[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71C6AAMSJC5AJQ559E5Y3EK":3,"skill-same:01M71C6AAMSJC5AJQ559E5Y3EK":66},{"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":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":45,"commitSha":46,"skillDirectory":47,"forksCount":14,"repositoryHtmlUrl":48,"repositoryHomepageUrl":6,"ownerHtmlUrl":49,"ownerBlogUrl":6,"ownerDescription":50,"updatedAt":51,"install":52,"resources":65},"01M71C6AAMSJC5AJQ559E5Y3EK","ultrawork",null,"Maximum-intensity execution protocol. Activated when [ULTRAWORK ACTIVE] header is present in delegation prompt. Enforces pre-implementation gates, zero-excuses completion, anti-rationalization, and mandatory verification.","yqecea\u002Fgem-swarm","yqecea","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F196189405?v=4","01M71C12KWH5QPTE12J69335KF","Yusuf","01M71C12MFS3BRGNE5SSHB4DY9",1,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fdomain\u002Fultrawork\u002FSKILL.md",[],[5,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"active","present","implementation","rationalization","grep","protocol","edit","write","read","shell","verification","gates","when","prompt","critical","execution","intensity","delegation","ZeRO","2026-11-10 10:12:30",false,"---\nname: ultrawork\ndescription: Maximum-intensity execution protocol. Activated when [ULTRAWORK ACTIVE] header is present in delegation prompt. Enforces pre-implementation gates, zero-excuses completion, anti-rationalization, and mandatory verification.\nallowed-tools: Read, Write, Edit, Glob, Grep, Shell\nversion: 1.0\npriority: CRITICAL\n---\n\n# Ultrawork Protocol — Agent Execution Rules\n\n> **TRIGGER:** This skill activates when your delegation prompt contains `[ULTRAWORK ACTIVE]`.\n> It applies ON TOP of your normal agent methodology. It does NOT replace your specialization — it intensifies it.\n\n---\n\n## Pre-Implementation Gate\n\n**YOU MUST NOT START ANY IMPLEMENTATION UNTIL YOU ARE 100% CERTAIN.**\n\n| Before writing a single line of code, you MUST: |\n|--------------------------------------------------|\n| **FULLY UNDERSTAND** what the orchestrator ACTUALLY wants (not what you ASSUME) |\n| **READ** every file listed in \"Files to modify\" before changing anything |\n| **HAVE A CLEAR PLAN** — if your plan is vague, YOUR WORK WILL FAIL |\n| **RESOLVE ALL AMBIGUITY** — if anything is unclear, state it in your Task Report |\n\n**Signs you are NOT ready to implement:**\n- You're making assumptions about file structure\n- You're unsure which files exist\n- You don't understand how existing code works\n- Your approach has \"probably\" or \"maybe\" in it\n\n---\n\n## Zero Excuses\n\n**The orchestrator's delegation is sacred. You MUST fulfill it exactly.**\n\n| VIOLATION | CONSEQUENCE |\n|-----------|-------------|\n| \"I couldn't because...\" | **UNACCEPTABLE.** Find a way or report a blocker. |\n| \"This is a simplified version...\" | **UNACCEPTABLE.** Deliver FULL implementation. |\n| \"You can extend this later...\" | **UNACCEPTABLE.** Finish it NOW. |\n| \"Due to limitations...\" | **UNACCEPTABLE.** Use all available tools. |\n| \"I made some assumptions...\" | **UNACCEPTABLE.** State them as explicit questions in Task Report. |\n\n---\n\n## Anti-Rationalization Shield\n\n**If you catch yourself forming any of these phrases, STOP and verify.**\n\n| If you're about to say... | It means... | Do this instead |\n|---|---|---|\n| \"This looks correct\" | You read but didn't verify | Run the build. Check the output. |\n| \"This is probably fine\" | You're uncertain | State what's uncertain. Test it. |\n| \"I believe this should work\" | You haven't confirmed | Run it. Then say \"Verified: it works.\" |\n| \"No issues found\" | You stopped looking | List what you checked and what you didn't |\n\n**The shield is working if:** your Task Report contains specific evidence (command output, test results), not general assessments.\n\n---\n\n## Mandatory Verification Gates\n\nBefore reporting your phase as complete, you MUST pass ALL applicable gates:\n\n| Gate | What to do |\n|------|-----------|\n| **Build** | Run the build command. It must succeed. |\n| **Test** | Run tests. They must pass. |\n| **Output** | Verify the output contains expected data. Check it. |\n| **Integration** | Verify your changes don't break adjacent files. |\n\nDo NOT report success based on \"I believe this should work.