u002F no-go judgment\n- escalation threshold\n- postmortem calibration\n\nAlso use it when the active task matches this operating principle:\n\n> Separate forecast, goal, plan, confidence, and cost of error before action. Update the estimate when evidence changes. Calibrate against outcomes after the decision resolves.\n\n## When not to use\n\nDo not use this skill when:\n\n- the task is deterministic and local evidence fully decides it\n- the decision is identity-level, relationship-critical, or owner-only\n- the only available estimate would be fake precision\n- the action is high or critical risk and lacks owner approval\n- the outcome cannot be observed later enough to calibrate\n\n## Operating doctrine\n\n1. Forecasts are not goals.\n2. Probability is a decision input, not a permission slip.\n3. Base rates come before local signals.\n4. Local signals must name direction and strength.\n5. Confidence describes evidence quality, not emotional certainty.\n6. Expected value must include cost of being wrong.\n7. High-impact irreversible actions require guard routing even when probability is favorable.\n8. Forecasts must be logged when the outcome can later be checked.\n\n## Required output shape\n\n```yaml\nprobabilistic_decision:\n  decision: string\n  outcome_being_estimated: string\n  time_horizon: string\n  base_rate:\n    estimate: number | unknown\n    source: string\n    confidence: low | medium | high\n  predictability:\n    factor_understanding: low | medium | high\n    data_available: none | weak | adequate | strong\n    future_similarity: low | medium | high\n    observer_effect: none | weak | material\n    verdict: low | medium | high\n  signals_for:\n    - signal: string\n      strength: weak | medium | strong\n      source: string\n  signals_against:\n    - signal: string\n      strength: weak | medium | strong\n      source: string\n  estimate:\n    probability: number | range | not_estimable\n    confidence: low | medium | high\n    rationale: string\n  decision_value:\n    upside: string\n    downside: string\n    cost_if_wrong: low | medium | high | critical\n    reversibility: low | medium | high\n    expected_value: negative | neutral | positive | unclear\n  action:\n    recommendation: proceed | defer | gather_more_evidence | escalate_to_owner | block\n    next_step: string\n    calibration_log_required: true | false\n```\n\n## Execution steps\n\n1. Define the decision and the concrete outcome being estimated.\n2. Separate goal from forecast: name what Marcel wants and what seems likely.\n3. Check predictability before assigning a probability.\n4. Establish a base rate from local history, external evidence, or mark it `unknown`.\n5. Add local evidence as signals for and against, with strength.\n6. Estimate probability as a range unless evidence supports a point estimate.\n7. Evaluate expected value, reversibility, and cost of error.\n8. Select action: proceed, defer, gather evidence, escalate, or block.\n9. If the outcome is observable, append a calibration row to the active probability calibration log.\n\n## Probability bands\n\n| Band | Use |\n|---|---|\n| 0-20% | unlikely |\n| 20-40% | doubtful |\n| 40-60% | uncertain \u002F coin-flip zone |\n| 60-80% | likely |\n| 80-95% | strong likelihood |\n| 95%+ | only for deterministic or heavily evidenced cases |\n\n## Calibration rules\n\n- Prefer ranges for first-pass estimates.\n- Do not use decimals unless produced by a real model or calculation.\n- Mark `not_estimable` when predictability is low and evidence is weak.\n- Score resolved binary forecasts with Brier score: `(forecast_probability - outcome)^2`.\n- Review calibration in buckets: 20-40%, 40-60%, 60-80%, 80-95%.\n- Treat overconfidence as a failure pattern and route repeated failures through `failure-evolution-loop`.\n\n## Integration with Yuri operations\n\n- Use before `execution-domain-core` when uncertainty shapes the domain boundary.\n- Use before `non-destructive-infinity-guard` when probability affects risk acceptance.\n- Use after `failure-evolution-loop` when a wrong estimate caused bad routing, priority, or action.\n- Use alongside market intelligence, operational planning, research prioritization, and other work where outcome probabilities drive decisions.