[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71CFP32RT6GD3GKNHCJP8VP":3,"skill-same:01M71CFP32RT6GD3GKNHCJP8VP":74},{"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":33,"linkOnly":34,"markdownContent":35,"markdownBody":36,"frontMatter":37,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":41,"repositoryHtmlUrl":42,"repositoryHomepageUrl":6,"ownerHtmlUrl":43,"ownerBlogUrl":6,"ownerDescription":44,"updatedAt":45,"install":46,"resources":59},"01M71CFP32RT6GD3GKNHCJP8VP","yuri-adw",null,"Invoke when user asks to run the ADW loop \u002F agentic engineering loop \u002F software factory workflow.","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\u002Fyuri-adw\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32],"yuri adw","run","engineering loop","software factory","software","agentic engineering","factory","loop","invoke","when","engineering","workflow","agentic","user","2026-11-10 10:12:30",false,"---\nname: yuri-adw\ndescription: Invoke when user asks to run the ADW loop \u002F agentic engineering loop \u002F software factory workflow.\n---\n\n## When to use\n- Invoke for end-to-end engineering tickets that need orchestrated planning, deterministic validation, and review.\n- Use when output must pass gates, include evidence, and route failures through BUILD payloads.\n- Prefer this loop when deterministic artifacts should dominate over agent claims.\n- Do not run blindly; pause for owner clarification on missing scope, no acceptance criteria, or blocked dependencies.\n\n## The loop\n- `INTAKE -> SCOUT -> PLAN -> BUILD -> VALIDATE -> REVIEW -> SHIP`\n- INTAKE: orchestrator decodes + sizes ticket.\n- SCOUT: cheap R0 lane (Phoenix, deepseek-v4-flash) produces evidence pack; output is hypothesis.\n- PLAN: orchestrator writes the micro-task plan AND, for the gate, an executable decision manifest — `{name, discrete:{dim:[options]}, paramSpace:{param:[lo,hi]}, value:\"(config,params)=>number\", ...}` (see `adw-gate.mjs --help` for the full schema; the gate scores a decision problem, not prose).\n- PLAN GATE: run `node _SYSTEM\u002FScripts\u002Fadw-gate.mjs plan --input \u003Cmanifest.json>` returning `gate, pass, score, cornerLawBite, predictionId, reasons`. Records a prediction to the ledger.\n- BUILD: fan out micro-tasks to producer lanes (gpt-5.4-mini mechanical, gpt-5.3-codex-spark creative, Sonnet semantic; gpt-5.6-luna for rescue\u002Fhard delegation — note the provider-route registry classifies luna verifier\u002Fstrategist, so prefer it for rescue over volume production).\n- VALIDATE: deterministic code checks first (lint\u002Ftype\u002Ftests), then ENERGY GATE.\n- VALIDATE GATE: run `node _SYSTEM\u002FScripts\u002Fadw-gate.mjs validate --input \u003Coutcome.json>` returning `gate, pass, deltaU, contributions, reasons`. Records the outcome to the prediction-ledger ONLY when the input carries the plan's `predictionId`; writes an energy trace only when `YURI_ENERGY_OBSERVABILITY=1` (advisory-unless-armed).\n- REVIEW: \u002Fcodex:review with profile `sol` (`gpt-5.6-sol`) plus native `\u002Fcode-review`; adjudicate claim-by-claim.\n- SHIP: orchestrator only, scoped-pathspec commit, prediction-ledger report, closeout.\n\n## Phase reference table\n| Phase | Actor\u002Flane | Command | Gate | Failure route |\n|---|---|---|---|---|\n| INTAKE | Orchestrator | decode ticket, define constraints, size tasks | none | missing clarity → request owner input |\n| SCOUT | R0 lane: Phoenix \u002F deepseek-v4-flash | gather evidence pack | none; all outputs are hypothesis | false\u002Fweak claim → ignore or recollect evidence |\n| PLAN | Orchestrator | build manifest.json then run plan gate | `adw-gate.mjs plan --input \u003Cmanifest.json>` | fail -> replan with counter-evidence (max 2 cycles), then HOLD |\n| BUILD | Producer lanes: mini, spark, Sonnet (luna = rescue only — registry: verifier-strategist, mayExecuteWorkerTasks false) | execute micro-tasks; no self-verification | none | submit failure payload back to orchestrator for BUILD loop |\n| VALIDATE | Orchestrator + deterministic tooling | lint\u002Ftype\u002Ftests, then `adw-gate.mjs validate --input \u003Coutcome.json>` | validate gate pass -> deltaU+contributions record | fail -> immediate payload back to BUILD in same session |\n| REVIEW | \u002Fcodex:review (sol) + \u002Fcode-review + third-family tie-break (Helios\u002FGrok) on disputes | claim-by-claim adjudication | cross-family adversarial pass only | unresolved claim -> return to BUILD or owner |\n| SHIP | Orchestrator | scoped `git add`, scoped `git commit`, ledger report | owner-ready, no blocking claims | conflict\u002Funcertain evidence -> HOLD, no commit |\n\n## Roles and lanes\n- Keep roles per-task, not fixed per-lane.\n- Scout, builder, validator, and verifier are separate responsibilities in each workcell.\n- A person\u002Fagent may switch roles only with explicit handoff and role boundary log.\n\n## Evidence rules\n- Tag every claim as `CONFIRMED` or `PLAUSIBLE`.\n- Lane output from SCOUT\u002FBUILD is `PLAUSIBLE` by default and advisory only.\n- Deterministic artifacts (`git diff`, tests, type\u002Flint logs, execution logs) are `CONFIRMED`.\n- Orchestrator must spot-check at least one claimed path before using SCOUT claims.\n- Only `CONFIRMED` claims can pass REVIEW gates and drive SHIP.\n\n## Bounded loops\n- Replan only inside PLAN failure.\n- Max 2 PLAN replans.\n- Third PLAN failure => `HOLD` is a valid terminal state and stops autonomous progression.\n- `HOLD` requires owner resolution, then resume as a fresh loop turn.\n\n## Invocation per surface\n- Claude Code session: invoke skill `yuri-adw` with the ticket text.\n- Codex terminal: invoke skill `yuri-adw` with manifest\u002Fticket context; run worker lanes as needed.\n- OMP terminal: invoke skill `yuri-adw` with task + acceptance criteria.\n- For all surfaces, pass inputs as plain task text + explicit constraints and expected deliverables.\n","## When to use\n- Invoke for end-to-end engineering tickets that need orchestrated planning, deterministic validation, and review.\n- Use when output must pass gates, include evidence, and route failures through BUILD payloads.\n- Prefer this loop when deterministic artifacts should dominate over agent claims.\n- Do not run blindly; pause for owner clarification on missing scope, no acceptance criteria, or blocked dependencies.\n\n## The loop\n- `INTAKE -> SCOUT -> PLAN -> BUILD -> VALIDATE -> REVIEW -> SHIP`\n- INTAKE: orchestrator decodes + sizes ticket.\n- SCOUT: cheap R0 lane (Phoenix, deepseek-v4-flash) produces evidence pack; output is hypothesis.\n- PLAN: orchestrator writes the micro-task plan AND, for the gate, an executable decision manifest — `{name, discrete:{dim:[options]}, paramSpace:{param:[lo,hi]}, value:\"(config,params)=>number\", ...}` (see `adw-gate.mjs --help` for the full schema; the gate scores a decision problem, not prose).\n- PLAN GATE: run `node _SYSTEM\u002FScripts\u002Fadw-gate.mjs plan --input \u003Cmanifest.json>` returning `gate, pass, score, cornerLawBite, predictionId, reasons`. Records a prediction to the ledger.\n- BUILD: fan out micro-tasks to producer lanes (gpt-5.4-mini mechanical, gpt-5.3-codex-spark creative, Sonnet semantic; gpt-5.6-luna for rescue\u002Fhard delegation — note the provider-route registry classifies luna verifier\u002Fstrategist, so prefer it for rescue over volume production).\n- VALIDATE: deterministic code checks first (lint\u002Ftype\u002Ftests), then ENERGY GATE.