@jaimevalasek/aioson 1.5.1 → 1.6.0

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 (131) hide show
  1. package/README.md +6 -0
  2. package/docs/design-previews/aurora-command-ui-website.html +884 -0
  3. package/docs/design-previews/aurora-command-ui.html +682 -0
  4. package/docs/design-previews/bold-editorial-ui-website.html +658 -0
  5. package/docs/design-previews/bold-editorial-ui.html +717 -0
  6. package/docs/design-previews/clean-saas-ui-website.html +1202 -0
  7. package/docs/design-previews/clean-saas-ui.html +549 -0
  8. package/docs/design-previews/cognitive-core-ui-website.html +1009 -0
  9. package/docs/design-previews/cognitive-core-ui.html +463 -0
  10. package/docs/design-previews/glassmorphism-ui-website.html +572 -0
  11. package/docs/design-previews/glassmorphism-ui.html +886 -0
  12. package/docs/design-previews/index.html +699 -0
  13. package/docs/design-previews/interface-design-website.html +1187 -0
  14. package/docs/design-previews/interface-design.html +513 -0
  15. package/docs/design-previews/neo-brutalist-ui-website.html +621 -0
  16. package/docs/design-previews/neo-brutalist-ui.html +797 -0
  17. package/docs/design-previews/premium-command-center-ui-website.html +1217 -0
  18. package/docs/design-previews/premium-command-center-ui.html +552 -0
  19. package/docs/design-previews/warm-craft-ui-website.html +684 -0
  20. package/docs/design-previews/warm-craft-ui.html +739 -0
  21. package/docs/en/cli-reference.md +20 -9
  22. package/docs/pt/README.md +7 -0
  23. package/docs/pt/agent-sharding.md +132 -0
  24. package/docs/pt/agentes.md +8 -2
  25. package/docs/pt/busca-de-contexto.md +129 -0
  26. package/docs/pt/cache-de-contexto.md +156 -0
  27. package/docs/pt/comandos-cli.md +28 -0
  28. package/docs/pt/design-hybrid-forge.md +107 -0
  29. package/docs/pt/inicio-rapido.md +54 -3
  30. package/docs/pt/inteligencia-adaptativa.md +324 -0
  31. package/docs/pt/monitor-de-contexto.md +104 -0
  32. package/docs/pt/recuperacao-de-sessao.md +125 -0
  33. package/docs/pt/sandbox.md +125 -0
  34. package/docs/pt/skills.md +98 -6
  35. package/package.json +1 -1
  36. package/src/agent-loader.js +280 -0
  37. package/src/cli.js +94 -0
  38. package/src/commands/agent-loader.js +85 -0
  39. package/src/commands/context-cache.js +90 -0
  40. package/src/commands/context-monitor.js +92 -0
  41. package/src/commands/context-search.js +66 -0
  42. package/src/commands/design-hybrid-options.js +385 -0
  43. package/src/commands/health.js +214 -0
  44. package/src/commands/init.js +54 -13
  45. package/src/commands/install.js +52 -13
  46. package/src/commands/learning-evolve.js +355 -0
  47. package/src/commands/live.js +34 -0
  48. package/src/commands/recovery.js +43 -0
  49. package/src/commands/sandbox.js +37 -0
  50. package/src/commands/setup-context.js +22 -2
  51. package/src/commands/setup.js +178 -0
  52. package/src/commands/skill.js +79 -32
  53. package/src/commands/tool-registry-cmd.js +232 -0
  54. package/src/commands/update.js +7 -0
  55. package/src/constants.js +9 -0
  56. package/src/context-cache.js +159 -0
  57. package/src/context-search.js +326 -0
  58. package/src/design-variation-catalog.js +503 -0
  59. package/src/i18n/messages/en.js +32 -2
  60. package/src/i18n/messages/es.js +30 -2
  61. package/src/i18n/messages/fr.js +30 -2
  62. package/src/i18n/messages/pt-BR.js +32 -2
  63. package/src/install-animation.js +260 -0
  64. package/src/install-profile.js +143 -0
