up-cc 0.16.1 → 2.0.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (134) hide show
  1. package/README.md +87 -577
  2. package/package.json +5 -3
  3. package/up/CHANGELOG.md +110 -0
  4. package/up/agents/up-arquiteto.md +95 -39
  5. package/up/agents/up-auditor.md +218 -0
  6. package/up/agents/up-executor.md +94 -31
  7. package/up/agents/up-mapeador-codigo.md +63 -10
  8. package/up/agents/up-pesquisador.md +278 -0
  9. package/up/agents/up-revisor.md +249 -0
  10. package/up/agents/up-sintetizador.md +156 -179
  11. package/up/agents/up-tester.md +280 -0
  12. package/up/agents/up-verificador.md +95 -11
  13. package/up/bin/install.js +182 -19
  14. package/up/bin/lib/core.cjs +17 -43
  15. package/up/bin/lib/github.cjs +495 -0
  16. package/up/bin/lib/multica.cjs +424 -0
  17. package/up/bin/up-tools.cjs +167 -46
  18. package/up/commands/auditar.md +66 -0
  19. package/up/commands/build.md +54 -43
  20. package/up/commands/depurar.md +1 -1
  21. package/up/commands/plan.md +52 -38
  22. package/up/commands/rapido.md +15 -9
  23. package/up/commands/testar.md +81 -122
  24. package/up/commands/up.md +106 -0
  25. package/up/hooks/up-session-start.js +107 -0
  26. package/up/references/engineering-principles.md +1 -1
  27. package/up/references/governance-rules.md +5 -5
  28. package/up/references/production-requirements.md +1 -1
  29. package/up/references/severity-levels.md +2 -2
  30. package/up/references/tdd-evidence-types.md +81 -0
  31. package/up/skills/up-brainstorm/SKILL.md +39 -0
  32. package/up/skills/up-tdd/SKILL.md +39 -0
  33. package/up/skills/up-verificar-antes-de-concluir/SKILL.md +49 -0
  34. package/up/skills/usando-up/SKILL.md +26 -0
  35. package/up/templates/audit-plan.md +3 -3
  36. package/up/templates/audit-report.md +2 -2
  37. package/up/templates/design-tokens.md +2 -2
  38. package/up/workflows/auditar.md +255 -0
  39. package/up/workflows/build.md +600 -386
  40. package/up/workflows/dcrv.md +183 -99
  41. package/up/workflows/governance.md +112 -220
  42. package/up/workflows/plan.md +169 -399
  43. package/up/workflows/rapido.md +7 -1
  44. package/up/workflows/up.md +447 -0
  45. package/up/agents/up-analista-codigo.md +0 -446
  46. package/up/agents/up-api-tester.md +0 -405
  47. package/up/agents/up-architecture-supervisor.md +0 -126
  48. package/up/agents/up-audit-supervisor.md +0 -83
  49. package/up/agents/up-auditor-modernidade.md +0 -378
  50. package/up/agents/up-auditor-performance.md +0 -426
  51. package/up/agents/up-auditor-ux.md +0 -396
  52. package/up/agents/up-backend-specialist.md +0 -175
  53. package/up/agents/up-blind-validator.md +0 -259
  54. package/up/agents/up-chief-architect.md +0 -184
  55. package/up/agents/up-chief-engineer.md +0 -202
  56. package/up/agents/up-chief-operations.md +0 -123
  57. package/up/agents/up-chief-product.md +0 -103
  58. package/up/agents/up-chief-quality.md +0 -211
  59. package/up/agents/up-clone-crawler.md +0 -234
  60. package/up/agents/up-clone-design-extractor.md +0 -227
  61. package/up/agents/up-clone-feature-mapper.md +0 -225
  62. package/up/agents/up-clone-prd-writer.md +0 -169
  63. package/up/agents/up-clone-verifier.md +0 -227
  64. package/up/agents/up-code-reviewer.md +0 -229
  65. package/up/agents/up-consolidador-ideias.md +0 -493
  66. package/up/agents/up-database-specialist.md +0 -169
  67. package/up/agents/up-delivery-auditor.md +0 -247
  68. package/up/agents/up-devops-agent.md +0 -203
  69. package/up/agents/up-execution-supervisor.md +0 -315
  70. package/up/agents/up-exhaustive-tester.md +0 -348
  71. package/up/agents/up-frontend-specialist.md +0 -152
  72. package/up/agents/up-operations-supervisor.md +0 -94
  73. package/up/agents/up-pesquisador-mercado.md +0 -350
  74. package/up/agents/up-pesquisador-projeto.md +0 -358
  75. package/up/agents/up-planning-auditor.md +0 -284
  76. package/up/agents/up-planning-supervisor.md +0 -260
  77. package/up/agents/up-product-analyst.md +0 -192
  78. package/up/agents/up-product-supervisor.md +0 -83
