[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C46JY7XGNX83E7J75SS170":3,"skill-same:01M6C46JY7XGNX83E7J75SS170":73},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"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":6,"updatedAt":44,"install":45,"resources":58},"01M6C46JY7XGNX83E7J75SS170","refactoring-methodology",null,"Proactive refactoring workflow for touched files and nearby debt. Use when hunting extraction opportunities, evaluating duplication or oversized files, or improving structure without changing intended behavior. DO NOT USE FOR: adversarial defect review (use adversarial-review) or net-new feature implementation (use implementation-discipline)","Grimblaz\u002Fagent-orchestra","Grimblaz","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F74755849?v=4","01M6C430M9XZDHGHPWQVTQ3KMA","01M6C430RJY3MMX0QAZ3X8DR3Y",3,"MIT","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Frefactoring-methodology\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[23,24,25,26,27,28,29,30,31,32],"methodology","refactoring","refactoring methodology","review","workflow","extraction","behavior","new feature","use","discipline","2026-11-10 10:12:30",false,"---\nname: refactoring-methodology\ndescription: \"Proactive refactoring workflow for touched files and nearby debt. Use when hunting extraction opportunities, evaluating duplication or oversized files, or improving structure without changing intended behavior. DO NOT USE FOR: adversarial defect review (use adversarial-review) or net-new feature implementation (use implementation-discipline)\"\n---\n\n\u003C!-- platform-assumptions: markdown skill guidance for VS Code custom agents in Agent Orchestra; assumes repository refactors are validated against existing tests and architecture rules. -->\n\u003C!-- markdownlint-disable-file MD041 MD003 -->\n\n# Refactoring Methodology\n\nReusable methodology for behavior-preserving structural improvement.\n\n## When to Use\n\n- When recently modified files need a proactive refactoring pass\n- When duplication, oversized functions, or structural debt may have been introduced or exposed\n- When integration gaps or symmetric mechanical fixes should be checked while the relevant code is already open\n- When refactoring decisions need a repeatable analysis and verification flow\n\n## Purpose\n\nFind real maintainability improvements in touched code without widening into speculative rewrites. Refactoring here is an active search for cleaner structure, not a passive approval pass.\n\n## Analysis Workflow\n\n1. Inspect every file modified by the current change unless the plan explicitly marks it out of scope.\n2. Check file size, function size, duplication, conditional complexity, and extraction opportunities.\n3. Inspect immediate neighbors when the touched file clearly participates in a broader repeated pattern.\n4. Check for unused added data or unwired integration points and treat them as incomplete work, not future debt.\n5. Before closing, grep for symmetric occurrences when the fix pattern is mechanical and repo-wide.\n6. Validate that the refactor preserved behavior and respected project architecture rules.\n\n## Mandatory Checks\n\n### Size And Structure\n\n- Check line count for each modified file\n- Flag files above 80% of the configured limit\n- Treat functions above roughly 50 lines as extraction candidates\n- Split responsibilities when a class or file is becoming multipurpose\n\n### Duplication And Readability\n\n- Extract repeated blocks or repeated control-flow fragments\n- Replace copied formulas or validations with delegation to one source of truth\n- Rename unclear variables and replace magic numbers with named constants\n- Simplify long expressions or nested conditionals when comprehension is suffering\n\n### Testability And Architecture\n\n- Prefer extracting private logic into testable units when behavior stays covered\n- Inject dependencies instead of copying behavior into new classes or helpers\n- Review layer placement and composition using `software-architecture` before significant extraction\n\n## Integration Gap Rule\n\nIf the change adds data, metadata, or configuration that is not consumed where the feature requires it, treat that as incomplete implementation. Fix it now when the integration is local and bounded; only defer when the remaining work truly requires a larger design decision.\n\n## Proactive Hunting Stance\n\nDo not rubber-stamp a touched file as \"fine\" without inspection.\n\n- Presume the touched code is improvable until checked.