  65. package/src/install-wizard.js +474 -0
  66. package/src/installer.js +38 -10
  67. package/src/parser.js +7 -1
  68. package/src/recovery-context-session.js +154 -0
  69. package/src/runtime-store.js +97 -1
  70. package/src/sandbox.js +177 -0
  71. package/src/tool-executor.js +94 -0
  72. package/src/updater.js +11 -3
  73. package/template/.aioson/agents/analyst.md +58 -3
  74. package/template/.aioson/agents/architect.md +38 -0
  75. package/template/.aioson/agents/design-hybrid-forge.md +127 -0
  76. package/template/.aioson/agents/dev.md +103 -0
  77. package/template/.aioson/agents/deyvin.md +57 -0
  78. package/template/.aioson/agents/pm.md +58 -0
  79. package/template/.aioson/agents/product.md +28 -0
  80. package/template/.aioson/agents/qa.md +79 -0
  81. package/template/.aioson/agents/setup.md +65 -3
  82. package/template/.aioson/agents/sheldon.md +107 -6
  83. package/template/.aioson/agents/tester.md +156 -0
  84. package/template/.aioson/config.md +15 -0
  85. package/template/.aioson/context/forensics/.gitkeep +0 -0
  86. package/template/.aioson/context/seeds/seed-example.md +27 -0
  87. package/template/.aioson/context/user-profile.md +42 -0
  88. package/template/.aioson/locales/en/agents/setup.md +33 -1
  89. package/template/.aioson/locales/es/agents/setup.md +33 -1
  90. package/template/.aioson/locales/fr/agents/setup.md +33 -1
  91. package/template/.aioson/locales/pt-BR/agents/setup.md +33 -1
  92. package/template/.aioson/skills/design/aurora-command-ui/SKILL.md +243 -0
  93. package/template/.aioson/skills/design/aurora-command-ui/references/art-direction.md +293 -0
  94. package/template/.aioson/skills/design/aurora-command-ui/references/components.md +827 -0
  95. package/template/.aioson/skills/design/aurora-command-ui/references/dashboards.md +250 -0
  96. package/template/.aioson/skills/design/aurora-command-ui/references/design-tokens.md +585 -0
  97. package/template/.aioson/skills/design/aurora-command-ui/references/motion.md +365 -0
  98. package/template/.aioson/skills/design/aurora-command-ui/references/patterns.md +482 -0
  99. package/template/.aioson/skills/design/aurora-command-ui/references/websites.md +387 -0
  100. package/template/.aioson/skills/design/glassmorphism-ui/SKILL.md +222 -0
  101. package/template/.aioson/skills/design/glassmorphism-ui/references/art-direction.md +159 -0
  102. package/template/.aioson/skills/design/glassmorphism-ui/references/components.md +498 -0
  103. package/template/.aioson/skills/design/glassmorphism-ui/references/dashboards.md +236 -0
  104. package/template/.aioson/skills/design/glassmorphism-ui/references/design-tokens.md +274 -0
  105. package/template/.aioson/skills/design/glassmorphism-ui/references/motion.md +355 -0
  106. package/template/.aioson/skills/design/glassmorphism-ui/references/patterns.md +198 -0
  107. package/template/.aioson/skills/design/glassmorphism-ui/references/websites.md +307 -0
  108. package/template/.aioson/skills/design/neo-brutalist-ui/SKILL.md +213 -0
  109. package/template/.aioson/skills/design/neo-brutalist-ui/references/art-direction.md +228 -0
  110. package/template/.aioson/skills/design/neo-brutalist-ui/references/components.md +855 -0
  111. package/template/.aioson/skills/design/neo-brutalist-ui/references/dashboards.md +334 -0
  112. package/template/.aioson/skills/design/neo-brutalist-ui/references/design-tokens.md +342 -0
  113. package/template/.aioson/skills/design/neo-brutalist-ui/references/motion.md +286 -0