  79. package/up/agents/up-project-ceo.md +0 -352
  80. package/up/agents/up-qa-agent.md +0 -171
  81. package/up/agents/up-quality-supervisor.md +0 -178
  82. package/up/agents/up-requirements-validator.md +0 -230
  83. package/up/agents/up-security-reviewer.md +0 -137
  84. package/up/agents/up-sintetizador-melhorias.md +0 -407
  85. package/up/agents/up-system-designer.md +0 -332
  86. package/up/agents/up-technical-writer.md +0 -188
  87. package/up/agents/up-verification-supervisor.md +0 -111
  88. package/up/agents/up-visual-critic.md +0 -358
  89. package/up/commands/adicionar-fase.md +0 -47
  90. package/up/commands/adicionar-testes.md +0 -145
  91. package/up/commands/ajuda.md +0 -176
  92. package/up/commands/atualizar.md +0 -103
  93. package/up/commands/clone-builder.md +0 -67
  94. package/up/commands/configurar.md +0 -219
  95. package/up/commands/custos.md +0 -67
  96. package/up/commands/dashboard.md +0 -48
  97. package/up/commands/discutir-fase.md +0 -35
  98. package/up/commands/executar-fase.md +0 -40
  99. package/up/commands/ideias.md +0 -49
  100. package/up/commands/iniciar.md +0 -31
  101. package/up/commands/mapear-codigo.md +0 -63
  102. package/up/commands/melhorias.md +0 -45
  103. package/up/commands/mobile-first.md +0 -71
  104. package/up/commands/modo-builder.md +0 -186
  105. package/up/commands/novo-projeto.md +0 -40
  106. package/up/commands/onboard.md +0 -69
  107. package/up/commands/pausar.md +0 -33
  108. package/up/commands/planejar-fase.md +0 -45
  109. package/up/commands/progresso.md +0 -33
  110. package/up/commands/remover-fase.md +0 -34
  111. package/up/commands/resetar.md +0 -27
  112. package/up/commands/retomar.md +0 -35
  113. package/up/commands/saude.md +0 -103
  114. package/up/commands/ux-tester.md +0 -63
  115. package/up/commands/verificar-trabalho.md +0 -35
  116. package/up/workflows/adicionar-fase.md +0 -112
  117. package/up/workflows/builder-e2e.md +0 -501
  118. package/up/workflows/builder.md +0 -3419
  119. package/up/workflows/ceo-intake.md +0 -305
  120. package/up/workflows/ceo-updates.md +0 -183
  121. package/up/workflows/clone-builder.md +0 -320
  122. package/up/workflows/discutir-fase.md +0 -336
  123. package/up/workflows/executar-fase.md +0 -358
  124. package/up/workflows/executar-plano.md +0 -659
  125. package/up/workflows/ideias.md +0 -381
  126. package/up/workflows/iniciar.md +0 -235
  127. package/up/workflows/melhorias.md +0 -409
  128. package/up/workflows/mobile-first.md +0 -692
  129. package/up/workflows/novo-projeto.md +0 -778
  130. package/up/workflows/planejar-fase.md +0 -293
  131. package/up/workflows/progresso.md +0 -226
  132. package/up/workflows/retomar.md +0 -231
  133. package/up/workflows/ux-tester.md +0 -526
  134. package/up/workflows/verificar-trabalho.md +0 -308
@@ -1,35 +0,0 @@
1
- ---
2
- name: up:discutir-fase
3
- description: Coletar contexto da fase por questionamento antes do planejamento
4
- argument-hint: "<fase>"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Bash
9
- - Glob
10
- - Grep
11
- - AskUserQuestion
12
- ---
13
- <objective>
14
- Collect phase context through structured questioning before planning begins.
15
-
16
- **Default flow:** Load phase info -> Ask targeted questions -> Capture decisions -> Write CONTEXT.md
17
-
18
- **Orchestrator role:** Read roadmap phase description, ask clarifying questions about scope/approach/constraints, capture all decisions and assumptions in CONTEXT.md for the planner to consume.
19
- </objective>
20
-
21
- <execution_context>
22
- @~/.claude/up/workflows/discutir-fase.md
23
- @~/.claude/up/references/ui-brand.md
24
- </execution_context>
25
-
26
- <context>
27
- Phase: $ARGUMENTS
28
-
29
- Discussion surfaces assumptions, clarifies gray areas, and captures decisions before planning.
30
- </context>
31
-
32
- <process>
33
- Execute the discutir-fase workflow from @~/.claude/up/workflows/discutir-fase.md end-to-end.
34
- Preserve all workflow gates (phase validation, questioning, decision capture, CONTEXT.md creation).
