[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MJHG8WHHNQC40XV437NWP8":3,"skill-same:01M6MJHG8WHHNQC40XV437NWP8":114},{"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":23,"publishedAt":45,"linkOnly":46,"markdownContent":47,"markdownBody":48,"frontMatter":49,"refName":90,"commitSha":91,"skillDirectory":92,"forksCount":93,"repositoryHtmlUrl":94,"repositoryHomepageUrl":95,"ownerHtmlUrl":96,"ownerBlogUrl":97,"ownerDescription":98,"updatedAt":99,"install":100,"resources":113},"01M6MJHG8WHHNQC40XV437NWP8","auto-pilot",null,"Intake-first autonomous project execution for Codex. Use when the user wants a short project request turned into a long-running build workflow with minimal interruptions, or explicitly mentions 오토파일럿\u002Fautopilot.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1575,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fminsu0707\u002Fauto-pilot\u002Fskills\u002Fauto-pilot\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],"auto pilot","auto","when","workflow","request","project execution","lock","mode","make","progress","app","project","autonomous","short","ask","save","start","shipping","build workflow","run","create","2026-11-10 10:12:30",false,"---\nname: auto-pilot\ndescription: Intake-first autonomous project execution for Codex. Use when the user wants a short project request turned into a long-running build workflow with minimal interruptions, or explicitly mentions 오토파일럿\u002Fautopilot.\nsummary: Ask once, lock spec, save state, keep shipping.\nmetadata:\n  priority: 8\n  promptSignals:\n    phrases:\n      - \"오토파일럿\"\n      - \"autopilot\"\n      - \"auto-pilot\"\n      - \"24시간 계속\"\n      - \"with ap\"\n      - \"using ap\"\n    allOf:\n      - [ap, build]\n      - [build, ap]\n      - [ap, make]\n      - [make, ap]\n      - [ap, create]\n      - [create, ap]\n      - [ap, app]\n      - [app, ap]\n      - [ap, 만들어줘]\n      - [만들어줘, ap]\n    anyOf:\n      - \"오토파일럿\"\n      - \"autopilot\"\n      - \"auto-pilot\"\n    minScore: 7\nretrieval:\n  aliases:\n    - 오토파일럿\n    - autopilot mode\n    - auto-pilot mode\n    - with ap\n    - intake first execution\n    - continuous project execution\n  intents:\n    - start an autopilot project\n    - run intake and keep building\n    - continue a project with minimal interruptions\n  entities:\n    - spec\n    - progress\n    - blockers\n    - state\n---\n\n# Auto Pilot\n\nUse this skill as the umbrella workflow for long-running project execution.\nTreat it as the **manager** role for the whole system.\n\n## Goal\n\nTurn a short product request into a resumable execution loop that:\n\n- gathers the minimum required inputs once\n- locks a written spec and definition of done\n- keeps implementing the next highest-priority work\n- escalates only for true human-only blockers\n\n## Operating Model\n\n### If state does not exist yet\n\n1. Route to the intake workflow first.\n2. Ask only for high-impact decisions.\n3. Infer safe defaults for the rest.\n4. After intake, detect whether the selected auth or managed services require upfront integration env values.\n5. If env values are already present, write:\n   - `docs\u002Fspec.md`\n   - `docs\u002Fdesign.md` for user-facing projects\n   - `docs\u002Fprogress.md`\n   - `docs\u002Fnext.md`\n   - `autopilot\u002Fstate.json`\n   - `autopilot\u002Fblockers.json`\n   - `autopilot\u002Fsecrets-status.json`\n6. If env values are still missing, treat the project as being in the normal `setup-secrets` phase and write:\n   - `docs\u002Fnext.md`\n   - `autopilot\u002Fstate.json`\n   - `autopilot\u002Fblockers.json`\n   - `autopilot\u002Fsecrets-status.json`\n   - request the consolidated env payload\n7. Start implementation immediately only after the required env values are ready.\n\n### If state already exists\n\n1. Read `autopilot\u002Fstate.json`, `autopilot\u002Fblockers.json`, and `autopilot\u002Fsecrets-status.json` if it exists.\n2. If `setupStatus` is `pending` or `autopilot\u002Fsecrets-status.json` is still pending, treat the project as an existing project waiting on setup continuation.