[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78EXZX4GR65QDANT77BCAN2":3,"skill-same:01M78EXZX4GR65QDANT77BCAN2":45},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M78EXZX4GR65QDANT77BCAN2","build-orchestration",null,"Drives an approved IDSD run by delegating deterministic state transitions to the build orchestrator, routing actions to specialist agents\u002Fcommands, and recording structured evidence.","Faraaz54\u002Fagent_harness","Faraaz54","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F21173756?v=4","01M78EXNGY3KPCX396VRDE8T8P","Mohd Faraaz","01M78EXNHJ6JFPC63B0FN6PSWY",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002F.cursor\u002Fskills\u002Fbuild-orchestration\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: build-orchestration\ndescription: Drives an approved IDSD run by delegating deterministic state transitions to the build orchestrator, routing actions to specialist agents\u002Fcommands, and recording structured evidence.\n---\n\n# Build Orchestration\n\n## Purpose\n\nCoordinate the approved implementation loop. Do not implement, review, validate, or commit directly. The orchestrator owns state; specialist agents own lifecycle work.\n\n## Required inputs\n\n- `AGENTS.md`\n- `harness.config.json`\n- `docs\u002Fbootstrap\u002Fenvironment.md`\n- approved Intent, Context, Expectations, and task contracts\n- `tasks\u002Frun_manifest.json`\n- `tasks\u002Frun_state.json`\n- `tasks\u002Ffeature_list.json`\n- active project-pack README and public review criteria when configured\n\n## Method\n\n1. Run `python -B scripts\u002Fbuild_orchestrator.py status`.\n2. If no approved run exists, stop.\n3. If bootstrap is stale, route `REFRESH_BOOTSTRAP`.\n4. Ask the deterministic controller for the next action.\n5. Execute exactly that action with the specified agent\u002Fcommand.\n6. Require a structured result artifact for every agent action.\n7. Record the result using `record-result`.\n8. Repeat until terminal.\n\n## Action routing\n\n| Action | Route |\n|---|---|\n| `REFRESH_BOOTSTRAP` | `\u002Fbootstrap-environment` |\n| `RUN_PREFLIGHT` | `\u002Fstart-run` |\n| `IMPLEMENT_TASK` | `build-agent` |\n| `REPAIR_TASK` | `build-agent` with repair packet |\n| `RUN_POSTFLIGHT` | `\u002Fpostflight` |\n| `RUN_REVIEW:test-agent` | `test-agent` |\n| `RUN_REVIEW:principal-engineer-agent` | `principal-engineer-agent` |\n| `RUN_DOMAIN_REVIEW` | `domain-reviewer-agent` |\n| `RUN_VALIDATOR` | `validator-agent` |\n| `COMMIT_TASK` | `\u002Fcommit-task` |\n| `FINALIZE_SESSION` | `\u002Ffinalize-session` |\n| `COMPLETE` | stop and summarize |\n\n## State discipline\n\n- Conversation text is never durable state.\n- Do not update task status manually unless a controller command explicitly asks for that artifact.\n- Every action receives an action ID.\n- Every action result must reference the same run ID, task ID, and attempt.\n- Do not continue from an action whose result failed schema or consistency checks.\n\n## Repair discipline\n\n- Classify failure before repair.\n- Consolidate findings across test\u002Fprincipal\u002Fdomain\u002Fvalidator gates.\n- After repair, previous postflight, review, and validator evidence is stale.\n- Stop when the repair budget is exhausted.\n- Update repair memory for repeated defects.\n\n## Stop conditions\n\n- missing required artifact;\n- project-pack config cannot be loaded;\n- required reviewer unavailable;\n- manifest\u002Fsource hash mismatch;\n- task contract conflict;\n- private eval requested by builder;\n- repair budget exhausted;\n- branch\u002Fworktree\u002Fpreflight failure;\n- command would require remote\u002Fprod authority not present.\n\n## Verification\n\n- [ ] Every action came from `build_orchestrator.py next`.\n- [ ] Every agent used appropriate skills.\n- [ ] Every result was recorded.\n- [ ] No role self-approved.\n- [ ] Build progressed only through durable evidence.\n\n\n## Model routing\n\nThe orchestrator must select the model alias from `harness.config.json:model_routing` for every action. `\u002Fbuild-auto` itself should use `orchestration_large`. Implementation actions should use `execution_medium` unless risk\u002Fescalation policy says otherwise. Test and principal review should use `review_reasoning`; validator should use `validation_reasoning`; domain review should use `domain_reasoning`.\n\nEvery action packet must include `model_routing` with alias\u002Fprovider\u002Fmodel\u002Freasoning metadata. Every structured result should record the actual model used and any fallback.\n\nDo not silently downgrade validation\u002Freview on high-risk tasks. Prefer independent model aliases for build, review and validation.\n","# Build Orchestration\n\n## Purpose\n\nCoordinate the approved implementation loop. Do not implement, review, validate, or commit directly. The orchestrator owns state; specialist agents own lifecycle work.\n\n## Required inputs\n\n- `AGENTS.md`\n- `harness.config.json`\n- `docs\u002Fbootstrap\u002Fenvironment.md`\n- approved Intent, Context, Expectations, and task contracts\n- `tasks\u002Frun_manifest.json`\n- `tasks\u002Frun_state.json`\n- `tasks\u002Ffeature_list.json`\n- active project-pack README and public review criteria when configured\n\n## Method\n\n1. Run `python -B scripts\u002Fbuild_orchestrator.py status`.\n2. If no approved run exists, stop.\n3. If bootstrap is stale, route `REFRESH_BOOTSTRAP`.\n4. Ask the deterministic controller for the next action.\n5. Execute exactly that action with the specified agent\u002Fcommand.\n6. Require a structured result artifact for every agent action.\n7. Record the result using `record-result`.