35
- </process>
@@ -1,40 +0,0 @@
1
- ---
2
- name: up:executar-fase
3
- description: Executar todos os planos de uma fase com paralelizacao por ondas
4
- argument-hint: "<fase> [--gaps-only]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Edit
9
- - Glob
10
- - Grep
11
- - Bash
12
- - Task
13
- - AskUserQuestion
14
- ---
15
- <objective>
16
- Execute all plans in a phase using wave-based parallel execution.
17
-
18
- Orchestrator stays lean: discover plans, analyze dependencies, group into waves, spawn subagents, collect results. Each subagent loads the full execute-plan context and handles its own plan.
19
-
20
- Context budget: ~15% orchestrator, 100% fresh per subagent.
21
- </objective>
22
-
23
- <execution_context>
24
- @~/.claude/up/workflows/executar-fase.md
25
- @~/.claude/up/references/ui-brand.md
26
- </execution_context>
27
-
28
- <context>
29
- Phase: $ARGUMENTS
30
-
31
- **Flags:**
32
- - `--gaps-only` -- Execute only gap closure plans (plans with `gap_closure: true` in frontmatter). Use after verificar-trabalho creates fix plans.
33
-
34
- Context files are resolved inside the workflow and per-subagent `<files_to_read>` blocks.
35
- </context>
36
-
37
- <process>
38
- Execute the executar-fase workflow from @~/.claude/up/workflows/executar-fase.md end-to-end.
39
- Preserve all workflow gates (wave execution, checkpoint handling, verification, state updates, routing).
40
- </process>
@@ -1,49 +0,0 @@
1
- ---
2
- name: up:ideias
3
- description: Sugestoes de features novas com pesquisa de mercado e analise de codigo
4
- argument-hint: ""
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Edit
9
- - Glob
10
- - Grep
11
- - Bash
12
- - Task
13
- - WebSearch
14
- - WebFetch
15
- - AskUserQuestion
16
- ---
17
- <objective>
18
- Run feature ideation with 2 parallel specialized agents (code analysis + market research), consolidate findings with ICE scoring into a prioritized report with anti-features.
19
-
20
- Standalone: works without /up:novo-projeto. Creates .plano/ideias/ automatically if not exists.
21
- Detects project stack and domain to contextualize analysis.
22
-
23
- Output: .plano/ideias/RELATORIO.md with suggestions ranked by ICE score (Impact x Confidence x Ease) and mandatory anti-features section.
24
-
25
- Optionally, after reviewing results, user can select ideas to convert into executable phases in ROADMAP.md.
26
- </objective>
27
-
28
- <execution_context>
29
- @~/.claude/up/workflows/ideias.md
30
- @~/.claude/up/references/ui-brand.md
31
- </execution_context>
32
-
33
- <context>
34
- $ARGUMENTS
35
-
36
- **Standalone mode:** This command does NOT require /up:novo-projeto or .plano/ to exist.
37
- It creates .plano/ideias/ automatically and runs the full ideation pipeline.
38
-
39
- **If .plano/ideias/ already exists:** Asks user if they want to overwrite previous ideation or cancel.
40
-
41
- **Pipeline:** 2 agents in parallel (analista-codigo + pesquisador-mercado) -> consolidador-ideias -> report
42
-
43
- **Integration with roadmap:** After ideation completes, the user can optionally select ideas to convert into ROADMAP.md phases. Requires the `phase generate-from-report` CLI subcommand.
44
- </context>
45
-
46
- <process>
47
- Execute the ideias workflow from @~/.claude/up/workflows/ideias.md end-to-end.
48
- Preserve all workflow gates (init, stack detection, agent spawn, consolidation, report presentation).
49
- </process>
@@ -1,31 +0,0 @@
1
- ---
2
- name: up:iniciar
3
- description: Registrar projeto existente no UP (leve, sem questionario)
4
- argument-hint: ""
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Bash
9
- - Glob
10
- - Grep
11
- - AskUserQuestion
12
- ---
13
- <objective>
14
- Registrar projeto no UP de forma leve. Detecta o que existe, cria PROJECT.md automaticamente sem questionario, e para. Ideal para quem quer adotar UP num projeto existente sem definir roadmap/requisitos de cara.
15
-
16
- Diferente de /up:novo-projeto que faz questionamento profundo + requisitos + roadmap, este comando apenas documenta o estado atual e configura o minimo necessario.
17
- </objective>
18
-
19
- <execution_context>
20
- @~/.claude/up/workflows/iniciar.md
21
- </execution_context>
22
-
23
- <context>
24
- $ARGUMENTS
25
-
26
- Modo leve: registra projeto, documenta o que existe, e para. O usuario vai planejando fases incrementalmente depois.
27
- </context>
28
-
29
- <process>
30
- Execute the iniciar workflow from @~/.claude/up/workflows/iniciar.md end-to-end.