\n3. In that setup continuation path, collect only the missing env payload and do not assume `docs\u002Fspec.md` or `docs\u002Fprogress.md` already exist.\n4. Otherwise read `docs\u002Fspec.md`, `docs\u002Fprogress.md`, `docs\u002Fnext.md`, and `docs\u002Fdesign.md` when relevant.\n5. Read `executionMode`, `teamRoles`, `currentOwner`, `qualityGates`, and `lastReviewerVerdict` from runtime state.\n6. Confirm whether there is an active `human-required` blocker.\n7. If not blocked, continue with the top unfinished task.\n8. Save results back to the same files after each loop.\n\n## Manager Runtime Contract\n\nThe manager owns the real orchestration loop.\nDo not treat `executionMode`, `teamRoles`, or `qualityGates` as decorative metadata.\nUse them to decide which specialist runs next and when a slice can advance.\n\nBefore the first specialist step:\n\n1. Read the saved `executionMode`.\n2. Check whether Codex native sub-agents are actually available in the current runtime.\n3. If the saved mode is `team-product` or `team-lite` but native sub-agents are unavailable, downgrade to `serial-fallback`.\n4. Immediately update:\n   - `autopilot\u002Fstate.json`\n   - `docs\u002Fprogress.md`\n   - `docs\u002Fnext.md`\n   so the recorded execution mode matches the real runtime path.\n\n## Intake Rules\n\nAsk for only the minimum complete project contract:\n\n- product summary\n- target user\n- core features\n- non-goals for this version\n- stack preference or default stack permission\n- architecture preset\n- auth requirement\n- payment requirement\n- admin requirement\n- deploy target\n- data store\n- theme preset\n- visual vibe\n- design direction refinement\n- definition of done\n\nKeep the intake compact, but ask exactly one question at a time.\nTreat one-question-at-a-time intake as the default behavior, not an optional preference.\nFor the exact interaction format, follow `..\u002Fautopilot-intake\u002FSKILL.md`.\n\n## Execution Rules\n\n1. The manager owns user-facing communication and final state updates.\n2. Do not send work directly to Builder before Planner has defined the current slice.\n3. Always choose the smallest shippable slice.\n4. Prefer working software over broad scaffolding with no validation.\n5. Run relevant verification after each meaningful change.\n6. Update progress and next-step files after each loop.\n7. Continue automatically after intake unless the run is waiting on the upfront integration setup phase or a true `human-required` blocker is active.\n8. Ask the user only when:\n   - the upfront integration setup phase needs a consolidated env payload\n   - external account setup is required and no existing key can be provided yet\n   - payment or production deployment approval is required\n   - the product direction is ambiguous in a way that changes the build materially\n\nTreat `setup-secrets` pending as a normal pre-execution phase, not as a blocker entry by itself.\n\n## Team Orchestration\n\nDefault orchestration is conditional:\n\n- `team-product`\n  - user-facing product\n  - or `Scalable` architecture\n  - or auth \u002F payments \u002F admin are required\n- `team-lite`\n  - CLI-only, backend-only, library-only, or simple internal tooling\n- `serial-fallback`\n  - use the same role order in one loop when specialist sub-agents are unavailable\n\n### Role Activation\n\n- `Planner`: always active at project start and again at milestone transitions\n- `Architect`: active only for `Scalable` architecture or cross-cutting structure decisions\n- `Builder`: always active\n- `Designer`: active only for user-facing projects\n- `QA`: always active before a slice can be marked complete\n\n### Specialist Responsibilities\n\n- `Planner`\n  - break the spec into milestones and current-slice acceptance criteria\n  - draft `docs\u002Fnext.md` and the initial quality gates\n- `Architect`\n  - define shared vs feature boundaries and integration seams\n  - prevent structure drift, especially under `Scalable`\n- `Builder`\n  - implement the next smallest shippable slice\n  - stay inside the planner\u002Farchitect contract\n- `Designer`\n  - strengthen `docs\u002Fdesign.md` into the binding UI brief\n  - run one explicit visual review after the first UI pass\n- `QA`\n  - validate tests, builds, key flows, regressions, and definition-of-done alignment\n  - leave a pass\u002Ffail verdict before completion\n\n### Codex Native Sub-Agent Policy\n\nUse Codex native sub-agents when they are available:\n\n- `Planner`, `Architect`, `Designer`, `QA`: `default`\n- `Builder`: `worker`\n- repo analysis or structural inspection: `explorer`\n\nKeep the manager in control of ordering:\n\n1. planner first\n2. architect and designer only when needed\n3. builder implementation\n4. QA review\n\nDesigner review is mandatory for user-facing UI slices.\nQA pass is mandatory before a slice is treated as complete.\n\n## Actual Execution Sequence\n\nUse this exact sequence on every slice.\n\n### Step 1: Planner\n\n- Always run planner first.\n- Use `..\u002Fautopilot-planner\u002FSKILL.md`.\n- Expected planner output:\n  - planner checkpoint\n  - current slice\n  - acceptance criteria\n  - risks\n  - required follow-on roles\n  - QA checks\n- Write planner results into:\n  - `docs\u002Fprogress.md`\n  - `docs\u002Fnext.md`\n  - `autopilot\u002Fstate.json`\n- Record the handoff with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role planner \\\n  --summary \"\u003Cplanner summary>\" \\\n  --planner-checkpoint \"\u003Cplanner checkpoint>\" \\\n  --planner-intent \"\u003Cplanner intent>\" \\\n  --builder-target \"\u003Cbuilder target>\" \\\n  --qa-check \"\u003Cqa check 1>\" \\\n  --qa-check \"\u003Cqa check 2>\" \\\n  --next-owner \"\u003CArchitect or Builder>\" \\\n  --next-task \"\u003Cnext current task>\"\n```\n\n### Step 2: Architect if needed\n\nRun architect only when:\n\n- `Scalable` architecture is active, or\n- planner identifies cross-cutting structure risk\n\nUse `..\u002Fautopilot-architect\u002FSKILL.md`.\nArchitect refines boundaries before builder starts and updates the same state files.\nRecord the result with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role architect \\\n  --summary \"\u003Carchitect summary>\" \\\n  --architect-guardrails \"\u003Carchitect guardrails>\" \\\n  --next-owner \"Builder\" \\\n  --next-task \"\u003Cbuilder task>\"\n```\n\n### Step 3: Builder\n\n- Builder only runs after planner, and after architect when architect was required.\n- If native sub-agents are available, delegate builder work to a `worker` sub-agent.\n- If not, execute the builder role directly inside the manager loop.\n- Builder must stay inside the planner and architect contract.\n- After a meaningful builder step, record progress with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role builder \\\n  --summary \"\u003Cbuilder summary>\" \\\n  --builder-progress \"\u003Cbuilder progress>\" \\\n  --next-owner \"\u003CDesigner or QA>\" \\\n  --next-task \"\u003Cnext current task>\"\n```\n\n### Step 4: Designer if needed\n\nRun designer only for user-facing slices.\n\n- Use `..\u002Fautopilot-designer\u002FSKILL.md`.\n- Before the first UI slice, designer strengthens `docs\u002Fdesign.md`.\n- After the first UI pass, designer performs one explicit visual review.\n- Manager must record the design verdict into:\n  - `docs\u002Fdesign.md`\n  - `docs\u002Fprogress.md`\n  - `autopilot\u002Fstate.json`\n- Record the result with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role designer \\\n  --summary \"\u003Cdesigner summary>\" \\\n  --designer-notes \"\u003Cdesigner notes>\" \\\n  --reviewer-verdict \"\u003Cdesigner verdict>\" \\\n  --design-review-note \"\u003Cdesign review note>\" \\\n  --next-owner \"QA\" \\\n  --next-task \"\u003Cqa task>\"\n```\n\n### Step 5: QA\n\n- Always run QA before a slice is marked complete.\n- Use `..\u002Fautopilot-qa\u002FSKILL.md`.\n- QA must return pass or fail.