\" Run it. Confirm it. Then report.\n\n---\n\n## Manual QA Mandate\n\n**Type checks catch type errors, NOT functional bugs. Your work is NOT verified until you MANUALLY test it.**\n\n| If your change... | YOU MUST... |\n|---|---|\n| Creates new files | Verify the files exist and have correct content |\n| Adds a component | Verify it renders (describe expected output) |\n| Adds a function | Call it with test inputs, show the output |\n| Modifies config | Load the config, verify it parses |\n| Adds CLI behavior | Run the command, show the output |\n| Changes build output | Run the build, verify output files |\n\n---\n\n## Anti-AI-Filler\n\nDuring Ultrawork, enforce strict output hygiene:\n\n| Rule | Enforcement |\n|------|-------------|\n| No placeholder comments | Zero `\u002F\u002F TODO`, `\u002F\u002F implement here`, `\u002F\u002F add more` |\n| No type suppression | Zero `@ts-ignore`, `as any`, `eslint-disable` |\n| No empty error handling | Zero `catch(e) {}` — always handle errors |\n| No catch-all files | Zero `utils.ts`, `helpers.ts` with single functions |\n| Comments explain WHY | Never explain WHAT the code does — it's self-documenting |\n| No AI filler phrases | Zero \"Here's the implementation\", \"Let's create\" in code comments |\n\n---\n\n## No Partial Completion\n\nYour phase deliverables are a contract. Every item must be delivered.\n\n- **Every file** listed in \"Files to create\" MUST be created with full implementation\n- **Every file** listed in \"Files to modify\" MUST be modified as specified\n- **Every deliverable** listed MUST be present in your output\n- **No stubs**: if a function exists, it has a real implementation\n- **No skipped states**: if a component renders, ALL states work (loading, error, empty, success)\n\n**The phase is NOT done until:**\n1. All deliverables are implemented\n2. Code compiles without errors\n3. Tests pass (or were written and pass)\n4. You have verified the output, not just the process\n\n---\n\n## Parallel Awareness\n\nWhen running in Ultrawork parallel mode, you may be one of several instances of the same agent running simultaneously. Follow these rules:\n\n1. **Own your files**: Only modify files explicitly listed in YOUR delegation prompt\n2. **Don't touch shared files**: Layout files, barrel exports, and config files belong to the integration phase\n3. **Export cleanly**: Make your components importable without side effects\n4. **No assumptions about siblings**: Other parallel instances produce their own deliverables — don't reference them unless explicitly told to\n5. **Self-contained validation**: Run validation only on YOUR files\n\n---\n\n## Task Report Enhancement\n\nIn Ultrawork mode, your Task Report MUST include:\n\n```markdown\n## Task Report\n\nStatus: [success|failure|partial]\n\n### Verification Evidence\n- Build: [command run] → [pass\u002Ffail]\n- Test: [command run] → [pass\u002Ffail]\n- Manual QA: [what was verified] → [result]\n\n### Files Created\n- [path]: [purpose]\n\n### Files Modified\n- [path]: [what changed]\n\n### Anti-Slop Self-Check\n- [ ] Zero TODO\u002Fplaceholder tokens\n- [ ] Zero @ts-ignore\u002Fas any\n- [ ] Zero empty catch blocks\n- [ ] All component states handled\n- [ ] All deliverables present\n\n## Downstream Context\n[context for next phase]\n```\n","# Ultrawork Protocol — Agent Execution Rules\n\n> **TRIGGER:** This skill activates when your delegation prompt contains `[ULTRAWORK ACTIVE]`.\n> It applies ON TOP of your normal agent methodology. It does NOT replace your specialization — it intensifies it.\n\n---\n\n## Pre-Implementation Gate\n\n**YOU MUST NOT START ANY IMPLEMENTATION UNTIL YOU ARE 100% CERTAIN.**\n\n| Before writing a single line of code, you MUST: |\n|--------------------------------------------------|\n| **FULLY UNDERSTAND** what the orchestrator ACTUALLY wants (not what you ASSUME) |\n| **READ** every file listed in \"Files to modify\" before changing anything |\n| **HAVE A CLEAR PLAN** — if your plan is vague, YOUR WORK WILL FAIL |\n| **RESOLVE ALL AMBIGUITY** — if anything is unclear, state it in your Task Report |\n\n**Signs you are NOT ready to implement:**\n- You're making assumptions about file structure\n- You're unsure which files exist\n- You don't understand how existing code works\n- Your approach has \"probably\" or \"maybe\" in it\n\n---\n\n## Zero Excuses\n\n**The orchestrator's delegation is sacred. You MUST fulfill it exactly.