\n\n## Quantum-probability instrument (gate-passed 2026-06-11)\n\nFor ORDER-SENSITIVE or frame-dependent evidence, classical Bayes is provably blind to\nsequence effects. The quantum-probability layer passed its falsification gate\n(`node _SYSTEM\u002FScripts\u002Fquantum-vs-bayes-benchmark.mjs` → exit 0; 4\u002F4 gates: machinery\nexact, QQ-coincidence \u003C1% under the classical family on Gallup order-effect data, no\nspurious win on order-free control). Module: `_SYSTEM\u002FScripts\u002Fquantum-hypothesis-tracker.mjs`.\n\n- WHEN: combining evidence whose framing\u002Forder changes the judgment (sequential\n  advisor opinions, owner corrections after partial conclusions, poll-style A-then-B\n  assessments). With commuting\u002Fcompatible evidence it reduces EXACTLY to Bayes — use\n  plain Bayes there.\n- API: `stateVector`, `projector`, `hypothesisPosteriors(psi, hypotheses, evidenceSeq)`\n  (order-aware), `qqEquality` (the structural test), `schmidtDecomposition` (coupling\n  test for \"are these two variables genuinely entangled or independent\").\n- KNOWN LIMITS (recorded in the benchmark): 2D models force p(yy)\u002Fp(yn)=p(nn)\u002Fp(ny)\n  per order — quantitative joint fits of heterogeneous populations need dim>2; the\n  Clinton\u002FGore joint cells used by gates G2\u002FG4 are literature-recalled and flagged\n  OWNER-VERIFY; one real dataset so far (literature has 72 — extend when verified\n  tables are available).\n\n## Research basis\n\n- Forecasting practice should distinguish forecasts, goals, and plans; planning responds to forecasts and goals.\n- Predictability depends on factor understanding, data availability, future similarity to the past, and whether forecasts affect the outcome.\n- Distributional forecasts should be evaluated against outcomes, not just stated.\n- Decision analysis should propagate uncertainty through alternatives instead of relying only on point estimates.\n- Proper scoring rules such as Brier score reward honest probabilistic forecasts and support calibration review.\n\nSources:\n- https:\u002F\u002Fotexts.com\u002Ffpp3\u002Fplanning.html\n- https:\u002F\u002Fotexts.com\u002Ffpp3\u002Fwhat-can-be-forecast.html\n- https:\u002F\u002Fotexts.com\u002Ffpp3\u002Fdistaccuracy.html\n- https:\u002F\u002Fwww.nist.gov\u002Fpublications\u002Fincorporating-attribute-value-uncertainty-decision-analysis\n- https:\u002F\u002Flink.springer.com\u002Farticle\u002F10.1007\u002Fs10994-023-06336-7\n\n## Safety rules\n\n- Do not let favorable probability bypass explicit owner approval.\n- Do not convert owner preferences into forecasts.\n- Do not treat model confidence as evidence confidence.\n- Do not log personal, identity-level, or relationship-sensitive predictions unless explicitly requested.\n- Do not write to protected state surfaces.\n\n## Success criteria\n\nThe skill succeeds when it produces:\n\n- a clear decision recommendation\n- separated forecast, goal, plan, and confidence\n- explicit base rate or `unknown`\n- named evidence for and against\n- cost-of-error and reversibility assessment\n- calibration log proposal when the outcome can be checked\n\n## Failure handling\n\nIf execution fails, return:\n\n```yaml\nfailure:\n  extension_id: \"probabilistic-decision-core\"\n  halt_reason: string\n  missing_inputs: []\n  unsafe_precision_risk: true | false\n  recommended_next_action: string\n```\n\n## Session Notes\n\n### 2026-06-16\n- session: 183m | peak ctx: 0% | compacts: 0\n- tools: Bash×400, Read×122, Edit×62, Write×19, WebSearch×19, Agent×10, WebFetch×8, ToolSearch×5, TodoWrite×4, ExitPlanMode×3, AskUserQuestion×2, Workflow×1, Skill×1\n- corrections: none\n- errors: none\n\n### 2026-06-15\n- session: 136m | peak ctx: 0% | compacts: 0\n- tools: Bash×272, Read×67, Edit×26, WebSearch×19, Write×9, WebFetch×8, Agent×7, ToolSearch×5, ExitPlanMode×2, TodoWrite×2, AskUserQuestion×1, Workflow×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\n### 2026-06-11\n- session: 50m | peak ctx: 100% | compacts: 8\n- tools: Bash×66, Read×24, Edit×15, Write×14, WebFetch×7, WebSearch×3, TodoWrite×3, ToolSearch×2, Agent×1\n- corrections: none\n- errors: none\n\n### 2026-05-16\n- session: 41m | peak ctx: 0% | compacts: 0\n- tools: Bash×44, Write×25, Edit×21, TodoWrite×6, Read×5, mcp×1, Skill×1\n- corrections: Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fend-of-transmission\n\n# End of Transmission\n\nContinuous background reflection engine for YURI. Runs **two modes**:\n\n1. **Micro-EOT**\n- errors: none\n\n### 2026-05-10\n- tools used: web research, local file inspection, apply_patch\n- corrections: none\n- errors: none\n- notes: Created as an operational decision layer for Yuri OS probability, uncertainty, expected value, and calibration discipline.