\n- If QA fails:\n  - do not mark the slice complete\n  - route rework back to builder\n  - update `lastReviewerVerdict`\n- Record the verdict with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role qa \\\n  --summary \"\u003Cqa summary>\" \\\n  --reviewer-verdict \"\u003Cpass or fail verdict>\" \\\n  --qa-check \"\u003Cqa check 1>\" \\\n  --qa-check \"\u003Cqa check 2>\" \\\n  --next-owner \"\u003CManager or Builder>\" \\\n  --next-task \"\u003Cnext current task>\"\n```\n\n### Step 6: Manager update\n\nAfter the final specialist for the slice:\n\n- update `currentOwner`\n- update `currentTask`\n- update `lastPlannerCheckpoint` when planner completes\n- update `lastReviewerVerdict` when designer or QA completes\n- update `docs\u002Fprogress.md`\n- update `docs\u002Fnext.md`\n- append or resolve blockers in `autopilot\u002Fblockers.json`\n\nUse the same checkpoint script to record blockers when needed:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role manager \\\n  --summary \"\u003Cmanager status update>\" \\\n  --blocker-classification \"human-required\" \\\n  --blocker-summary \"\u003Cminimal blocker summary>\" \\\n  --blocker-owner-role \"\u003Ccurrent owner role>\"\n```\n\n## Serial Fallback Rule\n\nIf native sub-agents are unavailable, run the exact same role order in one manager loop:\n\n1. planner\n2. architect if needed\n3. builder\n4. designer if needed\n5. QA\n\nDo not skip roles just because the runtime is serial.\n`serial-fallback` changes the backend, not the quality bar.\n\n## Design Execution Rule\n\nFor user-facing products, design synthesis is required before the first UI build.\n\n- Create and follow `docs\u002Fdesign.md`\n- Use the `frontend-ui-ux` skill as the default design helper\n- Prefer Figma references when provided\n- Otherwise use curated design inspiration sources rather than broad web search\n- Keep the base UI bright: white or light-neutral surfaces, compact readable typography, and restrained chrome\n- Choose the navigation shell from the project type: landing pages use top headers, dashboard\u002Fadmin surfaces use sidebars, and mobile-style products use app-header or bottom-tab shells\n- Treat theme and vibe as modulation of accent, contrast, spacing tone, and component shape rather than a full override of the base surface system\n- Run one explicit post-build design review after the first UI pass\n- Treat `docs\u002Fdesign.md` as the designer-owned binding brief\n\nTreat these as user-facing by default:\n\n- landing pages\n- marketing sites\n- dashboards\n- consumer web or mobile apps\n- admin interfaces\n\nSkip dedicated design synthesis only for clearly CLI-only, backend-only, or library-only work.\n\n## Blocker Policy\n\n### `retryable`\n\n- dependency install errors\n- failing tests\n- build issues\n- small runtime bugs\n\nBehavior:\n\n- retry automatically\n- try one or more alternative fixes\n- only escalate after retry budget is exhausted\n\n### `deferable`\n\n- placeholder copy\n- icon selection\n- optional polish\n- non-critical analytics\n\nBehavior:\n\n- keep moving with a reasonable default\n- write the deferred decision into `docs\u002Fnext.md` when useful\n\n### `human-required`\n\n- API keys or secrets\n- OAuth app creation\n- payment account setup\n- production deployment approval\n- ambiguous product direction that changes scope materially\n\nBehavior:\n\n- stop only at the minimum blocking decision\n- ask one focused question\n- continue immediately after the answer\n- write the blocker with `ownerRole` in `autopilot\u002Fblockers.json`\n\nEach blocker entry must include:\n\n- `ownerRole`\n- `classification`\n- `summary`\n","# Auto Pilot\n\nUse this skill as the umbrella workflow for long-running project execution.\nTreat it as the **manager** role for the whole system.\n\n## Goal\n\nTurn a short product request into a resumable execution loop that:\n\n- gathers the minimum required inputs once\n- locks a written spec and definition of done\n- keeps implementing the next highest-priority work\n- escalates only for true human-only blockers\n\n## Operating Model\n\n### If state does not exist yet\n\n1. Route to the intake workflow first.