[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71R6AXZ705VV2M222XJ6JR8":3,"skill-same:01M71R6AXZ705VV2M222XJ6JR8":63},{"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":17,"publishedAt":29,"linkOnly":30,"markdownContent":31,"markdownBody":32,"frontMatter":33,"refName":37,"commitSha":38,"skillDirectory":39,"forksCount":13,"repositoryHtmlUrl":40,"repositoryHomepageUrl":6,"ownerHtmlUrl":41,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":42,"install":43,"resources":56},"01M71R6AXZ705VV2M222XJ6JR8","subagent-isolation",null,"Clean context boundaries between agents - prevent hallucination cascades, cross-contamination, enforce error isolation.","LuisAlbertoMK\u002Fgentleman-agent-gh","LuisAlbertoMK","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F96153499?v=4","01M71QZX0CF9KKGQZ8ZN9BME63","01M71QZX11P5X6Z27HE1GQDF71",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsubagent-isolation\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28],"subagent isolation","subagent","isolation","error","enforce","clean","clean context","contamination","docs","context","cascades","2026-11-10 13:02:02",false,"---\nname: subagent-isolation\ndescription: \"Clean context boundaries between agents - prevent hallucination cascades, cross-contamination, enforce error isolation.\"\ntriggers: \"Subagent isolation, context boundaries\"\nchangelog: docs\u002Fciclos\u002Fcycle28-20260815.md\ntoken_budget: 3100\n---\n\n## When to Use\nMaintain clean context boundaries between delegated agents —\n\n## Isolation Rules\n### Fresh context per delegation\nEach `delegate` starts CLEAN. Include ALL context needed. Reference Engram IDs. File context -> paths + what to look for.\n\n### No cross-contamination\n| Rule | Why |\n|------|-----|\n| Never share state between subagents | Prevents hallucination cascades |\n| Independent tool access per subagent | Error isolation |\n| Serialize if B depends on A | Don't parallelize dependent work |\n\n### Dependency declaration\nDeclare what subagent needs (file paths, Engram IDs). NOT: full history, system prompt.\n\n### Result isolation\nEach delegate returns OWN output. Conflicts -> raise to orchestrator. Subagents NEVER modify global state without explicit instructions.\n\n### Preservation contract\nEvery delegation output MUST include this 4-field block AS-IS (never summarized):\n```\n## Decision Taken    ## Files Changed\n## Key Findings      ## Nuance (what detail would be lost if summarized)\n```\n\n### Error boundaries\n| Error | Action |\n|-------|--------|\n| Timeout | Retry once with cleaner prompt, then escalate |\n| Wrong output | Log to Engram, re-delegate with corrected context |\n| Hallucinates | Flag as contamination -> check isolation rules |\n\n## Anti-Rationalization\n\n| Rationalization | Red Flag | Verification |\n|-----------------|----------|--------------|\n| \"compartir contexto entre subagentes ahorra tokens\" | Reutilizar historial completo entre delegations | Cada delegate arranca CLEAN: solo paths + Engram IDs necesarios, nunca historial completo |\n| \"hallucination cascade se detecta sola\" | No aislar, asumir errores no se propagan | Aplicar isolation rules: no shared state, error boundaries (timeout→retry once, hallucinate→flag contamination) |\n| \"un subagente puede editar archivos de otro sin coordinar\" | Overlap de escritura sin dependency graph | Verificar no file overlap antes de paralelizar; si B depende de A → serial, conflictos → escalar a orchestrator |\n\n## Red Flags\n- Doing work without checking output format → STOP, re-read skill\n- Second occurrence of same rationalization → force RED zone\n\n## Verification\n- Output matches skill ## Output contract + file:line citaton\n- cross-ref-check.ps1 → SKILL.md OK\n## Refs\ndelivery-harness · command-wrapper · lean-context · context-watchdog · execution-mode\n\n## Anti-Patterns\nShare state between subagents · Parallelize dependent tasks · Skip 4-field contract · Keep full subagent output in context · No error isolation\n\n---\n\n> See [reference.md](docs\u002Fskills\u002Fsubagent-isolation\u002Freference.md) for extended details, examples, and detailed patterns.\n\n","## When to Use\nMaintain clean context boundaries between delegated agents —\n\n## Isolation Rules\n### Fresh context per delegation\nEach `delegate` starts CLEAN. Include ALL context needed. Reference Engram IDs. File context -> paths + what to look for.