\n","# Probabilistic Decision Core Skill\n\n## When to use\n\nUse this skill when Yuri OS \u002F YURI must make or support an operational decision under uncertainty:\n\n- task priority\n- route selection\n- risk acceptance\n- opportunity ranking\n- forecast-backed planning\n- resource allocation\n- go \u002F no-go judgment\n- escalation threshold\n- postmortem calibration\n\nAlso use it when the active task matches this operating principle:\n\n> Separate forecast, goal, plan, confidence, and cost of error before action. Update the estimate when evidence changes. Calibrate against outcomes after the decision resolves.\n\n## When not to use\n\nDo not use this skill when:\n\n- the task is deterministic and local evidence fully decides it\n- the decision is identity-level, relationship-critical, or owner-only\n- the only available estimate would be fake precision\n- the action is high or critical risk and lacks owner approval\n- the outcome cannot be observed later enough to calibrate\n\n## Operating doctrine\n\n1. Forecasts are not goals.\n2. Probability is a decision input, not a permission slip.\n3. Base rates come before local signals.\n4. Local signals must name direction and strength.\n5. Confidence describes evidence quality, not emotional certainty.\n6. Expected value must include cost of being wrong.\n7. High-impact irreversible actions require guard routing even when probability is favorable.\n8. Forecasts must be logged when the outcome can later be checked.\n\n## Required output shape\n\n```yaml\nprobabilistic_decision:\n  decision: string\n  outcome_being_estimated: string\n  time_horizon: string\n  base_rate:\n    estimate: number | unknown\n    source: string\n    confidence: low | medium | high\n  predictability:\n    factor_understanding: low | medium | high\n    data_available: none | weak | adequate | strong\n    future_similarity: low | medium | high\n    observer_effect: none | weak | material\n    verdict: low | medium | high\n  signals_for:\n    - signal: string\n      strength: weak | medium | strong\n      source: string\n  signals_against:\n    - signal: string\n      strength: weak | medium | strong\n      source: string\n  estimate:\n    probability: number | range | not_estimable\n    confidence: low | medium | high\n    rationale: string\n  decision_value:\n    upside: string\n    downside: string\n    cost_if_wrong: low | medium | high | critical\n    reversibility: low | medium | high\n    expected_value: negative | neutral | positive | unclear\n  action:\n    recommendation: proceed | defer | gather_more_evidence | escalate_to_owner | block\n    next_step: string\n    calibration_log_required: true | false\n```\n\n## Execution steps\n\n1. Define the decision and the concrete outcome being estimated.\n2. Separate goal from forecast: name what Marcel wants and what seems likely.\n3. Check predictability before assigning a probability.\n4. Establish a base rate from local history, external evidence, or mark it `unknown`.\n5. Add local evidence as signals for and against, with strength.\n6. Estimate probability as a range unless evidence supports a point estimate.\n7. Evaluate expected value, reversibility, and cost of error.\n8. Select action: proceed, defer, gather evidence, escalate, or block.\n9. If the outcome is observable, append a calibration row to the active probability calibration log.\n\n## Probability bands\n\n| Band | Use |\n|---|---|\n| 0-20% | unlikely |\n| 20-40% | doubtful |\n| 40-60% | uncertain \u002F coin-flip zone |\n| 60-80% | likely |\n| 80-95% | strong likelihood |\n| 95%+ | only for deterministic or heavily evidenced cases |\n\n## Calibration rules\n\n- Prefer ranges for first-pass estimates.\n- Do not use decimals unless produced by a real model or calculation.\n- Mark `not_estimable` when predictability is low and evidence is weak.\n- Score resolved binary forecasts with Brier score: `(forecast_probability - outcome)^2`.\n- Review calibration in buckets: 20-40%, 40-60%, 60-80%, 80-95%.\n- Treat overconfidence as a failure pattern and route repeated failures through `failure-evolution-loop`.\n\n## Integration with Yuri operations\n\n- Use before `execution-domain-core` when uncertainty shapes the domain boundary.\n- Use before `non-destructive-infinity-guard` when probability affects risk acceptance.