\n2. Ask only for high-impact decisions.\n3. Infer safe defaults for the rest.\n4. After intake, detect whether the selected auth or managed services require upfront integration env values.\n5. If env values are already present, write:\n   - `docs\u002Fspec.md`\n   - `docs\u002Fdesign.md` for user-facing projects\n   - `docs\u002Fprogress.md`\n   - `docs\u002Fnext.md`\n   - `autopilot\u002Fstate.json`\n   - `autopilot\u002Fblockers.json`\n   - `autopilot\u002Fsecrets-status.json`\n6. If env values are still missing, treat the project as being in the normal `setup-secrets` phase and write:\n   - `docs\u002Fnext.md`\n   - `autopilot\u002Fstate.json`\n   - `autopilot\u002Fblockers.json`\n   - `autopilot\u002Fsecrets-status.json`\n   - request the consolidated env payload\n7. Start implementation immediately only after the required env values are ready.\n\n### If state already exists\n\n1. Read `autopilot\u002Fstate.json`, `autopilot\u002Fblockers.json`, and `autopilot\u002Fsecrets-status.json` if it exists.\n2. If `setupStatus` is `pending` or `autopilot\u002Fsecrets-status.json` is still pending, treat the project as an existing project waiting on setup continuation.\n3. In that setup continuation path, collect only the missing env payload and do not assume `docs\u002Fspec.md` or `docs\u002Fprogress.md` already exist.\n4. Otherwise read `docs\u002Fspec.md`, `docs\u002Fprogress.md`, `docs\u002Fnext.md`, and `docs\u002Fdesign.md` when relevant.\n5. Read `executionMode`, `teamRoles`, `currentOwner`, `qualityGates`, and `lastReviewerVerdict` from runtime state.\n6. Confirm whether there is an active `human-required` blocker.\n7. If not blocked, continue with the top unfinished task.\n8. Save results back to the same files after each loop.\n\n## Manager Runtime Contract\n\nThe manager owns the real orchestration loop.\nDo not treat `executionMode`, `teamRoles`, or `qualityGates` as decorative metadata.\nUse them to decide which specialist runs next and when a slice can advance.\n\nBefore the first specialist step:\n\n1. Read the saved `executionMode`.\n2. Check whether Codex native sub-agents are actually available in the current runtime.\n3. If the saved mode is `team-product` or `team-lite` but native sub-agents are unavailable, downgrade to `serial-fallback`.\n4. Immediately update:\n   - `autopilot\u002Fstate.json`\n   - `docs\u002Fprogress.md`\n   - `docs\u002Fnext.md`\n   so the recorded execution mode matches the real runtime path.\n\n## Intake Rules\n\nAsk for only the minimum complete project contract:\n\n- product summary\n- target user\n- core features\n- non-goals for this version\n- stack preference or default stack permission\n- architecture preset\n- auth requirement\n- payment requirement\n- admin requirement\n- deploy target\n- data store\n- theme preset\n- visual vibe\n- design direction refinement\n- definition of done\n\nKeep the intake compact, but ask exactly one question at a time.\nTreat one-question-at-a-time intake as the default behavior, not an optional preference.\nFor the exact interaction format, follow `..\u002Fautopilot-intake\u002FSKILL.md`.\n\n## Execution Rules\n\n1. The manager owns user-facing communication and final state updates.\n2. Do not send work directly to Builder before Planner has defined the current slice.\n3. Always choose the smallest shippable slice.\n4. Prefer working software over broad scaffolding with no validation.\n5. Run relevant verification after each meaningful change.\n6. Update progress and next-step files after each loop.\n7. Continue automatically after intake unless the run is waiting on the upfront integration setup phase or a true `human-required` blocker is active.\n8. Ask the user only when:\n   - the upfront integration setup phase needs a consolidated env payload\n   - external account setup is required and no existing key can be provided yet\n   - payment or production deployment approval is required\n   - the product direction is ambiguous in a way that changes the build materially\n\nTreat `setup-secrets` pending as a normal pre-execution phase, not as a blocker entry by itself.