\n\n### No cross-contamination\n| Rule | Why |\n|------|-----|\n| Never share state between subagents | Prevents hallucination cascades |\n| Independent tool access per subagent | Error isolation |\n| Serialize if B depends on A | Don't parallelize dependent work |\n\n### Dependency declaration\nDeclare what subagent needs (file paths, Engram IDs). NOT: full history, system prompt.\n\n### Result isolation\nEach delegate returns OWN output. Conflicts -> raise to orchestrator. Subagents NEVER modify global state without explicit instructions.\n\n### Preservation contract\nEvery delegation output MUST include this 4-field block AS-IS (never summarized):\n```\n## Decision Taken    ## Files Changed\n## Key Findings      ## Nuance (what detail would be lost if summarized)\n```\n\n### Error boundaries\n| Error | Action |\n|-------|--------|\n| Timeout | Retry once with cleaner prompt, then escalate |\n| Wrong output | Log to Engram, re-delegate with corrected context |\n| Hallucinates | Flag as contamination -> check isolation rules |\n\n## Anti-Rationalization\n\n| Rationalization | Red Flag | Verification |\n|-----------------|----------|--------------|\n| \"compartir contexto entre subagentes ahorra tokens\" | Reutilizar historial completo entre delegations | Cada delegate arranca CLEAN: solo paths + Engram IDs necesarios, nunca historial completo |\n| \"hallucination cascade se detecta sola\" | No aislar, asumir errores no se propagan | Aplicar isolation rules: no shared state, error boundaries (timeout→retry once, hallucinate→flag contamination) |\n| \"un subagente puede editar archivos de otro sin coordinar\" | Overlap de escritura sin dependency graph | Verificar no file overlap antes de paralelizar; si B depende de A → serial, conflictos → escalar a orchestrator |\n\n## Red Flags\n- Doing work without checking output format → STOP, re-read skill\n- Second occurrence of same rationalization → force RED zone\n\n## Verification\n- Output matches skill ## Output contract + file:line citaton\n- cross-ref-check.ps1 → SKILL.md OK\n## Refs\ndelivery-harness · command-wrapper · lean-context · context-watchdog · execution-mode\n\n## Anti-Patterns\nShare state between subagents · Parallelize dependent tasks · Skip 4-field contract · Keep full subagent output in context · No error isolation\n\n---\n\n> See [reference.md](docs\u002Fskills\u002Fsubagent-isolation\u002Freference.md) for extended details, examples, and detailed patterns.\n\n",{"name":5,"triggers":34,"changelog":35,"description":7,"token_budget":36},"Subagent isolation, context boundaries","docs\u002Fciclos\u002Fcycle28-20260815.md",3100,"main","4af6946b30783e2c8f9182082324d530fc28f838",".agents\u002Fskills\u002Fsubagent-isolation","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK","2026-11-10 18:34:27",{"source":8,"skillName":5,"commands":44,"prompt":54,"zipAvailable":55},[45,48,51],{"tool":46,"command":47},"npx","npx skills add LuisAlbertoMK\u002Fgentleman-agent-gh --skill subagent-isolation",{"tool":49,"command":50},"bunx","bunx skills add LuisAlbertoMK\u002Fgentleman-agent-gh --skill subagent-isolation",{"tool":52,"command":53},"pnpm","pnpm dlx skills add LuisAlbertoMK\u002Fgentleman-agent-gh --skill subagent-isolation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsubagent-isolation\u002FSKILL.md",true,[57],{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002Fdelegation-template.md","REFERENCE","参考文档",869,"https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsubagent-isolation\u002Freferences\u002Fdelegation-template.md",[64,107,130,150,175,201],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":86,"publishedAt":29},"01M71R7K35P668DYEPYGDQP2QA","web-quality-audit","Comprehensive web audit: performance, a11y, SEO, responsive, animation, design tokens.","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002Fdocs\u002Fciclos\u002Fc28-w1-backup\u002Fweb-quality-audit\u002FSKILL.md",[70,75,78,83],{"categoryNo":71,"categoryName":72,"taxonomyType":73,"taxonomyTypeDesc":74},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",{"categoryNo":76,"categoryName":77,"taxonomyType":73,"taxonomyTypeDesc":74},"01JBSEEDCAT000000000000019","营销推广",{"categoryNo":79,"categoryName":80,"taxonomyType":81,"taxonomyTypeDesc":82},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",{"categoryNo":84,"categoryName":85,"taxonomyType":81,"taxonomyTypeDesc":82},"01JBSEEDCAT000000000000035","市场营销",[87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106],"quality","quality