31
- </process>
@@ -1,63 +0,0 @@
1
- ---
2
- name: up:mapear-codigo
3
- description: Analisar codebase existente com agentes mapeadores paralelos
4
- argument-hint: "[opcional: area especifica, ex: 'api' ou 'auth']"
5
- allowed-tools:
6
- - Read
7
- - Bash
8
- - Glob
9
- - Grep
10
- - Write
11
- - Task
12
- ---
13
-
14
- <objective>
15
- Analyze existing codebase using parallel up-mapeador-codigo agents to produce structured codebase documents.
16
-
17
- Each mapper agent explores a focus area and **writes documents directly** to `.plano/codebase/`. The orchestrator only receives confirmations, keeping context usage minimal.
18
-
19
- Output: .plano/codebase/ folder with 7 structured documents about the codebase state.
20
- </objective>
21
-
22
- <execution_context>
23
- @~/.claude/up/workflows/mapear-codigo.md
24
- @~/.claude/up/references/ui-brand.md
25
- </execution_context>
26
-
27
- <context>
28
- Focus area: $ARGUMENTS (optional - if provided, tells agents to focus on specific subsystem)
29
-
30
- **Load project state if exists:**
31
- Check for .plano/STATE.md - loads context if project already initialized
32
-
33
- **This command can run:**
34
- - Before /up:novo-projeto (brownfield codebases) - creates codebase map first
35
- - After /up:novo-projeto (greenfield codebases) - updates codebase map as code evolves
36
- - Anytime to refresh codebase understanding
37
- </context>
38
-
39
- <when_to_use>
40
- **Use mapear-codigo for:**
41
- - Projetos brownfield antes da inicializacao (entender codigo existente primeiro)
42
- - Atualizar mapa do codebase apos mudancas significativas
43
- - Onboarding em codebase desconhecido
44
- - Antes de refatoracao grande (entender estado atual)
45
-
46
- **Skip mapear-codigo for:**
47
- - Projetos greenfield sem codigo ainda (nada para mapear)
48
- - Codebases triviais (<5 arquivos)
49
- </when_to_use>
50
-
51
- <process>
52
- 1. Check if .plano/codebase/ already exists (offer to refresh or skip)
53
- 2. Create .plano/codebase/ directory structure
54
- 3. Spawn 4 parallel up-mapeador-codigo agents:
55
- - Agent 1: tech focus -> writes STACK.md, INTEGRATIONS.md
56
- - Agent 2: arch focus -> writes ARCHITECTURE.md, STRUCTURE.md
57
- - Agent 3: quality focus -> writes CONVENTIONS.md, TESTING.md
58
- - Agent 4: concerns focus -> writes CONCERNS.md
59
- 4. Wait for agents to complete, collect confirmations (NOT document contents)
60
- 5. Verify all 7 documents exist with line counts
61
- 6. Commit codebase map
62
- 7. Offer next steps (typically: /up:novo-projeto or /up:planejar-fase)
63
- </process>
@@ -1,45 +0,0 @@
1
- ---
2
- name: up:melhorias
3
- description: Auditoria completa do codebase com sugestoes priorizadas (UX, performance, modernidade)
4
- argument-hint: ""
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Edit
9
- - Glob
10
- - Grep
11
- - Bash
12
- - Task
13
- - AskUserQuestion
14
- ---
15
- <objective>
16
- Run a full codebase audit with 3 parallel specialized agents (UX, Performance, Modernidade), synthesize findings into a single prioritized report.
17
-
18
- Standalone: works without /up:novo-projeto. Creates .plano/melhorias/ automatically if not exists.
19
- Detects project stack (React/Vue/Next/Tailwind/etc.) and adjusts analysis heuristics.
20
-
21
- Output: .plano/melhorias/RELATORIO.md with suggestions organized in effort x impact matrix (4 quadrants).
22
-
23
- Optionally, after reviewing results, user can select suggestions to convert into executable phases in ROADMAP.md.
24
- </objective>
25
-
26
- <execution_context>
27
- @~/.claude/up/workflows/melhorias.md
28
- @~/.claude/up/references/ui-brand.md
29
- </execution_context>
30
-
31
- <context>
32
- $ARGUMENTS
33
-
34
- **Standalone mode:** This command does NOT require /up:novo-projeto or .plano/ to exist.
35
- It creates .plano/melhorias/ automatically and runs the full audit pipeline.
36
-
37
- **If .plano/melhorias/ already exists:** Asks user if they want to overwrite previous audit or cancel.
38
-
39
- **Integration with roadmap:** After the audit completes, the user can optionally select suggestions to convert into ROADMAP.md phases. Requires the `phase generate-from-report` CLI subcommand.
40
- </context>
41
-
42
- <process>
43
- Execute the melhorias workflow from @~/.claude/up/workflows/melhorias.md end-to-end.