  114. package/template/.aioson/skills/design/neo-brutalist-ui/references/patterns.md +458 -0
  115. package/template/.aioson/skills/design/neo-brutalist-ui/references/websites.md +723 -0
  116. package/template/.aioson/skills/process/aioson-spec-driven/SKILL.md +45 -0
  117. package/template/.aioson/skills/process/aioson-spec-driven/references/approval-gates.md +109 -0
  118. package/template/.aioson/skills/process/aioson-spec-driven/references/artifact-map.md +44 -0
  119. package/template/.aioson/skills/process/aioson-spec-driven/references/classification-map.md +37 -0
  120. package/template/.aioson/skills/process/aioson-spec-driven/references/hardening-lane.md +49 -0
  121. package/template/.aioson/skills/process/aioson-spec-driven/references/maintenance-and-state.md +66 -0
  122. package/template/.aioson/skills/process/aioson-spec-driven/references/ui-language.md +75 -0
  123. package/template/.aioson/skills/process/design-hybrid-forge/SKILL.md +144 -0
  124. package/template/.aioson/skills/process/design-hybrid-forge/references/crossover-protocol.md +221 -0
  125. package/template/.aioson/skills/process/design-hybrid-forge/references/naming-registry.md +88 -0
  126. package/template/.aioson/skills/process/design-hybrid-forge/references/output-contract.md +291 -0
  127. package/template/.aioson/skills/process/design-hybrid-forge/references/pair-compatibility.md +117 -0
  128. package/template/.aioson/skills/process/design-hybrid-forge/references/quality-gates.md +188 -0
  129. package/template/.aioson/skills/process/design-hybrid-forge/references/variation-library.md +125 -0
  130. package/template/AGENTS.md +23 -1
  131. package/template/CLAUDE.md +1 -0
@@ -0,0 +1,45 @@
1
+ # Skill: aioson-spec-driven
2
+
3
+ > Process methodology skill. Covers: phase sequencing, artifact contracts, approval gates, and hardening lane.
4
+ > Load this file first. Then load only the `references/` file relevant to your current role and phase.
5
+
6
+ ## When to use
7
+
8
+ Load this skill when:
9
+ - starting spec work for a new feature or project (any agent)
10
+ - deciding phase depth based on classification (MICRO / SMALL / MEDIUM)
11
+ - preparing a clean handoff to the next agent
12
+ - retaking work after a session break (check `last_checkpoint` + `phase_gates` first)
13
+
14
+ Do NOT load the entire `references/` folder. Load only the file matching your current need.
15
+
16
+ ## What phases exist
17
+
18
+ | Phase | AIOSON artifact | Primary agent | MICRO | SMALL | MEDIUM |
19
+ |-------|----------------|---------------|-------|-------|--------|
20
+ | Specify | `prd*.md` | @product | lite | full | full |
21
+ | Research/Discuss | `sheldon-enrichment*.md` | @sheldon | optional | recommended | required |
22
+ | Requirements | `requirements-{slug}.md` | @analyst | skip | required | required |
23
+ | Design | `architecture.md`, `design-doc*.md` | @architect | skip | selective | required |
24
+ | Tasks/Plan | `implementation-plan*.md` | @dev | optional | recommended | required |
25
+ | Execute | code, commits, spec updates | @dev, @deyvin | — | — | — |
26
+ | State/Resume | `spec*.md`, runtime | @dev, @deyvin | — | — | — |
27
+
28
+ ## Phase depth by classification
29
+
30
+ - **MICRO**: Specify (lite) + Execute. Skip Requirements, Design, Plan unless complexity warrants it.
31
+ - **SMALL**: Specify + Requirements + selective Design + Plan. @sheldon recommended before downstream.