\n\n## Team Orchestration\n\nDefault orchestration is conditional:\n\n- `team-product`\n  - user-facing product\n  - or `Scalable` architecture\n  - or auth \u002F payments \u002F admin are required\n- `team-lite`\n  - CLI-only, backend-only, library-only, or simple internal tooling\n- `serial-fallback`\n  - use the same role order in one loop when specialist sub-agents are unavailable\n\n### Role Activation\n\n- `Planner`: always active at project start and again at milestone transitions\n- `Architect`: active only for `Scalable` architecture or cross-cutting structure decisions\n- `Builder`: always active\n- `Designer`: active only for user-facing projects\n- `QA`: always active before a slice can be marked complete\n\n### Specialist Responsibilities\n\n- `Planner`\n  - break the spec into milestones and current-slice acceptance criteria\n  - draft `docs\u002Fnext.md` and the initial quality gates\n- `Architect`\n  - define shared vs feature boundaries and integration seams\n  - prevent structure drift, especially under `Scalable`\n- `Builder`\n  - implement the next smallest shippable slice\n  - stay inside the planner\u002Farchitect contract\n- `Designer`\n  - strengthen `docs\u002Fdesign.md` into the binding UI brief\n  - run one explicit visual review after the first UI pass\n- `QA`\n  - validate tests, builds, key flows, regressions, and definition-of-done alignment\n  - leave a pass\u002Ffail verdict before completion\n\n### Codex Native Sub-Agent Policy\n\nUse Codex native sub-agents when they are available:\n\n- `Planner`, `Architect`, `Designer`, `QA`: `default`\n- `Builder`: `worker`\n- repo analysis or structural inspection: `explorer`\n\nKeep the manager in control of ordering:\n\n1. planner first\n2. architect and designer only when needed\n3. builder implementation\n4. QA review\n\nDesigner review is mandatory for user-facing UI slices.\nQA pass is mandatory before a slice is treated as complete.\n\n## Actual Execution Sequence\n\nUse this exact sequence on every slice.\n\n### Step 1: Planner\n\n- Always run planner first.\n- Use `..\u002Fautopilot-planner\u002FSKILL.md`.\n- Expected planner output:\n  - planner checkpoint\n  - current slice\n  - acceptance criteria\n  - risks\n  - required follow-on roles\n  - QA checks\n- Write planner results into:\n  - `docs\u002Fprogress.md`\n  - `docs\u002Fnext.md`\n  - `autopilot\u002Fstate.json`\n- Record the handoff with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role planner \\\n  --summary \"\u003Cplanner summary>\" \\\n  --planner-checkpoint \"\u003Cplanner checkpoint>\" \\\n  --planner-intent \"\u003Cplanner intent>\" \\\n  --builder-target \"\u003Cbuilder target>\" \\\n  --qa-check \"\u003Cqa check 1>\" \\\n  --qa-check \"\u003Cqa check 2>\" \\\n  --next-owner \"\u003CArchitect or Builder>\" \\\n  --next-task \"\u003Cnext current task>\"\n```\n\n### Step 2: Architect if needed\n\nRun architect only when:\n\n- `Scalable` architecture is active, or\n- planner identifies cross-cutting structure risk\n\nUse `..\u002Fautopilot-architect\u002FSKILL.md`.\nArchitect refines boundaries before builder starts and updates the same state files.\nRecord the result with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role architect \\\n  --summary \"\u003Carchitect summary>\" \\\n  --architect-guardrails \"\u003Carchitect guardrails>\" \\\n  --next-owner \"Builder\" \\\n  --next-task \"\u003Cbuilder task>\"\n```\n\n### Step 3: Builder\n\n- Builder only runs after planner, and after architect when architect was required.\n- If native sub-agents are available, delegate builder work to a `worker` sub-agent.\n- If not, execute the builder role directly inside the manager loop.\n- Builder must stay inside the planner and architect contract.