44
- Preserve all workflow gates (init, stack detection, agent spawn, synthesis, report presentation).
45
- </process>
@@ -1,71 +0,0 @@
1
- ---
2
- name: up:mobile-first
3
- description: Tornar sistema responsivo para mobile/tablet sem quebrar desktop — detecta problemas via Playwright e corrige automaticamente
4
- argument-hint: "[--no-fix] [--page /rota] [porta]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Edit
9
- - Glob
10
- - Grep
11
- - Bash
12
- - Task
13
- - AskUserQuestion
14
- - mcp__plugin_playwright_playwright__*
15
- ---
16
- <objective>
17
- Mobile First: abrir o sistema no browser em multiplos viewports, detectar o que quebra no mobile/tablet e corrigir automaticamente SEM mexer na experiencia desktop.
18
-
19
- Desktop e a referencia sagrada. Cada correcao e verificada: se desktop mudou, reverte.
20
-
21
- **Standalone:** Funciona em qualquer projeto, qualquer momento. NAO requer /up:novo-projeto ou .plano/.
22
- **Builder:** Tambem integrado no modo builder (Estagio 4 — Polish).
23
-
24
- **Detecta problemas:** overflow horizontal, texto ilegivel, alvos de toque pequenos, grid/flex quebrado, imagens distorcidas, navegacao que nao cabe, modais que estourem, sidebar fixa, sobreposicao de elementos.
25
-
26
- **Corrige com:** classes Tailwind responsivas (sm:/md:/lg:), media queries CSS, componentes novos (hamburger menu, drawer, tabela responsiva), ajustes de layout.
27
-
28
- **Output:** `.plano/mobile-review/MOBILE-REPORT.md` com score de responsividade, problemas por pagina, screenshots antes/depois em todos viewports.
29
- </objective>
30
-
31
- <execution_context>
32
- @~/.claude/up/workflows/mobile-first.md
33
- @~/.claude/up/references/ui-brand.md
34
- </execution_context>
35
-
36
- <context>
37
- $ARGUMENTS
38
-
39
- **Flags:**
40
- - `--no-fix` — Apenas escanear e gerar relatorio, NAO corrigir
41
- - `--page /rota` — Testar apenas uma pagina especifica (ex: `--page /dashboard`)
42
-
43
- **Argumentos opcionais:**
44
- - Porta: `3000` ou `http://localhost:3000` (default: detecta automaticamente)
45
-
46
- **Exemplos:**
47
- ```
48
- /up:mobile-first # Escanear tudo e corrigir
49
- /up:mobile-first --no-fix # Apenas relatorio
50
- /up:mobile-first --page /dashboard # Apenas uma pagina
51
- /up:mobile-first 5173 # Especificar porta
52
- /up:mobile-first --page /settings 3000 # Combinar flags
53
- ```
54
-
55
- **Cria .plano/mobile-review/ automaticamente** (standalone, sem pre-requisitos).
56
- </context>
57
-
58
- <process>
59
- Execute the mobile-first workflow from @~/.claude/up/workflows/mobile-first.md end-to-end.
60
-
61
- Preserve all workflow gates:
62
- 1. Setup (detectar CSS stack, subir server, descobrir paginas)
63
- 2. Scan (capturar todas paginas em mobile/tablet/desktop, detectar problemas)
64
- 3. Analyze (mapear problema → arquivo → estrategia de fix)
65
- 4. Fix (corrigir cada problema, verificar desktop intacto apos cada um)
66
- 5. Report (MOBILE-REPORT.md com screenshots comparativos)
67
- 6. Cleanup (matar server, fechar browser)
68
-
69
- **Flag --no-fix:** Se presente, pular passo 4 (Fix). Apenas escanear e reportar.
70
- **Flag --page:** Se presente, filtrar para apenas a pagina especificada.
71
- </process>
@@ -1,186 +0,0 @@
1
- ---
2
- name: up:modo-builder
3
- description: Construir projeto completo de forma autonoma — greenfield (do zero) ou brownfield (feature nova em projeto existente)
4
- argument-hint: "[--light] [descricao do projeto ou feature]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Edit
9
- - Glob
10
- - Grep
11
- - Bash
12
- - Task
13
- - WebFetch
14
- - WebSearch
15
- - AskUserQuestion
16
- - mcp__context7__*
17
- - mcp__plugin_playwright_playwright__*
18
- ---
19
- <objective>
20
- Modo Builder: construir projeto completo de forma totalmente autonoma.