32
+ - **MEDIUM**: Full pack — all phases, all artifacts, @sheldon validation before @analyst, implementation plan required.
33
+
34
+ ## References available
35
+
36
+ Load the file that matches your current context — do not load all at once:
37
+
38
+ | File | Load when |
39
+ |------|-----------|
40
+ | `references/artifact-map.md` | You need to know which artifact lives where and who owns it |
41
+ | `references/classification-map.md` | You need to decide phase depth for a project or feature |
42
+ | `references/approval-gates.md` | You are preparing a handoff and need to know what must be ready |
43
+ | `references/hardening-lane.md` | The input is vague, exploratory, or "vibe-style" and needs to be converted |
44
+ | `references/maintenance-and-state.md` | You are writing or reading `spec*.md`, checkpoints, or retaking work |
45
+ | `references/ui-language.md` | You need to present options, status symbols, or checkpoints to the user with consistent visual standards |
@@ -0,0 +1,109 @@
1
+ # Approval Gates — Spec-Driven Phase Transitions
2
+
3
+ > Use when preparing a handoff or checking if a phase is ready to proceed.
4
+ > Gates are blocking in MEDIUM. Informational in MICRO/SMALL.
5
+
6
+ ## Gate A — Requirements approval
7
+
8
+ **Before leaving @product / @analyst → @architect or @dev**
9
+
10
+ Required for the gate to pass:
11
+ - [ ] Objectives are clear and unambiguous
12
+ - [ ] Expected behaviors are described (not just features — what happens when X)
13
+ - [ ] Constraints are explicit (what this version does NOT do)
14
+ - [ ] Out of scope is listed
15
+ - [ ] Open ambiguities are documented (not silently ignored)
16
+ - [ ] Requirement IDs exist for all business rules: `REQ-{slug}-{N}`
17
+ - [ ] Acceptance criteria exist for all behavioral requirements: `AC-{slug}-{N}`
18
+
19
+ **Signal in `spec-{slug}.md`:** set `phase_gates.requirements: approved`
20
+
21
+ **Not ready if:**
22
+ - Requirements reference behavior that hasn't been decided
23
+ - ACs are not independently verifiable (e.g., "works correctly" is not an AC)
24
+ - There are contradictions between PRD and requirements
25
+
26
+ ---
27
+
28
+ ## Gate B — Design approval
29
+
30
+ **Before leaving @architect → @dev**
31
+
32
+ Required for the gate to pass:
33
+ - [ ] Technical approach is chosen and documented
34
+ - [ ] Module/folder structure is defined
35
+ - [ ] Dependencies between components are explicit
36
+ - [ ] Risks and non-goals are documented
37
+ - [ ] Decision rationale exists for non-obvious choices
38
+ - [ ] @dev can start without having to invent business logic or architecture
39
+
40
+ **Signal in `spec-{slug}.md`:** set `phase_gates.design: approved`
41
+ **Signal in `architecture.md`:** closing line `> **Gate B:** Architecture approved — @dev can proceed.`
42
+
43
+ **Not ready if:**
44
+ - There are open decisions that @dev would need to make up during implementation
45
+ - The architecture introduces patterns not supported by the project's stack
46
+ - There are circular dependencies in the module structure
47
+
48
+ ---
49
+
50
+ ## Gate C — Plan approval
51
+
52
+ **Before @dev or @deyvin executes a significant batch**
53
+
54
+ Required for the gate to pass:
55
+ - [ ] Execution sequence is defined
56
+ - [ ] Checkpoints are listed with criteria of done
57
+ - [ ] Context package is listed (which files to read before each phase)
58
+ - [ ] Review / QA requirements are noted
59
+ - [ ] Decisions marked "pre-taken" are FINAL — @dev does not re-discuss
60
+
61
+ **Signal in `spec-{slug}.md`:** set `phase_gates.plan: approved`
62
+ **Signal in `implementation-plan-{slug}.md`:** status field = `approved`
63
+
64
+ **Not ready if:**
65
+ - The plan has phases with unclear done criteria
66
+ - There are external dependencies not yet resolved
67
+ - @dev would need to re-read the full spec from scratch to start
68
+
69
+ ---
70
+
71
+ ## Gate D — Execution verification (must_haves)
72
+
73
+ **Before marking a phase or feature as complete**
74
+
75
+ Required:
76
+ - [ ] All truths verified (behavioral — not just "I think it works")
77
+ - [ ] All artifacts verified (substantive — not stubs)
78
+ - [ ] All key_links verified (wiring — imports, registrations, middleware)
79
+
80
+ **Not complete if:**
81
+ - Any truth has no passing test or manual verification step
82
+ - Any artifact is empty, a stub, or missing required exports
83
+ - Any key_link shows the code exists but isn't connected
84
+
85
+ **Signal:** update `last_checkpoint` in `spec-{slug}.md` with which must_haves passed and which failed.