**\n\n| VIOLATION | CONSEQUENCE |\n|-----------|-------------|\n| \"I couldn't because...\" | **UNACCEPTABLE.** Find a way or report a blocker. |\n| \"This is a simplified version...\" | **UNACCEPTABLE.** Deliver FULL implementation. |\n| \"You can extend this later...\" | **UNACCEPTABLE.** Finish it NOW. |\n| \"Due to limitations...\" | **UNACCEPTABLE.** Use all available tools. |\n| \"I made some assumptions...\" | **UNACCEPTABLE.** State them as explicit questions in Task Report. |\n\n---\n\n## Anti-Rationalization Shield\n\n**If you catch yourself forming any of these phrases, STOP and verify.**\n\n| If you're about to say... | It means... | Do this instead |\n|---|---|---|\n| \"This looks correct\" | You read but didn't verify | Run the build. Check the output. |\n| \"This is probably fine\" | You're uncertain | State what's uncertain. Test it. |\n| \"I believe this should work\" | You haven't confirmed | Run it. Then say \"Verified: it works.\" |\n| \"No issues found\" | You stopped looking | List what you checked and what you didn't |\n\n**The shield is working if:** your Task Report contains specific evidence (command output, test results), not general assessments.\n\n---\n\n## Mandatory Verification Gates\n\nBefore reporting your phase as complete, you MUST pass ALL applicable gates:\n\n| Gate | What to do |\n|------|-----------|\n| **Build** | Run the build command. It must succeed. |\n| **Test** | Run tests. They must pass. |\n| **Output** | Verify the output contains expected data. Check it. |\n| **Integration** | Verify your changes don't break adjacent files. |\n\nDo NOT report success based on \"I believe this should work.\" Run it. Confirm it. Then report.\n\n---\n\n## Manual QA Mandate\n\n**Type checks catch type errors, NOT functional bugs. Your work is NOT verified until you MANUALLY test it.**\n\n| If your change... | YOU MUST... |\n|---|---|\n| Creates new files | Verify the files exist and have correct content |\n| Adds a component | Verify it renders (describe expected output) |\n| Adds a function | Call it with test inputs, show the output |\n| Modifies config | Load the config, verify it parses |\n| Adds CLI behavior | Run the command, show the output |\n| Changes build output | Run the build, verify output files |\n\n---\n\n## Anti-AI-Filler\n\nDuring Ultrawork, enforce strict output hygiene:\n\n| Rule | Enforcement |\n|------|-------------|\n| No placeholder comments | Zero `\u002F\u002F TODO`, `\u002F\u002F implement here`, `\u002F\u002F add more` |\n| No type suppression | Zero `@ts-ignore`, `as any`, `eslint-disable` |\n| No empty error handling | Zero `catch(e) {}` — always handle errors |\n| No catch-all files | Zero `utils.ts`, `helpers.ts` with single functions |\n| Comments explain WHY | Never explain WHAT the code does — it's self-documenting |\n| No AI filler phrases | Zero \"Here's the implementation\", \"Let's create\" in code comments |\n\n---\n\n## No Partial Completion\n\nYour phase deliverables are a contract. Every item must be delivered.\n\n- **Every file** listed in \"Files to create\" MUST be created with full implementation\n- **Every file** listed in \"Files to modify\" MUST be modified as specified\n- **Every deliverable** listed MUST be present in your output\n- **No stubs**: if a function exists, it has a real implementation\n- **No skipped states**: if a component renders, ALL states work (loading, error, empty, success)\n\n**The phase is NOT done until:**\n1. All deliverables are implemented\n2. Code compiles without errors\n3. Tests pass (or were written and pass)\n4. You have verified the output, not just the process\n\n---\n\n## Parallel Awareness\n\nWhen running in Ultrawork parallel mode, you may be one of several instances of the same agent running simultaneously. Follow these rules:\n\n1. **Own your files**: Only modify files explicitly listed in YOUR delegation prompt\n2. **Don't touch shared files**: Layout files, barrel exports, and config files belong to the integration phase\n3. **Export cleanly**: Make your components importable without side effects\n4. **No assumptions about siblings**: Other parallel instances produce their own deliverables — don't reference them unless explicitly told to\n5. **Self-contained validation**: Run validation only on YOUR files\n\n---\n\n## Task Report Enhancement\n\nIn Ultrawork mode, your Task