21
-
22
- Detecta automaticamente o modo:
23
- - **Greenfield** (sem codigo existente): cria projeto do zero
24
- - **Brownfield** (codigo existente): mapeia codebase, planeja e implementa feature/mudanca
25
-
26
- **Dois niveis de execucao:**
27
- - **Full (padrao):** Pipeline completo com pesquisa, polish, UX review, ideias, delivery
28
- - **Light (`--light`):** Pipeline enxuto — planeja, constroi, testa. Sem gordura. Ideal para features medias em projetos existentes.
29
-
30
- O usuario fornece um briefing, responde perguntas criticas (apenas sobre credenciais/APIs/ambiguidades), e o sistema faz TUDO sozinho.
31
-
32
- **Modo Full (padrao):**
33
- 1. Detecta modo → Pesquisa/Mapeia → Estrutura → Planeja → Executa → Verifica → E2E
34
- 2. Polish (melhorias + UX tester + ideias) → DELIVERY.md
35
-
36
- **Modo Light (`--light`):**
37
- 1. Detecta modo → Mini-scan → Estrutura inline → Planeja → Executa → Verifica → E2E
38
- 2. Fim. Sem polish, sem delivery, sem pesquisa pesada.
39
- </objective>
40
-
41
- <execution_context>
42
- @~/.claude/up/workflows/builder.md
43
- @~/.claude/up/workflows/builder-e2e.md
44
- @~/.claude/up/workflows/ux-tester.md
45
- @~/.claude/up/workflows/mobile-first.md
46
- @~/.claude/up/references/ui-brand.md
47
- </execution_context>
48
-
49
- <context>
50
- $ARGUMENTS
51
-
52
- **Flags:**
53
- - `--light` — Modo enxuto. Pula pesquisa, polish, UX tester, ideias, delivery, reassessment, captures. Mantém: planejar, executar, verificar, E2E Playwright.
54
-
55
- O restante do argumento e o briefing em texto livre. Pode incluir:
56
-
57
- **Greenfield (projeto novo):**
58
- - O que quer construir
59
- - Para quem (publico)
60
- - Stack desejada (ou usa builder-defaults.md)
61
- - Features principais
62
- - Credenciais/APIs que ja tem
63
-
64
- **Brownfield (projeto existente):**
65
- - Que feature/mudanca quer implementar
66
- - Como se integra com o existente
67
- - Novas APIs/integrações necessarias
68
- - Restricoes ou areas que nao devem ser tocadas
69
-
70
- Se $ARGUMENTS estiver vazio (alem de flags), o workflow pedira o briefing interativamente.
71
-
72
- **Deteccao automatica:** Se ha codigo existente no diretorio (package.json, src/, etc.) ou .plano/ existente, o builder entra em modo brownfield automaticamente.
73
-
74
- **Defaults:** Lidos de `~/.claude/up/builder-defaults.md` (se existir).
75
- Em brownfield, convencoes do codebase existente tem prioridade sobre defaults.
76
- </context>
77
-
78
- <process>
79
- **GATE OBRIGATORIO — Owner Profile:**
80
- Antes de qualquer coisa, verificar se `~/.claude/up/owner-profile.md` existe.
81
- Se NAO existir: rodar `/up:onboard` primeiro (workflow onboarding.md) pra criar profile.
82
- Sem profile, o CEO nao pode conduzir intake.
83
-
84
- **Parsear flags primeiro:** Extrair `--light` dos $ARGUMENTS se presente. O restante e o briefing.
85
-
86
- **GUARD: Light mode SOMENTE se `--light` esta presente LITERALMENTE nos argumentos.**
87
- NAO inferir light baseado no tamanho do briefing. Briefing curto = FULL com poucas fases.
88
-
89
- **Se `--light` PRESENTE nos argumentos:**
90
- Execute o builder workflow em modo light (ver secao `<light_process>` no workflow).
91
- Light ainda verifica (up-verificador), testa (E2E + DCRV 1 ciclo), mas sem polish/delivery.
92
-
93
- **Se `--light` AUSENTE (default = FULL):**
94
- Execute the builder workflow from @~/.claude/up/workflows/builder.md end-to-end.
95
-
96
- **CRITICO:** A partir do Estagio 2, ZERO interacao com usuario. NAO use AskUserQuestion apos coletar o briefing e respostas criticas. Toda decisao e tomada autonomamente.
97
-
98
- **Modo Full — Preserve all workflow gates:**
99
- - Estagio 1: Intake (detectar modo + briefing + perguntas criticas)
100
- - Estagio 2: Arquitetura (mapear codebase OU pesquisar ecossistema + up-arquiteto)
101
- - Estagio 3: Build (loop planejar → executar → verificar → E2E por fase)
102
- - Estagio 4: Polish (melhorias + quick wins + UX tester + ideias)
103
- - Estagio 5: Entrega (E2E final + captures + DELIVERY.md + resumo)
104
-
105
- Falhas sao contornadas, nunca bloqueiam. O builder SEMPRE entrega algo.