86
+
87
+ ---
88
+
89
+ ## Checkpoint taxonomy reference
90
+
91
+ Quando qualquer agente precisa de interação humana durante execução:
92
+
93
+ | Tipo | Quando | Frequência esperada |
94
+ |------|--------|---------------------|
95
+ | `verify` | Confirmar comportamento visível | Comum — após cada entrega testável |
96
+ | `decision` | Bifurcação arquitetural ou de produto | Rara — apenas quando não especificado |
97
+ | `action` | Passo que o agente não consegue executar | Muito rara — <1% dos steps |
98
+
99
+ Regra: se o agente consegue executar sem risco, execute. Não peça confirmação desnecessária.
100
+
101
+ ---
102
+
103
+ ## How agents communicate gate status
104
+
105
+ Each agent, at session end, should:
106
+ 1. Update `phase_gates` in `spec-{slug}.md`
107
+ 2. Tell the user clearly: "Gate [A/B/C] passed — activate [@next-agent]" OR "Gate [A/B/C] blocked — [reason]. Resolve before continuing."
108
+
109
+ Never silently assume a gate is passed.
@@ -0,0 +1,44 @@
1
+ # Artifact Map — AIOSON Spec-Driven
2
+
3
+ > Source of truth: which artifact lives where, who writes it, who reads it.
4
+
5
+ ## Artifact chain
6
+
7
+ ```
8
+ prd*.md
9
+ → sheldon-enrichment-{slug}.md (optional enrichment layer)
10
+ → requirements-{slug}.md (entities, rules, ACs — @analyst)
11
+ → spec-{slug}.md (feature memory — @analyst seeds, @dev fills)
12
+ → architecture.md (tech decisions — @architect)
13
+ → design-doc*.md (scope-specific decisions — @architect)
14
+ → implementation-plan-{slug}.md (execution sequence — @dev)
15
+ → spec-{slug}.md (updated) (living state during execution — @dev)
16
+ ```
17
+
18
+ ## Artifact ownership
19
+
20
+ | Artifact | Written by | Enriched by | Read by |
21
+ |----------|-----------|-------------|---------|
22
+ | `prd.md` / `prd-{slug}.md` | @product | @sheldon (via sheldon-enrichment) | all downstream |
23
+ | `sheldon-enrichment-{slug}.md` | @sheldon | @sheldon | @sheldon (re-entry), @analyst |
24
+ | `sheldon-validation.md` | @sheldon | — | all agents before starting MEDIUM |
25
+ | `discovery.md` | @analyst | — | @architect, @dev |
26
+ | `requirements-{slug}.md` | @analyst | — | @architect, @dev, @qa |
27
+ | `spec-{slug}.md` | @analyst (skeleton) | @dev (execution state) | @dev, @deyvin, @qa |
28
+ | `spec.md` | @dev | @dev | @dev, @deyvin |
29
+ | `architecture.md` | @architect | — | @dev, @ux-ui |
30
+ | `design-doc*.md` | @architect | — | @dev, @ux-ui |
31
+ | `implementation-plan-{slug}.md` | @dev | — | @dev, @deyvin |
32
+
33
+ ## Naming conventions
34
+
35
+ - Project-level artifacts: `prd.md`, `discovery.md`, `spec.md`, `architecture.md`
36
+ - Feature-level artifacts: always use `{slug}` suffix — `prd-{slug}.md`, `requirements-{slug}.md`, `spec-{slug}.md`