\n8. Repeat until terminal.\n\n## Action routing\n\n| Action | Route |\n|---|---|\n| `REFRESH_BOOTSTRAP` | `\u002Fbootstrap-environment` |\n| `RUN_PREFLIGHT` | `\u002Fstart-run` |\n| `IMPLEMENT_TASK` | `build-agent` |\n| `REPAIR_TASK` | `build-agent` with repair packet |\n| `RUN_POSTFLIGHT` | `\u002Fpostflight` |\n| `RUN_REVIEW:test-agent` | `test-agent` |\n| `RUN_REVIEW:principal-engineer-agent` | `principal-engineer-agent` |\n| `RUN_DOMAIN_REVIEW` | `domain-reviewer-agent` |\n| `RUN_VALIDATOR` | `validator-agent` |\n| `COMMIT_TASK` | `\u002Fcommit-task` |\n| `FINALIZE_SESSION` | `\u002Ffinalize-session` |\n| `COMPLETE` | stop and summarize |\n\n## State discipline\n\n- Conversation text is never durable state.\n- Do not update task status manually unless a controller command explicitly asks for that artifact.\n- Every action receives an action ID.\n- Every action result must reference the same run ID, task ID, and attempt.\n- Do not continue from an action whose result failed schema or consistency checks.\n\n## Repair discipline\n\n- Classify failure before repair.\n- Consolidate findings across test\u002Fprincipal\u002Fdomain\u002Fvalidator gates.\n- After repair, previous postflight, review, and validator evidence is stale.\n- Stop when the repair budget is exhausted.\n- Update repair memory for repeated defects.\n\n## Stop conditions\n\n- missing required artifact;\n- project-pack config cannot be loaded;\n- required reviewer unavailable;\n- manifest\u002Fsource hash mismatch;\n- task contract conflict;\n- private eval requested by builder;\n- repair budget exhausted;\n- branch\u002Fworktree\u002Fpreflight failure;\n- command would require remote\u002Fprod authority not present.\n\n## Verification\n\n- [ ] Every action came from `build_orchestrator.py next`.\n- [ ] Every agent used appropriate skills.\n- [ ] Every result was recorded.\n- [ ] No role self-approved.\n- [ ] Build progressed only through durable evidence.\n\n\n## Model routing\n\nThe orchestrator must select the model alias from `harness.config.json:model_routing` for every action. `\u002Fbuild-auto` itself should use `orchestration_large`. Implementation actions should use `execution_medium` unless risk\u002Fescalation policy says otherwise. Test and principal review should use `review_reasoning`; validator should use `validation_reasoning`; domain review should use `domain_reasoning`.\n\nEvery action packet must include `model_routing` with alias\u002Fprovider\u002Fmodel\u002Freasoning metadata. Every structured result should record the actual model used and any fallback.\n\nDo not silently downgrade validation\u002Freview on high-risk tasks. Prefer independent model aliases for build, review and validation.\n",{"name":5,"description":7},"main","214a047388c6c9f45aa486a02cac0773e837351c",".cursor\u002Fskills\u002Fbuild-orchestration","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness","https:\u002F\u002Fgithub.com\u002FFaraaz54","Data Scientist working on agentic AI.","2026-07-13 20:46:30",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add Faraaz54\u002Fagent_harness --skill build-orchestration",{"tool":37,"command":38},"bunx","bunx skills add Faraaz54\u002Fagent_harness --skill build-orchestration",{"tool":40,"command":41},"pnpm","pnpm dlx skills add Faraaz54\u002Fagent_harness --skill build-orchestration","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002F.cursor\u002Fskills\u002Fbuild-orchestration\u002FSKILL.md",true,[],[46,53,60,67,74,81],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M78F0TRN2PGMG6H3FD2DHNM8","invoice-domain-review","Domain review for contractor invoice governance, ingestion, reconciliation, attendance attribution and billing evidence.","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002Fproject-packs\u002Finvoice-governance\u002Fskills\u002Finvoice-domain-review\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M78F0QRMKHN9W06VV38BSNB5","python-domain-review","Reviews generic Python implementation against task intent, API expectations, simplicity and maintainability.","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002Fproject-packs\u002Fgeneric-python\u002Fskills\u002Fpython-domain-review\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M78F0N3WMW64CZ3Z21NN73AH","verified-commit","Creates a task-scoped commit only after validator and exact-tree postflight pass.","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002F.cursor\u002Fskills\u002Fverified-commit\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M78F0K4H11BVVZEXD2MSYVET","using-agent-skills","Selects and applies the smallest relevant set of skills for a task, preventing generic prompting and unused specialist guidance. Use at the start of every command, task, review, or repair cycle.","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002F.cursor\u002Fskills\u002Fusing-agent-skills\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M78F0FV1TQGXRZT1NVWFJ9HM","testing-patterns","Designs focused, hierarchical, regression, boundary, integration, and end-to-end tests for vertical-slice implementation. Use before and after implementation, and during test review.","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002F.cursor\u002Fskills\u002Ftesting-patterns\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M78F0E17M758PBFBE59Q45E5","test-hierarchy","Classifies and enforces the required unit, integration, local E2E, and cloud E2E test tier for a task or run.","https:\u002F\u002Fgithub.com\u002FFaraaz54\u002Fagent_harness\u002Fblob\u002Fmain\u002F.cursor\u002Fskills\u002Ftest-hierarchy\u002FSKILL.md",[],[]]