\n- Hunt for duplication, unclear naming, oversized units, and awkward control flow.\n- Apply the Boy Scout Rule within the touched area and its immediate neighbors.\n- If no improvements are needed, state what you checked and why.\n\nSuccess means finding proportionate maintainability improvements when they exist, not refactoring for its own sake.\n\n## Integration Gaps Are Not Tech Debt\n\nIf a PR adds data that is not consumed where the feature requires it, treat that as incomplete implementation, not future cleanup.\n\nExamples:\n\n- `supportedRegions` is added but consumers do not filter by it\n- `TIER_MULTIPLIERS` exists but the pricing path does not apply it\n- priority metadata is added but the scheduler ignores it\n- one map gains new entries while symmetric related maps do not\n\nWhen you find unused new data:\n\n1. Identify where it should be used.\n2. Estimate the integration effort.\n3. If the fix is local and bounded, include it now.\n4. If it is genuinely larger, document why it is large rather than calling it generic tech debt.\n\n## Output Structure\n\nUse a concise analysis with these sections when reporting refactor work:\n\n- `## Refactoring Analysis`\n- `### Files Analyzed`\n- `### Opportunities Found`\n- `### Actions Taken`\n- `### Deferred (Out of Scope)`\n- `### Verification`\n\n## Verification\n\n- Run the project-configured validation and test commands after refactoring\n- Confirm coverage expectations still hold for the affected behavior\n- Re-check the modified files for any remaining symmetric mechanical misses\n\n## Conductor Integration\n\nIn the Code-Conductor hub flow, **ALWAYS call Refactor-Specialist after Code-Smith completes.** Hub-mode Code-Conductor still runs the mandatory Refactor-Specialist pass.\n\nIn `\u002Fspine-run`, the `implement-refactor` port resolves through [skills\u002Frefactoring-methodology\u002Fadapters\u002Fimplement-refactor-adapter.md](adapters\u002Fimplement-refactor-adapter.md), and predicate adapters may resolve the port as auto-N\u002FA or explicit skip before the work adapter runs.\n\nRefactor-Specialist will:\n\n1. Analyze all files modified in the PR\n2. Hunt proactively for improvement opportunities\n3. Report findings (even if no action taken)\n4. Make improvements where beneficial\n\n**In the Code-Conductor hub flow, there is no \"skip refactoring\" option.** The Refactor-Specialist decides what needs improvement, not the plan or Code-Conductor.\n\n**Flow**: Code-Smith -> Refactor-Specialist -> Code-Critic\n\n**Clarification**: \"Avoid broad rewrites\" does NOT mean \"skip refactoring\" - it means keep refactoring proportionate to the PR's intent.\n\n**Proportionate refactoring (good)** means improving code you already touched (or its immediate neighbors) to reduce complexity\u002Fduplication without expanding the PR's goal. Examples:\n\n- Extract a small helper \u002F function when the change introduced duplication in the same file\n- Rename a confusing local symbol or tighten types in the files already modified\n- Simplify a conditional \u002F remove dead code encountered while making the change\n- Consolidate duplicated logic within the touched module(s) when it reduces future churn\n\n**Broad rewrite (avoid)** includes scope that changes the \"shape\" of the system beyond what the PR set out to do, such as:\n\n- Large file moves\u002Frenames or sweeping formatting churn across many files\n- Sweeping API changes (especially public\u002Fshared interfaces) just to \"clean things up\"\n- Re-architecting multiple systems\u002Fmodules as part of a small feature\u002Fbugfix\n- Wide refactors that require updating many call sites unrelated to the original change\n\n**Decision rule (guardrail)**: If refactoring would expand beyond the PR's change intent (e.g., many unrelated files, new cross-cutting abstractions, or broad API changes), pause and escalate with options (including capturing as a `tech-debt` issue for a separate, dedicated PR) and a recommended choice.