106
- </process>
107
-
108
- <light_mode>
109
- ## Modo Light — Pipeline Enxuto
110
-
111
- Quando `--light` esta presente, o builder roda um pipeline minimalista:
112
-
113
- ### Estagio 1 Light: Intake Rapido
114
-
115
- 1. Detectar modo (greenfield/brownfield) — mesmo processo
116
- 2. Receber briefing
117
- 3. Perguntas criticas — SIM, mesma logica (so perguntar o essencial)
118
- 4. **NAO carregar builder-defaults.md** — usar stack do codebase (brownfield) ou inferir (greenfield)
119
-
120
- ### Estagio 2 Light: Estrutura Inline
121
-
122
- **Brownfield:**
123
- - Se `.plano/codebase/` existe e tem < 7 dias: reutilizar
124
- - Se nao existe: mini-scan inline (sem spawnar 4 agentes mapeadores):
125
- ```bash
126
- # Mini-scan rapido
127
- ls package.json requirements.txt pyproject.toml 2>/dev/null
128
- cat package.json 2>/dev/null | head -50
129
- ls -d src/ app/ lib/ pages/ components/ 2>/dev/null | head -10
130
- find . -name "*.ts" -o -name "*.tsx" | head -30
131
- ```
132
- - **Estruturar inline** (sem spawnar up-arquiteto):
133
- - Criar/atualizar PROJECT.md com briefing e contexto minimo
134
- - Criar/atualizar REQUIREMENTS.md com requisitos da feature
135
- - Criar/atualizar ROADMAP.md com 1-3 fases max
136
- - config.json com `builder_mode: true, builder_type: light`
137
-
138
- **Greenfield:**
139
- - **Pular pesquisa** (sem 4 pesquisadores paralelos)
140
- - Estruturar inline com inferencia inteligente
141
- - 2-5 fases max
142
-
143
- ### Estagio 3 Light: Build + E2E
144
-
145
- Mesmo loop do full, mas:
146
- - **Sem LOCK.md** (sessao curta, nao precisa de crash recovery)
147
- - **Sem reassessment** (poucas fases, nao vale a pena)
148
- - **Sem captures** (sessao curta)
149
- - **COM teste E2E Playwright** (se tem UI) — mesmo processo do full
150
-
151
- ### Sem Estagio 4 (Polish)
152
-
153
- Nao roda melhorias, UX tester, nem ideias.
154
-
155
- ### Sem Estagio 5 (Entrega)
156
-
157
- Nao gera DELIVERY.md. Apenas exibe resumo inline:
158
-
159
- ```
160
- ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
161
- UP > BUILDER LIGHT — COMPLETO
162
- ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
163
-
164
- **Feature:** [resumo]
165
- **Fases:** [N] completadas
166
- **Commits:** [N]
167
- **E2E:** [N] testes, [X] passaram
168
- **Bugs E2E:** [N] encontrados, [M] corrigidos
169
-
170
- ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
171
- ```
172
-
173
- ### Comparacao de Tokens (estimativa)
174
-
175
- | Etapa | Full | Light |
176
- |-------|------|-------|
177
- | Pesquisa (4 agentes) | ~80k tokens | 0 |
178
- | Mapeamento (4 agentes) | ~60k tokens | ~5k (mini-scan) |
179
- | Arquiteto (agente) | ~40k tokens | ~10k (inline) |
180
- | Build (por fase) | ~100k tokens | ~100k tokens |
181
- | E2E (por fase) | ~30k tokens | ~30k tokens |
182
- | Polish (7+ agentes) | ~150k tokens | 0 |
183
- | Delivery | ~20k tokens | 0 |
184
- | **Total (3 fases)** | **~680k** | **~345k** (~50% menos) |
185
-
186
- </light_mode>
@@ -1,40 +0,0 @@
1
- ---
2
- name: up:novo-projeto
3
- description: Inicializar projeto (detecta greenfield/brownfield automaticamente)
4
- argument-hint: ""
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Bash
9
- - Glob
10
- - Grep
11
- - Task
12
- - WebFetch
13
- - WebSearch
14
- - AskUserQuestion
15
- - mcp__context7__*
16
- ---
17
- <objective>
18
- Inicializar projeto com coleta de contexto estruturada. Detecta automaticamente se e greenfield (sem codigo) ou brownfield (codigo existente) e adapta questionamento, pesquisa e requisitos.
19
-
20
- **Greenfield:** Questioning → Research → Requirements → Roadmap → PROJECT.md
21
- **Brownfield:** Codebase map → Questioning adaptado → Requisitos inferidos + novos → Roadmap → PROJECT.md
22
-
23
- **Orchestrator role:** Detectar modo, carregar mapeamento do codebase (se brownfield), guiar questionamento adaptado, sintetizar requisitos, gerar roadmap com contexto completo.