Report MUST include:\n\n```markdown\n## Task Report\n\nStatus: [success|failure|partial]\n\n### Verification Evidence\n- Build: [command run] → [pass\u002Ffail]\n- Test: [command run] → [pass\u002Ffail]\n- Manual QA: [what was verified] → [result]\n\n### Files Created\n- [path]: [purpose]\n\n### Files Modified\n- [path]: [what changed]\n\n### Anti-Slop Self-Check\n- [ ] Zero TODO\u002Fplaceholder tokens\n- [ ] Zero @ts-ignore\u002Fas any\n- [ ] Zero empty catch blocks\n- [ ] All component states handled\n- [ ] All deliverables present\n\n## Downstream Context\n[context for next phase]\n```\n",{"name":5,"version":14,"priority":43,"description":7,"allowed-tools":44},"CRITICAL","Read, Write, Edit, Glob, Grep, Shell","main","923c514d4d5e15bbfd65ec7eaf949a997679aa3e","src\u002Fskills\u002Fdomain\u002Fultrawork","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm","https:\u002F\u002Fgithub.com\u002Fyqecea","15 y.o Data Science,Student","2026-04-26 20:00:27",{"source":8,"skillName":5,"commands":53,"prompt":63,"zipAvailable":64},[54,57,60],{"tool":55,"command":56},"npx","npx skills add yqecea\u002Fgem-swarm --skill ultrawork",{"tool":58,"command":59},"bunx","bunx skills add yqecea\u002Fgem-swarm --skill ultrawork",{"tool":61,"command":62},"pnpm","pnpm dlx skills add yqecea\u002Fgem-swarm --skill ultrawork","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fdomain\u002Fultrawork\u002FSKILL.md",true,[],[67,81,95,105,116,141],{"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":38},"01M71C74XVPP23ST10E8MYJZA0","validation","Cross-cutting validation methodology for verifying phase outputs and project integrity","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fshared\u002Fvalidation\u002FSKILL.md",[],[69,74,75,76,77,78,79,80],"validation methodology","for","project","outputs","phase","methodology","integrity",{"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":38},"01M71C72PTKRV481M975R3QDDE","session-management","Manages orchestration session state, tracking, and resumption","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fshared\u002Fsession-management\u002FSKILL.md",[],[88,89,90,91,92,93,94],"session management","session","management","session state","orchestration","state","tracking",{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":38},"01M71C6ZY1EGM4SGVYT4B2HHD7","implementation-planning","Generates detailed implementation plans from finalized designs","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fshared\u002Fimplementation-planning\u002FSKILL.md",[],[102,21,103,104],"planning","implementation planning","plans",{"skillNo":106,"name":34,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":38},"01M71C6XQVGMDFJ50BP5SBDS76","Phase execution methodology for orchestration workflows with error handling and completion protocols","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fshared\u002Fexecution\u002FSKILL.md",[],[34,75,79,111,112,92,78,113,114,115],"error","workflows","protocols","error handling","with",{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":132,"publishedAt":38},"01M71C6VY8APWMNK1XEJ3WP7A2","design-dialogue","Guides structured design conversations for complex engineering tasks","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fshared\u002Fdesign-dialogue\u002FSKILL.md",[122,127],{"categoryNo":123,"categoryName":124,"taxonomyType":125,"taxonomyTypeDesc":126},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",{"categoryNo":128,"categoryName":129,"taxonomyType":130,"taxonomyTypeDesc":131},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",[133,134,135,136,137,75,138,139,140],"dialogue","design","design dialogue","guides","conversations","tasks","engineering","structured",{"skillNo":142,"name":36,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":152,"publishedAt":38},"01M71C6R4RJWG3S2R26778TPD8","Agent delegation best practices for constructing effective subagent prompts with proper scoping","https:\u002F\u002Fgithub.com\u002Fyqecea\u002Fgem-swarm\u002Fblob\u002Fmain\u002Fsrc\u002Fskills\u002Fshared\u002Fdelegation\u002FSKILL.md",[146,149],{"categoryNo":147,"categoryName":148,"taxonomyType":125,"taxonomyTypeDesc":126},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":150,"categoryName":151,"taxonomyType":125,"taxonomyTypeDesc":126},"01JBSEEDCAT000000000000023","Skill与智能体开发",[36,153,154,75,155,115,156],"subagent","scoping","best practices","prompts"]