\n\n## Related Guidance\n\n- Load `software-architecture` before extractions or structural splits\n- Load `systematic-debugging` when a refactor exposes a failing behavior and the root cause is unclear\n\n## Gotchas\n\n| Trigger                                        | Gotcha                                                                   | Fix                                                                   |\n| ---------------------------------------------- | ------------------------------------------------------------------------ | --------------------------------------------------------------------- |\n| Saying \"no refactor needed\" after a quick skim | Obvious extractions and duplication survive because analysis was shallow | Check file size, duplication, and function complexity before deciding |\n\n| Trigger                                   | Gotcha                                                    | Fix                                                                  |\n| ----------------------------------------- | --------------------------------------------------------- | -------------------------------------------------------------------- |\n| Treating unused new data as later cleanup | The PR ships an incomplete feature under a refactor label | Identify the missing integration point and fix it in the same change |\n\n## Frame Ports Filled By This Skill\n\n| Port | Work adapter | Auto-N\u002FA adapter | Explicit-skip adapter |\n| --- | --- | --- | --- |\n| `implement-refactor` | [adapters\u002Fimplement-refactor-adapter.md](adapters\u002Fimplement-refactor-adapter.md) | [adapters\u002Fimplement-refactor-auto-na-adapter.md](adapters\u002Fimplement-refactor-auto-na-adapter.md) | [adapters\u002Fimplement-refactor-explicit-skip-adapter.md](adapters\u002Fimplement-refactor-explicit-skip-adapter.md) |\n","\u003C!-- platform-assumptions: markdown skill guidance for VS Code custom agents in Agent Orchestra; assumes repository refactors are validated against existing tests and architecture rules. -->\n\u003C!-- markdownlint-disable-file MD041 MD003 -->\n\n# Refactoring Methodology\n\nReusable methodology for behavior-preserving structural improvement.\n\n## When to Use\n\n- When recently modified files need a proactive refactoring pass\n- When duplication, oversized functions, or structural debt may have been introduced or exposed\n- When integration gaps or symmetric mechanical fixes should be checked while the relevant code is already open\n- When refactoring decisions need a repeatable analysis and verification flow\n\n## Purpose\n\nFind real maintainability improvements in touched code without widening into speculative rewrites. Refactoring here is an active search for cleaner structure, not a passive approval pass.\n\n## Analysis Workflow\n\n1. Inspect every file modified by the current change unless the plan explicitly marks it out of scope.\n2. Check file size, function size, duplication, conditional complexity, and extraction opportunities.\n3. Inspect immediate neighbors when the touched file clearly participates in a broader repeated pattern.\n4. Check for unused added data or unwired integration points and treat them as incomplete work, not future debt.\n5. Before closing, grep for symmetric occurrences when the fix pattern is mechanical and repo-wide.\n6. Validate that the refactor preserved behavior and respected project architecture rules.\n\n## Mandatory Checks\n\n### Size And Structure\n\n- Check line count for each modified file\n- Flag files above 80% of the configured limit\n- Treat functions above roughly 50 lines as extraction candidates\n- Split responsibilities when a class or file is becoming multipurpose\n\n### Duplication And Readability\n\n- Extract repeated blocks or repeated control-flow fragments\n- Replace copied formulas or validations with delegation to one source of truth\n- Rename unclear variables and replace magic numbers with named constants\n- Simplify long expressions or nested conditionals when comprehension is suffering\n\n### Testability And Architecture\n\n- Prefer extracting private logic into testable units when behavior stays covered\n- Inject dependencies instead of copying behavior into new classes or helpers\n- Review layer placement and composition using `software-architecture` before significant extraction\n\n## Integration Gap Rule\n\nIf the change adds data, metadata, or configuration that is not consumed where the feature requires it, treat that as incomplete implementation. Fix it now when the integration is local and bounded; only defer when the remaining work truly requires a larger design decision.\n\n## Proactive Hunting Stance\n\nDo not rubber-stamp a touched file as \"fine\" without inspection.\n\n- Presume the touched code is improvable until checked.\n- Hunt for duplication, unclear naming, oversized units, and awkward control flow.\n- Apply the Boy Scout Rule within the touched area and its immediate neighbors.\n- If no improvements are needed, state what you checked and why.\n\nSuccess means finding proportionate maintainability improvements when they exist, not refactoring for its own sake.