\n- VALIDATE GATE: run `node _SYSTEM\u002FScripts\u002Fadw-gate.mjs validate --input \u003Coutcome.json>` returning `gate, pass, deltaU, contributions, reasons`. Records the outcome to the prediction-ledger ONLY when the input carries the plan's `predictionId`; writes an energy trace only when `YURI_ENERGY_OBSERVABILITY=1` (advisory-unless-armed).\n- REVIEW: \u002Fcodex:review with profile `sol` (`gpt-5.6-sol`) plus native `\u002Fcode-review`; adjudicate claim-by-claim.\n- SHIP: orchestrator only, scoped-pathspec commit, prediction-ledger report, closeout.\n\n## Phase reference table\n| Phase | Actor\u002Flane | Command | Gate | Failure route |\n|---|---|---|---|---|\n| INTAKE | Orchestrator | decode ticket, define constraints, size tasks | none | missing clarity → request owner input |\n| SCOUT | R0 lane: Phoenix \u002F deepseek-v4-flash | gather evidence pack | none; all outputs are hypothesis | false\u002Fweak claim → ignore or recollect evidence |\n| PLAN | Orchestrator | build manifest.json then run plan gate | `adw-gate.mjs plan --input \u003Cmanifest.json>` | fail -> replan with counter-evidence (max 2 cycles), then HOLD |\n| BUILD | Producer lanes: mini, spark, Sonnet (luna = rescue only — registry: verifier-strategist, mayExecuteWorkerTasks false) | execute micro-tasks; no self-verification | none | submit failure payload back to orchestrator for BUILD loop |\n| VALIDATE | Orchestrator + deterministic tooling | lint\u002Ftype\u002Ftests, then `adw-gate.mjs validate --input \u003Coutcome.json>` | validate gate pass -> deltaU+contributions record | fail -> immediate payload back to BUILD in same session |\n| REVIEW | \u002Fcodex:review (sol) + \u002Fcode-review + third-family tie-break (Helios\u002FGrok) on disputes | claim-by-claim adjudication | cross-family adversarial pass only | unresolved claim -> return to BUILD or owner |\n| SHIP | Orchestrator | scoped `git add`, scoped `git commit`, ledger report | owner-ready, no blocking claims | conflict\u002Funcertain evidence -> HOLD, no commit |\n\n## Roles and lanes\n- Keep roles per-task, not fixed per-lane.\n- Scout, builder, validator, and verifier are separate responsibilities in each workcell.\n- A person\u002Fagent may switch roles only with explicit handoff and role boundary log.\n\n## Evidence rules\n- Tag every claim as `CONFIRMED` or `PLAUSIBLE`.\n- Lane output from SCOUT\u002FBUILD is `PLAUSIBLE` by default and advisory only.\n- Deterministic artifacts (`git diff`, tests, type\u002Flint logs, execution logs) are `CONFIRMED`.\n- Orchestrator must spot-check at least one claimed path before using SCOUT claims.\n- Only `CONFIRMED` claims can pass REVIEW gates and drive SHIP.\n\n## Bounded loops\n- Replan only inside PLAN failure.\n- Max 2 PLAN replans.\n- Third PLAN failure => `HOLD` is a valid terminal state and stops autonomous progression.\n- `HOLD` requires owner resolution, then resume as a fresh loop turn.\n\n## Invocation per surface\n- Claude Code session: invoke skill `yuri-adw` with the ticket text.\n- Codex terminal: invoke skill `yuri-adw` with manifest\u002Fticket context; run worker lanes as needed.\n- OMP terminal: invoke skill `yuri-adw` with task + acceptance criteria.\n- For all surfaces, pass inputs as plain task text + explicit constraints and expected deliverables.\n",{"name":5,"description":7},"main","4c6d8d65f3d6096e2fc25b0845e481dda73c2794","skills\u002Fyuri-adw",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":47,"prompt":57,"zipAvailable":58},[48,51,54],{"tool":49,"command":50},"npx","npx skills add nexuslinkproductions\u002Fyuri-os --skill yuri-adw",{"tool":52,"command":53},"bunx","bunx skills add nexuslinkproductions\u002Fyuri-os --skill yuri-adw",{"tool":55,"command":56},"pnpm","pnpm dlx skills add nexuslinkproductions\u002Fyuri-os --skill