37
+ - Enrichment: `sheldon-enrichment.md` (project) or `sheldon-enrichment-{slug}.md` (feature)
38
+ - Plans: `.aioson/plans/{slug}/manifest.md` + `plan-{slug-fase}.md` files
39
+
40
+ ## What NOT to create
41
+
42
+ - Do not create `.specs/` folders — use the artifact names above
43
+ - Do not rename existing artifacts to match TLC conventions
44
+ - Do not create a new artifact if an existing one covers the same purpose
@@ -0,0 +1,37 @@
1
+ # Classification Map — Phase Depth by Project Size
2
+
3
+ > Use this when deciding which phases to run and how deep to go.
4
+
5
+ ## Depth table
6
+
7
+ | Phase | MICRO (0–1) | SMALL (2–3) | MEDIUM (4–6) |
8
+ |-------|-------------|-------------|--------------|
9
+ | Specify (PRD) | 1 conversation, lite template | Full PRD conversation | Full PRD + `## Specify depth` section |
10
+ | Research (@sheldon) | Skip unless links/external sources provided | Recommended — run before @analyst | Required — run Modo C (full validation) |
11
+ | Requirements (@analyst) | Skip — go @product → @dev | Required — `requirements-{slug}.md` | Required — with requirement IDs + ACs |
12
+ | Design (@architect) | Skip unless auth or external integration | Selective — only if new architecture pattern | Required — full `architecture.md` |
13
+ | Plan (implementation-plan) | Optional — suggest only if @dev asks | Recommended | Required — with gate approval + verification criteria |
14
+ | Execute (@dev) | Direct from PRD | From requirements + spec | From approved plan only |
15
+ | State (@dev, @deyvin) | Minimal `spec.md` note | `spec-{slug}.md` with phase_gates | Full spec pack — phase_gates + checkpoints + maintenance notes |
16
+
17
+ ## Scoring (from @analyst)
18
+
19
+ | Dimension | Score |
20
+ |-----------|-------|
21
+ | User types: 1 | 0 |
22
+ | User types: 2 | 1 |
23
+ | User types: 3+ | 2 |
24
+ | External integrations: 0 | 0 |
25
+ | External integrations: 1–2 | 1 |
26
+ | External integrations: 3+ | 2 |
27
+ | Business rule complexity: none | 0 |
28
+ | Business rule complexity: some | 1 |
29
+ | Business rule complexity: complex | 2 |
30
+
31
+ **0–1 = MICRO / 2–3 = SMALL / 4–6 = MEDIUM**
32
+
33
+ ## Gate behavior by classification
34
+
35
+ - **MICRO**: gates are informational, never blocking. @dev may proceed without explicit approval.
36
+ - **SMALL**: Gate A (requirements) is recommended before design. Gate B is optional.
37
+ - **MEDIUM**: Gates A, B, C are required. Do not proceed to next phase without explicit user confirmation.
@@ -0,0 +1,49 @@
1
+ # Hardening Lane — From Exploration to Reliable Execution
2
+
3
+ > Use when input is vague, exploratory, or "vibe-style" and needs to be converted into a spec pack before implementation.