\n\n## Integration Gaps Are Not Tech Debt\n\nIf a PR adds data that is not consumed where the feature requires it, treat that as incomplete implementation, not future cleanup.\n\nExamples:\n\n- `supportedRegions` is added but consumers do not filter by it\n- `TIER_MULTIPLIERS` exists but the pricing path does not apply it\n- priority metadata is added but the scheduler ignores it\n- one map gains new entries while symmetric related maps do not\n\nWhen you find unused new data:\n\n1. Identify where it should be used.\n2. Estimate the integration effort.\n3. If the fix is local and bounded, include it now.\n4. If it is genuinely larger, document why it is large rather than calling it generic tech debt.\n\n## Output Structure\n\nUse a concise analysis with these sections when reporting refactor work:\n\n- `## Refactoring Analysis`\n- `### Files Analyzed`\n- `### Opportunities Found`\n- `### Actions Taken`\n- `### Deferred (Out of Scope)`\n- `### Verification`\n\n## Verification\n\n- Run the project-configured validation and test commands after refactoring\n- Confirm coverage expectations still hold for the affected behavior\n- Re-check the modified files for any remaining symmetric mechanical misses\n\n## Conductor Integration\n\nIn the Code-Conductor hub flow, **ALWAYS call Refactor-Specialist after Code-Smith completes.** Hub-mode Code-Conductor still runs the mandatory Refactor-Specialist pass.\n\nIn `\u002Fspine-run`, the `implement-refactor` port resolves through [skills\u002Frefactoring-methodology\u002Fadapters\u002Fimplement-refactor-adapter.md](adapters\u002Fimplement-refactor-adapter.md), and predicate adapters may resolve the port as auto-N\u002FA or explicit skip before the work adapter runs.\n\nRefactor-Specialist will:\n\n1. Analyze all files modified in the PR\n2. Hunt proactively for improvement opportunities\n3. Report findings (even if no action taken)\n4. Make improvements where beneficial\n\n**In the Code-Conductor hub flow, there is no \"skip refactoring\" option.** The Refactor-Specialist decides what needs improvement, not the plan or Code-Conductor.\n\n**Flow**: Code-Smith -> Refactor-Specialist -> Code-Critic\n\n**Clarification**: \"Avoid broad rewrites\" does NOT mean \"skip refactoring\" - it means keep refactoring proportionate to the PR's intent.\n\n**Proportionate refactoring (good)** means improving code you already touched (or its immediate neighbors) to reduce complexity\u002Fduplication without expanding the PR's goal. Examples:\n\n- Extract a small helper \u002F function when the change introduced duplication in the same file\n- Rename a confusing local symbol or tighten types in the files already modified\n- Simplify a conditional \u002F remove dead code encountered while making the change\n- Consolidate duplicated logic within the touched module(s) when it reduces future churn\n\n**Broad rewrite (avoid)** includes scope that changes the \"shape\" of the system beyond what the PR set out to do, such as:\n\n- Large file moves\u002Frenames or sweeping formatting churn across many files\n- Sweeping API changes (especially public\u002Fshared interfaces) just to \"clean things up\"\n- Re-architecting multiple systems\u002Fmodules as part of a small feature\u002Fbugfix\n- Wide refactors that require updating many call sites unrelated to the original change\n\n**Decision rule (guardrail)**: If refactoring would expand beyond the PR's change intent (e.g., many unrelated files, new cross-cutting abstractions, or broad API changes), pause and escalate with options (including capturing as a `tech-debt` issue for a separate, dedicated PR) and a recommended choice.\n\n## Related Guidance\n\n- Load `software-architecture` before extractions or structural splits\n- Load `systematic-debugging` when a refactor exposes a failing behavior and the root cause is unclear\n\n## Gotchas\n\n| Trigger                                        | Gotcha                                                                   | Fix                                                                   |\n| ---------------------------------------------- | ------------------------------------------------------------------------ | --------------------------------------------------------------------- |\n| Saying \"no refactor needed\" after a quick skim | Obvious extractions and duplication survive because analysis was shallow | Check file size, duplication, and function complexity before deciding |\n\n| Trigger                                   | Gotcha                                                    | Fix                                                                  |\n| ----------------------------------------- | --------------------------------------------------------- | -------------------------------------------------------------------- |\n| Treating unused new data as later cleanup | The PR ships an incomplete feature under a refactor label | Identify the missing integration point and fix it in the same change |\n\n## Frame Ports Filled By This Skill\n\n| Port | Work adapter | Auto-N\u002FA adapter | Explicit-skip adapter |\n| --- | --- | --- | --- |\n| `implement-refactor` | [adapters\u002Fimplement-refactor-adapter.md](adapters\u002Fimplement-refactor-adapter.md) | [adapters\u002Fimplement-refactor-auto-na-adapter.md](adapters\u002Fimplement-refactor-auto-na-adapter.md) | [adapters\u002Fimplement-refactor-explicit-skip-adapter.md](adapters\u002Fimplement-refactor-explicit-skip-adapter.md) |\n",{"name":5,"description":7},"main","d5486aacc32c9ca1b07173cfd530f8fe75c29e58","skills\u002Frefactoring-methodology",0,"https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra","https:\u002F\u002Fgithub.com\u002FGrimblaz","2026-08-16 01:12:46",{"source":8,"skillName":5,"commands":46,"prompt":56,"zipAvailable":57},[47,50,53],{"tool":48,"command":49},"npx","npx skills add Grimblaz\u002Fagent-orchestra --skill refactoring-methodology",{"tool":51,"command":52},"bunx","bunx skills add Grimblaz\u002Fagent-orchestra --skill refactoring-methodology",{"tool":54,"command":55},"pnpm","pnpm dlx skills add Grimblaz\u002Fagent-orchestra --skill refactoring-methodology","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Frefactoring-methodology\u002FSKILL.md",true,[59,65,69],{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"adapters\u002Fimplement-refactor-adapter.md","REFERENCE","参考文档",3800,"https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Frefactoring-methodology\u002Fadapters\u002Fimplement-refactor-adapter.md",{"path":66,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":67,"sourceHtmlUrl":68},"adapters\u002Fimplement-refactor-auto-na-adapter.md",302,"https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Frefactoring-methodology\u002Fadapters\u002Fimplement-refactor-auto-na-adapter.md",{"path":70,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":71,"sourceHtmlUrl":72},"adapters\u002Fimplement-refactor-explicit-skip-adapter.md",347,"https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Frefactoring-methodology\u002Fadapters\u002Fimplement-refactor-explicit-skip-adapter.md",[74,105,129,144,170,191],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":86,"publishedAt":33},"01M6C48YMJVFHM7DVGG1WWW5RJ","webapp-testing","Playwright end-to-end testing guidance for web apps, with practical patterns and setup steps. Use when creating or improving browser-based E2E coverage, test stability, and CI execution. DO NOT USE FOR: React component tests (use ui-testing), canvas game interaction (use browser-canvas-testing), TDD workflow (use test-driven-development), debugging failures (use systematic-debugging), or randomized property tests (use property-based-testing).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fwebapp-testing\u002FSKILL.md",[80,83],{"categoryNo":81,"categoryName":82,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":84,"categoryName":85,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000009","前端开发",[87,88,89,90,91,27,92,93,94,95,96,97,98,99,100,101,102,103,104,31],"testing","webapp testing","webapp","canvas","tdd","playwright","test","web","react","component","debugging","development","browser","ci","execution","e2e","game","coverage",{"skillNo":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":33},"01M6C48TNNP2KCSQXV8YK7SK49","verification-before-completion","Evidence-based verification checklist before marking work complete. Use before PRs, releases, marking tickets done, or any \"I'm finished\" declaration; also when naming a baseline commit for a differential suite claim, when designing a mutation campaign or red-run evidence, when making an absence claim, a universal claim, or a one-time demonstration offered as a standing guarantee, when attributing a measurement or claiming a sweep or verification loop is complete, and when stating what a check or a CI suite will do about a change. DO NOT USE FOR: post-merge cleanup or archival (use post-pr-review) or processing GitHub review comments (use code-review-intake).