yuri-adw","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-adw\u002FSKILL.md",true,[60,66,70],{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"examples\u002Fplan-manifest.example.json","REFERENCE","参考文档",560,"https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-adw\u002Fexamples\u002Fplan-manifest.example.json",{"path":67,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":68,"sourceHtmlUrl":69},"examples\u002FREADME.md",764,"https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-adw\u002Fexamples\u002FREADME.md",{"path":71,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":72,"sourceHtmlUrl":73},"examples\u002Fvalidate-outcome.example.json",698,"https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fyuri-adw\u002Fexamples\u002Fvalidate-outcome.example.json",[75,97,120,150,179,204],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":33},"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",[],[82,83,84,85,86,87,88,89,90,91,92,28,93,94,95,96],"zoom out","zoom","with","harness","how","section","re","true","understand","use","perspective","you","code","it","context",{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":33},"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",[],[104,105,85,106,107,108,109,110,111,112,113,91,114,28,30,115,116,117,118,119],"yuri shura","parallel","architecture review","planning","for","deployment","vector","strategic review","decision","strategic","ops","decisions","architecture","review","perspectives","adversarial",{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":131,"publishedAt":33},"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",[126],{"categoryNo":127,"categoryName":128,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000005","邮件与沟通","SCENARIO","使用场景",[132,133,134,135,136,137,138,139,140,28,141,142,143,144,145,146,147,108,148,149],"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":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":164,"publishedAt":33},"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",[156,159],{"categoryNo":157,"categoryName":158,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":160,"categoryName":161,"taxonomyType":162,"taxonomyTypeDesc":163},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[165,94,166,132,85,167,168,169,108,170,171,172,173,174,175,176,177,178,91,28],"yuri code intelligence","code intelligence","reviews","code quality","feature","performance","code review","long","up","issues","clean","techniques","improve","design",{"skillNo":180,"name":181,"nameZh":6,"description":182,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":183,"categories":184,"keywords":197,"publishedAt":33},"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",[185,188,191,194],{"categoryNo":186,"categoryName":187,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":189,"categoryName":190,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":192,"categoryName":193,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":195,"categoryName":196,"taxonomyType":162,"taxonomyTypeDesc":163},"01JBSEEDCAT000000000000036","内容创作者",[198,199,200,28,30,201,85,202,109,91,203],"writing","writing skills","new","work","editing","Existing",{"skillNo":205,"name":206,"nameZh":6,"description":207,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":208,"categories":209,"keywords":212,"publishedAt":33},"01M71CFGS93M91C0J7SSPXM1NS","writing-plans","Use when you have a spec or requirements for a multi-step task, before touching code","https:\u002F\u002Fgithub.com\u002Fnexuslinkproductions\u002Fyuri-os\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-plans\u002FSKILL.md",[210,211],{"categoryNo":189,"categoryName":190,"taxonomyType":129,"taxonomyTypeDesc":130},{"categoryNo":195,"categoryName":196,"taxonomyType":162,"taxonomyTypeDesc":163},[213,198,214,85,108,91,28,30,93,215,94,216,217,218,219],"plans","writing plans","requirements","task","multi","spec","multi-step task"]