audit","web quality audit","audit","web","Animation","design audit","design","website","ui","responsive","seo","tokens","ui audit","Lighthouse","performance","review","optimize","a11y","page",{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":114,"publishedAt":29},"01M71R7H1RA7CQ568DHRK37PE2","ui-engine","UI system — Grid\u002FFlexbox\u002F@layer\u002F:has(), container queries, compositor-only animation, OKLCH tokens, component patterns","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002Fdocs\u002Fciclos\u002Fc28-w1-backup\u002Fui-engine\u002FSKILL.md",[113],{"categoryNo":79,"categoryName":80,"taxonomyType":81,"taxonomyTypeDesc":82},[115,116,96,92,117,118,119,106,120,99,121,122,123,94,124,125,126,127,128,129],"ui engine","engine","compositor","css","dark","system","mode","hooks","container","only","query","components","dark mode","grid","queries",{"skillNo":131,"name":132,"nameZh":6,"description":133,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":134,"categories":135,"keywords":138,"publishedAt":29},"01M71R7EPMZFKB833WE4B3GTHC","baseline-ui","Anti-slop UI — layout, typography, responsive, animation, tokens. Use for cleanup or polish.","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002Fdocs\u002Fciclos\u002Fc28-w1-backup\u002Fbaseline-ui\u002FSKILL.md",[136,137],{"categoryNo":71,"categoryName":72,"taxonomyType":73,"taxonomyTypeDesc":74},{"categoryNo":79,"categoryName":80,"taxonomyType":81,"taxonomyTypeDesc":82},[139,140,96,92,141,142,143,144,99,145,123,94,146,147,148,100,149,125,128,103],"baseline ui","baseline","generic","slop","for","anti-slop","fix","interface","cleanup","design review","use",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":159,"publishedAt":29},"01M71R7CWK3FBW77Z8WD9GXQFE","skill-creator","Create new AI agent skills following Agent Skills spec -- auto-bootstrap unknown tasks, define structure, register in AGENTS.md","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002F.archive\u002Fskills\u002Fskill-creator\u002FSKILL.md",[156],{"categoryNo":157,"categoryName":158,"taxonomyType":73,"taxonomyTypeDesc":74},"01JBSEEDCAT000000000000023","Skill与智能体开发",[160,161,162,163,164,165,166,167,168,169,170,171,172,173,174],"skill creator","creator","create","structure","register","2026","spec","new","ai skill","auto","define","ai agent","unknown","tasks","bootstrap",{"skillNo":176,"name":177,"nameZh":6,"description":178,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":179,"categories":180,"keywords":183,"publishedAt":29},"01M71R78DZM1GA16F1KN8ND926","senior-engineer","Apply senior engineering competencies — system design, trade-off analysis, delegation, mentoring, and tech debt strategy","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002F.archive\u002Fskills\u002Fsenior-engineer\u002FSKILL.md",[181,182],{"categoryNo":71,"categoryName":72,"taxonomyType":73,"taxonomyTypeDesc":74},{"categoryNo":79,"categoryName":80,"taxonomyType":81,"taxonomyTypeDesc":82},[184,185,186,187,188,189,190,94,191,192,165,193,194,195,120,196,197,198,199,200],"engineer","senior","senior engineer","system design","trade-offs","senior engineering","delegation","senior architect","analysis","debt","architect","tech","strategy","apply","trade","tech debt","engineering",{"skillNo":202,"name":203,"nameZh":6,"description":204,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":205,"categories":206,"keywords":211,"publishedAt":29},"01M71R74WQTXZMG3219RND96CE","prompt-engineering","Design effective prompts using SPEARS framework — scope, principles, examples, assertions, refinements, with ReAct and multi-agent patterns","https:\u002F\u002Fgithub.com\u002FLuisAlbertoMK\u002Fgentleman-agent-gh\u002Fblob\u002Fmain\u002F.archive\u002Fskills\u002Fprompt-engineering\u002FSKILL.md",[207,210],{"categoryNo":208,"categoryName":209,"taxonomyType":73,"taxonomyTypeDesc":74},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":157,"categoryName":158,"taxonomyType":73,"taxonomyTypeDesc":74},[212,213,200,214,215,94,216,165,217,218,219,220,221,222,223,224,225,226,227,228],"prompt","prompt engineering","scope","improve","react","security","patterns","multi-agent","framework","agent patterns","assertions","examples","with","multi","principles","improve prompt","prompts"]