24
- </objective>
25
-
26
- <execution_context>
27
- @~/.claude/up/workflows/novo-projeto.md
28
- @~/.claude/up/references/ui-brand.md
29
- </execution_context>
30
-
31
- <context>
32
- $ARGUMENTS
33
-
34
- Inicializacao coleta contexto por questionamento interativo. Se codigo existente for detectado, adapta o fluxo para brownfield: carrega mapeamento do codebase, infere requisitos validados, e pergunta sobre novos objetivos.
35
- </context>
36
-
37
- <process>
38
- Execute the novo-projeto workflow from @~/.claude/up/workflows/novo-projeto.md end-to-end.
39
- Preserve all workflow gates (mode detection, codebase loading, questioning, research, requirements, roadmap generation, PROJECT.md creation).
40
- </process>
@@ -1,69 +0,0 @@
1
- ---
2
- name: up:onboard
3
- description: Entrevista inicial pro CEO te conhecer e criar seu perfil (~/.claude/up/owner-profile.md)
4
- argument-hint: "[--update]"
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Bash
9
- - AskUserQuestion
10
- ---
11
- <objective>
12
- Onboarding do UP: entrevista conduzida pelo CEO que cria seu perfil pessoal.
13
-
14
- O perfil e lido por todos os agentes UP antes de cada projeto pra adaptar:
15
- - Tom de comunicacao
16
- - Stack padrao
17
- - Estilo de decisao
18
- - Nome do seu CEO
19
- - Restricoes permanentes
20
-
21
- Rodado automaticamente no primeiro uso de qualquer comando UP.
22
- Rodado manualmente via `/up:onboard` ou `/up:onboard --update` pra refazer.
23
- </objective>
24
-
25
- <execution_context>
26
- @~/.claude/up/workflows/onboarding.md
27
- @~/.claude/up/templates/owner-profile.md
28
- </execution_context>
29
-
30
- <context>
31
- $ARGUMENTS
32
-
33
- **Flags:**
34
- - `--update` — Refaz o onboarding, sobrescreve perfil existente
35
-
36
- Sem argumentos: roda se nao existe, pula se ja existe.
37
- </context>
38
-
39
- <process>
40
- Execute the onboarding workflow from @~/.claude/up/workflows/onboarding.md end-to-end.
41
-
42
- **Passo 0 obrigatorio:**
43
- Verificar se `~/.claude/up/owner-profile.md` ja existe.
44
-
45
- ```bash
46
- if [ -f ~/.claude/up/owner-profile.md ]; then
47
- if [[ "$ARGUMENTS" == *"--update"* ]]; then
48
- echo "Atualizando perfil existente..."
49
- # Prosseguir
50
- else
51
- echo "Perfil ja existe. Use /up:onboard --update pra refazer."
52
- cat ~/.claude/up/owner-profile.md | head -20
53
- exit 0
54
- fi
55
- fi
56
- ```
57
-
58
- Seguir workflow completo:
59
- 1. Apresentacao
60
- 2. Bloco 1: Identidade (nome, papel, localizacao, idioma)
61
- 3. Bloco 2: Nome do CEO (como voce me chama) + tom
62
- 4. Bloco 3: Contexto profissional
63
- 5. Bloco 4: Stack preferida
64
- 6. Bloco 5: Estilo de trabalho
65
- 7. Bloco 6: Restricoes permanentes
66
- 8. Bloco 7: Integracoes disponiveis
67
- 9. Gerar owner-profile.md
68
- 10. Confirmar resumo
69
- </process>
@@ -1,33 +0,0 @@
1
- ---
2
- name: up:pausar
3
- description: Criar handoff .continue-aqui.md ao pausar trabalho
4
- argument-hint: ""
5
- allowed-tools:
6
- - Read
7
- - Write
8
- - Bash
9
- - Glob
10
- - AskUserQuestion
11
- ---
12
- <objective>
13
- Create a handoff document when pausing work so the next session can resume seamlessly.
14
-
15
- **Default flow:** Capture current state -> Document in-progress work -> Write .continue-aqui.md -> Confirm
16
-
17
- **Orchestrator role:** Gather current task state, uncommitted changes, pending decisions, and next steps into a structured .continue-aqui.md handoff file.
18
- </objective>
19
-
20
- <execution_context>
21
- @~/.claude/up/workflows/pausar.md
22
- </execution_context>
23
-
24
- <context>
25
- $ARGUMENTS
26
-
27
- Creates .continue-aqui.md in the project root with all context needed to resume work.
28
- </context>
29
-
30
- <process>
31
- Execute the pausar workflow from @~/.claude/up/workflows/pausar.md end-to-end.
32
- Preserve all workflow gates (state capture, change detection, handoff creation, confirmation).
33
- </process>