\n- After a meaningful builder step, record progress with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role builder \\\n  --summary \"\u003Cbuilder summary>\" \\\n  --builder-progress \"\u003Cbuilder progress>\" \\\n  --next-owner \"\u003CDesigner or QA>\" \\\n  --next-task \"\u003Cnext current task>\"\n```\n\n### Step 4: Designer if needed\n\nRun designer only for user-facing slices.\n\n- Use `..\u002Fautopilot-designer\u002FSKILL.md`.\n- Before the first UI slice, designer strengthens `docs\u002Fdesign.md`.\n- After the first UI pass, designer performs one explicit visual review.\n- Manager must record the design verdict into:\n  - `docs\u002Fdesign.md`\n  - `docs\u002Fprogress.md`\n  - `autopilot\u002Fstate.json`\n- Record the result with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role designer \\\n  --summary \"\u003Cdesigner summary>\" \\\n  --designer-notes \"\u003Cdesigner notes>\" \\\n  --reviewer-verdict \"\u003Cdesigner verdict>\" \\\n  --design-review-note \"\u003Cdesign review note>\" \\\n  --next-owner \"QA\" \\\n  --next-task \"\u003Cqa task>\"\n```\n\n### Step 5: QA\n\n- Always run QA before a slice is marked complete.\n- Use `..\u002Fautopilot-qa\u002FSKILL.md`.\n- QA must return pass or fail.\n- If QA fails:\n  - do not mark the slice complete\n  - route rework back to builder\n  - update `lastReviewerVerdict`\n- Record the verdict with:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role qa \\\n  --summary \"\u003Cqa summary>\" \\\n  --reviewer-verdict \"\u003Cpass or fail verdict>\" \\\n  --qa-check \"\u003Cqa check 1>\" \\\n  --qa-check \"\u003Cqa check 2>\" \\\n  --next-owner \"\u003CManager or Builder>\" \\\n  --next-task \"\u003Cnext current task>\"\n```\n\n### Step 6: Manager update\n\nAfter the final specialist for the slice:\n\n- update `currentOwner`\n- update `currentTask`\n- update `lastPlannerCheckpoint` when planner completes\n- update `lastReviewerVerdict` when designer or QA completes\n- update `docs\u002Fprogress.md`\n- update `docs\u002Fnext.md`\n- append or resolve blockers in `autopilot\u002Fblockers.json`\n\nUse the same checkpoint script to record blockers when needed:\n\n```bash\npython3 ..\u002F..\u002Fscripts\u002Fteam_checkpoint.py \\\n  --workspace \u003Ctarget-workspace> \\\n  --role manager \\\n  --summary \"\u003Cmanager status update>\" \\\n  --blocker-classification \"human-required\" \\\n  --blocker-summary \"\u003Cminimal blocker summary>\" \\\n  --blocker-owner-role \"\u003Ccurrent owner role>\"\n```\n\n## Serial Fallback Rule\n\nIf native sub-agents are unavailable, run the exact same role order in one manager loop:\n\n1. planner\n2. architect if needed\n3. builder\n4. designer if needed\n5. QA\n\nDo not skip roles just because the runtime is serial.\n`serial-fallback` changes the backend, not the quality bar.\n\n## Design Execution Rule\n\nFor user-facing products, design synthesis is required before the first UI build.\n\n- Create and follow `docs\u002Fdesign.md`\n- Use the `frontend-ui-ux` skill as the default design helper\n- Prefer Figma references when provided\n- Otherwise use curated design inspiration sources rather than broad web search\n- Keep the base UI bright: white or light-neutral surfaces, compact readable typography, and restrained chrome\n- Choose the navigation shell from the project type: landing pages use top headers, dashboard\u002Fadmin surfaces use sidebars, and mobile-style products use app-header or bottom-tab shells\n- Treat theme and vibe as modulation of accent, contrast, spacing tone, and component shape rather than a full override of the base surface system\n- Run one explicit post-build design review after the first UI pass\n- Treat `docs\u002Fdesign.md` as the designer-owned binding brief\n\nTreat these as user-facing by default:\n\n- landing pages\n- marketing sites\n- dashboards\n- consumer web or mobile apps\n- admin interfaces\n\nSkip dedicated design synthesis only for clearly CLI-only, backend-only, or library-only work.