\n- Use after `failure-evolution-loop` when a wrong estimate caused bad routing, priority, or action.\n- Use alongside market intelligence, operational planning, research prioritization, and other work where outcome probabilities drive decisions.\n\n## Quantum-probability instrument (gate-passed 2026-06-11)\n\nFor ORDER-SENSITIVE or frame-dependent evidence, classical Bayes is provably blind to\nsequence effects. The quantum-probability layer passed its falsification gate\n(`node _SYSTEM\u002FScripts\u002Fquantum-vs-bayes-benchmark.mjs` → exit 0; 4\u002F4 gates: machinery\nexact, QQ-coincidence \u003C1% under the classical family on Gallup order-effect data, no\nspurious win on order-free control). Module: `_SYSTEM\u002FScripts\u002Fquantum-hypothesis-tracker.mjs`.\n\n- WHEN: combining evidence whose framing\u002Forder changes the judgment (sequential\n  advisor opinions, owner corrections after partial conclusions, poll-style A-then-B\n  assessments). With commuting\u002Fcompatible evidence it reduces EXACTLY to Bayes — use\n  plain Bayes there.\n- API: `stateVector`, `projector`, `hypothesisPosteriors(psi, hypotheses, evidenceSeq)`\n  (order-aware), `qqEquality` (the structural test), `schmidtDecomposition` (coupling\n  test for \"are these two variables genuinely entangled or independent\").\n- KNOWN LIMITS (recorded in the benchmark): 2D models force p(yy)\u002Fp(yn)=p(nn)\u002Fp(ny)\n  per order — quantitative joint fits of heterogeneous populations need dim>2; the\n  Clinton\u002FGore joint cells used by gates G2\u002FG4 are literature-recalled and flagged\n  OWNER-VERIFY; one real dataset so far (literature has 72 — extend when verified\n  tables are available).\n\n## Research basis\n\n- Forecasting practice should distinguish forecasts, goals, and plans; planning responds to forecasts and goals.\n- Predictability depends on factor understanding, data availability, future similarity to the past, and whether forecasts affect the outcome.\n- Distributional forecasts should be evaluated against outcomes, not just stated.\n- Decision analysis should propagate uncertainty through alternatives instead of relying only on point estimates.\n- Proper scoring rules such as Brier score reward honest probabilistic forecasts and support calibration review.\n\nSources:\n- https:\u002F\u002Fotexts.com\u002Ffpp3\u002Fplanning.html\n- https:\u002F\u002Fotexts.com\u002Ffpp3\u002Fwhat-can-be-forecast.html\n- https:\u002F\u002Fotexts.com\u002Ffpp3\u002Fdistaccuracy.html\n- https:\u002F\u002Fwww.nist.gov\u002Fpublications\u002Fincorporating-attribute-value-uncertainty-decision-analysis\n- https:\u002F\u002Flink.springer.com\u002Farticle\u002F10.1007\u002Fs10994-023-06336-7\n\n## Safety rules\n\n- Do not let favorable probability bypass explicit owner approval.\n- Do not convert owner preferences into forecasts.\n- Do not treat model confidence as evidence confidence.\n- Do not log personal, identity-level, or relationship-sensitive predictions unless explicitly requested.\n- Do not write to protected state surfaces.\n\n## Success criteria\n\nThe skill succeeds when it produces:\n\n- a clear decision recommendation\n- separated forecast, goal, plan, and confidence\n- explicit base rate or `unknown`\n- named evidence for and against\n- cost-of-error and reversibility assessment\n- calibration log proposal when the outcome can be checked\n\n## Failure handling\n\nIf execution fails, return:\n\n```yaml\nfailure:\n  extension_id: \"probabilistic-decision-core\"\n  halt_reason: string\n  missing_inputs: []\n  unsafe_precision_risk: true | false\n  recommended_next_action: string\n```\n\n## Session Notes\n\n### 2026-06-16\n- session: 183m | peak ctx: 0% | compacts: 0\n- tools: Bash×400, Read×122, Edit×62, Write×19, WebSearch×19, Agent×10, WebFetch×8, ToolSearch×5, TodoWrite×4, ExitPlanMode×3, AskUserQuestion×2, Workflow×1, Skill×1\n- corrections: none\n- errors: none\n\n### 2026-06-15\n- session: 136m | peak ctx: 0% | compacts: 0\n- tools: Bash×272, Read×67, Edit×26, WebSearch×19, Write×9, WebFetch×8, Agent×7, ToolSearch×5, ExitPlanMode×2, TodoWrite×2, AskUserQuestion×1, Workflow×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\n### 2026-06-11\n- session: 50m | peak ctx: 100% | compacts: 8\n- tools: Bash×66, Read×24, Edit×15, Write×14, WebFetch×7, WebSearch×3, TodoWrite×3, ToolSearch×2, Agent×1\n- corrections: none\n- errors: none\n\n### 2026-05-16\n- session: 41m | peak ctx: 0% | compacts: 0\n- tools: Bash×44, Write×25, Edit×21, TodoWrite×6, Read×5, mcp×1, Skill×1\n- corrections: Base directory for this skill: \u002FUsers\u002Fmarcelspatz\u002F.claude\u002Fskills\u002Fend-of-transmission\n\n# End of Transmission\n\nContinuous background reflection engine for YURI. Runs **two modes**:\n\n1. **Micro-EOT**\n- errors: none\n\n### 2026-05-10\n- tools used: web research, local file inspection, apply_patch\n- corrections: none\n- errors: none\n- notes: Created as an operational decision layer for Yuri OS probability, uncertainty, expected value, and calibration discipline.