4
+
5
+ ## Two lanes
6
+
7
+ ### Lane 1 — Exploration / Vibe
8
+ - Goal: discover value, direction, and feasibility
9
+ - Tolerates: back-and-forth, open questions, incomplete scope
10
+ - Good for: MVPs, ideas, experiments, early discovery
11
+ - Output: notes, rough PRD, conversation log
12
+
13
+ ### Lane 2 — Spec Hardening
14
+ - Goal: convert exploration into reliable, maintainable execution
15
+ - Requires: requirements, design decisions, checkpoints, and tests
16
+ - Good for: code that will survive, grow, and be maintained
17
+ - Output: spec pack (PRD + requirements + architecture + implementation plan)
18
+
19
+ **AIOSON's identity is Lane 2.** It does not compete with vibe coding — it is the hardening layer that converts exploration into software that works in production.
20
+
21
+ ## When to harden
22
+
23
+ Harden before coding when ANY of these are true:
24
+ - The feature involves new entities or database changes
25
+ - The feature integrates with an external service
26
+ - Multiple user types are involved
27
+ - The behavior in error or edge cases matters for production quality
28
+ - Another developer (or AI session) will need to continue this work later
29
+
30
+ Skip hardening (go direct from PRD to @dev) ONLY when:
31
+ - Classification is confirmed MICRO
32
+ - No new entities, no integrations, pure UI/CRUD
33
+ - The full spec can fit in one session context without losing state
34
+
35
+ ## Signals that input is in vibe mode (not yet hardenable)
36
+
37
+ - "I want something like X" without defining what X actually does
38
+ - Requirements expressed as UI descriptions ("there should be a button that does Y")
39
+ - No mention of what happens when things go wrong
40
+ - No mention of who can do what (permissions, roles)
41
+ - Scope keeps expanding during the conversation
42
+
43
+ ## What to do when input is in vibe mode
44
+
45
+ Do not start implementation. Instead:
46
+ 1. Acknowledge: "This is still in exploration mode — let's harden it before coding."
47
+ 2. Route to @product if no PRD exists yet.
48
+ 3. Route to @sheldon if PRD exists but has gaps.
49
+ 4. Only proceed to @analyst / @architect / @dev after Gate A is passed.
@@ -0,0 +1,66 @@
1
+ # Maintenance and State — Writing Useful Checkpoints
2
+
3
+ > Use when writing or reading `spec*.md`, updating checkpoints, or retaking work after a session break.
4
+
5
+ ## The purpose of spec-{slug}.md
6
+
7
+ It is not just an implementation log. It is the **living memory** of a feature across all sessions, agents, and tools.
8
+
9
+ A well-written `spec-{slug}.md` allows:
10
+ - @deyvin to resume from the last checkpoint without re-reading the entire spec pack
11
+ - @qa to know what was decided and what was deferred
12
+ - A future developer (or AI) to understand *why* the code was written the way it was
13
+
14
+ A poorly written `spec-{slug}.md` forces every new session to rediscover what was already decided.
15
+
16
+ ## What to write in phase_gates
17
+
18
+ ```yaml
19
+ phase_gates:
20
+ requirements: approved # requirements are locked — no new scope without PRD change
21
+ design: approved # architecture is locked — no structural changes without @architect
22
+ plan: approved # execution sequence is locked — @dev follows plan, not instinct
23
+ ```
24
+
25
+ Never mark a gate as `approved` if there are unresolved items in that phase.
26
+
27
+ ## What to write in last_checkpoint
28
+
29
+ Format: `{phase-name}: {what was completed} — {what is next}`
30
+
31
+ Examples:
32
+ - `migration: users table created and seeded — next: implement CreateUser action`
33
+ - `auth: login and register complete, password reset pending — next: implement ResetPassword action`
34
+ - `api: GET /products and POST /products done — next: implement PATCH /products/{id}`
35
+
36
+ This one line is what @deyvin reads first. Make it actionable.