\n\n## Blocker Policy\n\n### `retryable`\n\n- dependency install errors\n- failing tests\n- build issues\n- small runtime bugs\n\nBehavior:\n\n- retry automatically\n- try one or more alternative fixes\n- only escalate after retry budget is exhausted\n\n### `deferable`\n\n- placeholder copy\n- icon selection\n- optional polish\n- non-critical analytics\n\nBehavior:\n\n- keep moving with a reasonable default\n- write the deferred decision into `docs\u002Fnext.md` when useful\n\n### `human-required`\n\n- API keys or secrets\n- OAuth app creation\n- payment account setup\n- production deployment approval\n- ambiguous product direction that changes scope materially\n\nBehavior:\n\n- stop only at the minimum blocking decision\n- ask one focused question\n- continue immediately after the answer\n- write the blocker with `ownerRole` in `autopilot\u002Fblockers.json`\n\nEach blocker entry must include:\n\n- `ownerRole`\n- `classification`\n- `summary`\n",{"name":5,"summary":50,"metadata":51,"retrieval":76,"description":7},"Ask once, lock spec, save state, keep shipping.",{"priority":52,"promptSignals":53},8,{"allOf":54,"anyOf":68,"phrases":71,"minScore":75},[55,58,59,60,61,62,63,64,65,67],[56,57],"ap","build",[57,56],[56,32],[32,56],[56,44],[44,56],[56,34],[34,56],[56,66],"만들어줘",[66,56],[69,70,5],"오토파일럿","autopilot",[69,70,5,72,73,74],"24시간 계속","with ap","using ap",7,{"aliases":77,"intents":82,"entities":86},[69,78,79,73,80,81],"autopilot mode","auto-pilot mode","intake first execution","continuous project execution",[83,84,85],"start an autopilot project","run intake and keep building","continue a project with minimal interruptions",[87,33,88,89],"spec","blockers","state","main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002Fminsu0707\u002Fauto-pilot\u002Fskills\u002Fauto-pilot",491,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-14 14:06:13",{"source":8,"skillName":5,"commands":101,"prompt":111,"zipAvailable":112},[102,105,108],{"tool":103,"command":104},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill auto-pilot",{"tool":106,"command":107},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill auto-pilot",{"tool":109,"command":110},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill auto-pilot","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fminsu0707\u002Fauto-pilot\u002Fskills\u002Fauto-pilot\u002FSKILL.md",true,[],[115,125,133,141,149,157],{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":119,"licenseFileUrl":120,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":124},"01M7DV332YQMGD8WTRJ9DZQXTB","worktable-review-with-annotations","Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.","MIT","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":119,"licenseFileUrl":129,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":124},"01M7DV2YGHCZVBEB1AHYB6CGHE","worktable-find-and-synthesize-context","Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FSKILL.md",[],[],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":119,"licenseFileUrl":137,"sourceHtmlUrl":138,"categories":139,"keywords":140,"publishedAt":124},"01M7DV2TQN1H11ME2SAF9MXN7V","worktable-create-or-update-docs","Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FSKILL.md",[],[],{"skillNo":142,"name":143,"nameZh":6,"description":144,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":119,"licenseFileUrl":145,"sourceHtmlUrl":146,"categories":147,"keywords":148,"publishedAt":124},"01M7DV2NE452NEA8TXATKSNAJH","worktable-create-or-manage-records","Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FSKILL.md",[],[],{"skillNo":150,"name":151,"nameZh":6,"description":152,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":119,"licenseFileUrl":153,"sourceHtmlUrl":154,"categories":155,"keywords":156,"publishedAt":124},"01M7DV2HENRBR2KSCVM064WH0Y","worktable-create-or-manage-interactive-html","Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FSKILL.md",[],[],{"skillNo":158,"name":159,"nameZh":6,"description":160,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":119,"licenseFileUrl":161,"sourceHtmlUrl":162,"categories":163,"keywords":164,"publishedAt":124},"01M7DV2CADR5R7RQFFYPNAGR9Y","worktable-collaborate-threads","Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FSKILL.md",[],[]]