\n",{"name":5,"scope":44,"status":33,"version":45,"requires":46,"triggers":50,"invocation":54,"description":7,"enterprise_ready":55,"non_destructive_default":55},"harness","1.0.0",[47,48,49],"execution-domain-core","non-destructive-infinity-guard","failure-evolution-loop",[51,52,53],"\u002Fyuri probability","\u002Fprobability","\u002Fpdc","ability",true,"main","4c6d8d65f3d6096e2fc25b0845e481dda73c2794",".claude\u002Fskills\u002Fprobabilistic-decision-core",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":65,"prompt":75,"zipAvailable":55},[66,69,72],{"tool":67,"command":68},"npx","npx skills add nexuslinkproductions\u002Fyuri-os --skill probabilistic-decision-core",{"tool":70,"command":71},"bunx","bunx skills add nexuslinkproductions\u002Fyuri-os --skill probabilistic-decision-core",{"tool":73,"command":74},"pnpm","pnpm dlx skills add nexuslinkproductions\u002Fyuri-os --skill probabilistic-decision-core","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fprobabilistic-decision-core\u002FSKILL.md",[],[78,99,122,152,181,199],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"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",[],[85,86,87,44,88,89,90,91,92,93,94,24,95,96,97,98],"zoom out","zoom","with","how","section","re","true","understand","use","perspective","you","code","it","context",{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"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",[],[106,107,44,108,109,110,111,112,113,21,114,93,115,24,116,117,118,119,120,121],"yuri shura","parallel","architecture review","planning","for","deployment","vector","strategic review","strategic","ops","workflow","decisions","architecture","review","perspectives","adversarial",{"skillNo":123,"name":124,"nameZh":6,"description":125,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":126,"categories":127,"keywords":133,"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",[128],{"categoryNo":129,"categoryName":130,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000005","邮件与沟通","SCENARIO","使用场景",[134,135,136,137,138,139,140,141,142,24,143,144,145,146,147,148,149,110,150,151],"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":153,"name":154,"nameZh":6,"description":155,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":156,"categories":157,"keywords":166,"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",[158,161],{"categoryNo":159,"categoryName":160,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":162,"categoryName":163,"taxonomyType":164,"taxonomyTypeDesc":165},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[167,96,168,134,44,169,170,171,110,172,173,174,175,176,177,178,179,180,93,24],"yuri code intelligence","code intelligence","reviews","code quality","feature","performance","code review","long","up","issues","clean","techniques","improve","design",{"skillNo":182,"name":183,"nameZh":6,"description":184,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":185,"categories":186,"keywords":187,"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",[],[188,189,190,191,192,193,194,31,195,24,196,116,197,198],"yuri adw","run","engineering loop","software factory","software","agentic engineering","factory","invoke","engineering","agentic","user",{"skillNo":200,"name":201,"nameZh":6,"description":202,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":203,"categories":204,"keywords":217,"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",[205,208,211,214],{"categoryNo":206,"categoryName":207,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":209,"categoryName":210,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":212,"categoryName":213,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":215,"categoryName":216,"taxonomyType":164,"taxonomyTypeDesc":165},"01JBSEEDCAT000000000000036","内容创作者",[218,219,220,24,116,221,44,222,111,93,223],"writing","writing skills","new","work","editing","Existing"]