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fverification-before-completion\u002FSKILL.md",[],[112,113,26,114,115,116,117,118,119,120,121,122,100,123,124,125,126,31,127,128],"verification","verification before completion","change","commit","merge","pr","cleanup","run","code","code-review","intake","loop","verification loop","work","check","evidence","campaign",{"skillNo":130,"name":131,"nameZh":6,"description":132,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":33},"01M6C48QZWNTP7JGKD32QZAKKX","validation-methodology","Reusable validation and review methodology for staged validation, failure triage, and prosecution-depth setup. Use when running validation ladders, triaging failures, or executing adversarial review passes. DO NOT USE FOR: CE Gate orchestration, specialist dispatch ownership, or step execution flow (keep those in Code-Conductor).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fvalidation-methodology\u002FSKILL.md",[],[136,23,137,26,138,139,140,141,120,142,101,31,143],"validation","validation methodology","triage","orchestration","setup","adversarial review","flow","dispatch",{"skillNo":145,"name":146,"nameZh":6,"description":147,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":148,"categories":149,"keywords":156,"publishedAt":33},"01M6C48MCAHG5Z79BJ369CRYRZ","upstream-onboarding","Shared opening-phase protocol for upstream agents (Experience-Owner, Solution-Designer, Issue-Planner) and Code-Conductor when invoked on an existing GitHub issue. Renders a scaled context brief and runs a standards check on inherited work at each phase boundary. Use when a user-invocable agent receives a request referencing an existing GitHub issue and must load solution-authoring and upstream-onboarding at the opening phase. DO NOT USE FOR: subagent dispatches (which already operate within an assessed session context); post-merge review (use post-pr-review); research or non-tree subagents.","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fupstream-onboarding\u002FSKILL.md",[150,151],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":152,"categoryName":153,"taxonomyType":154,"taxonomyTypeDesc":155},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[157,158,26,159,160,116,161,117,120,162,163,164,165,166,167,125,126,168,169,31],"onboarding","upstream onboarding","github","research","standards check","protocol","brief","boundary","designer","subagent","planner","user","session",{"skillNo":171,"name":172,"nameZh":6,"description":173,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":174,"categories":175,"keywords":184,"publishedAt":33},"01M6C48GPVE120TGPMD1CS365H","ui-testing","Resilient React component testing strategies focusing on user behavior. Use when writing or reviewing component-level React tests, fixing flaky component tests, or establishing React testing patterns. DO NOT USE FOR: Playwright E2E tests (use webapp-testing), canvas game interaction (use browser-canvas-testing), TDD workflow (use test-driven-development), debugging failures (use systematic-debugging), UI component design (use frontend-design), or randomized property tests (use property-based-testing).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fui-testing\u002FSKILL.md",[176,177,178,181],{"categoryNo":81,"categoryName":82,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":84,"categoryName":85,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":179,"categoryName":180,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":182,"categoryName":183,"taxonomyType":154,"taxonomyTypeDesc":155},"01JBSEEDCAT000000000000032","设计师",[87,185,186,187,188,189,90,27,92,93,95,96,97,98,99,190,102,103,31,89],"ui","ui testing","frontend","design","writing","testing patterns",{"skillNo":192,"name":193,"nameZh":6,"description":194,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":195,"categories":196,"keywords":199,"publishedAt":33},"01M6C48DRJ9XBDR2FW0W37DY53","ui-iteration","Reusable screenshot-driven UI polish workflow for iterative visual refinement. Use when improving layout, hierarchy, spacing, feedback, or Tailwind-based presentation through repeated review-and-adjust passes. DO NOT USE FOR: functional bug fixing, accessibility-only audits, or major redesign strategy work (use Code-Smith or frontend-design).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fui-iteration\u002FSKILL.md",[197,198],{"categoryNo":84,"categoryName":85,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":182,"categoryName":183,"taxonomyType":154,"taxonomyTypeDesc":155},[185,200,187,188,26,27,201,120,202,203,204,205,206,125,207,208,31,209],"ui iteration","accessibility","screenshot","Tailwind","polish","ui polish","feedback","redesign","strategy","only"]