37
+
38
+ ## What to write in pending_review
39
+
40
+ List items that need human verification before the next phase begins:
41
+
42
+ ```yaml
43
+ pending_review:
44
+ - "Confirm: should password reset expire after 1h or 24h? (decision was deferred)"
45
+ - "Review: CreateUser action sends email — confirm SMTP config in staging before testing"
46
+ ```
47
+
48
+ ## What to write in Key decisions
49
+
50
+ Format: `[ISO-date] [decision] — [reason this decision reduces future debug or maintenance cost]`
51
+
52
+ Bad example:
53
+ - `2026-03-28 Used soft deletes`
54
+
55
+ Good example:
56
+ - `2026-03-28 Used soft deletes on users table — reason: billing history must remain intact after account deletion; hard delete would orphan invoices`
57
+
58
+ The reason is what makes the decision useful in 6 months.
59
+
60
+ ## How @deyvin should use this file
61
+
62
+ 1. Read `phase_gates` first — know which phases are locked
63
+ 2. Read `last_checkpoint` — start from there, not from the beginning
64
+ 3. Read `pending_review` — these may need user input before proceeding
65
+ 4. Read `Key decisions` only if the next task involves re-touching those areas
66
+ 5. Do NOT re-read the full spec pack unless `last_checkpoint` is null or unclear
@@ -0,0 +1,75 @@
1
+ # UI Language — AIOSON Visual Standards
2
+
3
+ > Carregue quando um agente precisa apresentar opções, status ou checkpoints ao usuário.
4
+
5
+ ## Status symbols
6
+ ✓ completo / aprovado
7
+ ✗ falhou / bloqueado
8
+ ◆ em progresso
9
+ ○ pendente
10
+ ⚠ atenção necessária
11
+ ⚡ auto-aprovado
12
+
13
+ ## Stage banner
14
+ Usar ao iniciar fase principal:
15
+ ```
16
+ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
17
+ AIOSON ► @{AGENT} — {FASE}
18
+ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
19
+ ```
20
+
21
+ ## Checkpoint verify (confirmação visual)
22
+ Usar: após implementação que o usuário precisa ver
23
+ ```
24
+ ┌─────────────────────────────────────────────┐
25
+ │ ✓ VERIFICAR: {título} │
26
+ │ {instrução específica} │
27
+ │ Confirmar? [s/n] │
28
+ └─────────────────────────────────────────────┘
29
+ ```
30
+
31
+ ## Checkpoint decision (AskUserQuestion — radio)
32
+ Usar: quando há bifurcação com outcomes diferentes
33
+ → AskUserQuestion com multiSelect: false, 2-4 opções
34
+
35
+ ```
36
+ ┌─────────────────────────────────────────────┐
37
+ │ ◆ DECISÃO NECESSÁRIA │
38
+ │ │
39
+ │ {contexto da decisão} │
40
+ │ │
41
+ │ 1. {opção A} — {consequências} │
42
+ │ 2. {opção B} — {consequências} │
43
+ │ │
44
+ │ Escolha [1/2]: │
45
+ └─────────────────────────────────────────────┘
46
+ ```
47
+
48
+ ## Checkpoint action (passo manual)
49
+ Usar: apenas para passos que o agente literalmente não pode executar
50
+ ```
51
+ ┌─────────────────────────────────────────────┐
52
+ │ ⚠ AÇÃO MANUAL NECESSÁRIA │
53
+ │ │
54
+ │ {instrução específica} │
55
+ │ {onde executar} │
56
+ │ │
57
+ │ Avise quando estiver pronto. │
58
+ └─────────────────────────────────────────────┘
59
+ ```
60
+
61
+ ## Checkpoint multi-select (AskUserQuestion — checkbox)
62
+ Usar: seleção múltipla (skills, requirements, itens de sprint)
63
+ → AskUserQuestion com multiSelect: true
64
+
65
+ ## Progress bar
66
+ Usar para fases longas com steps definidos:
67
+ ```
68
+ Progresso: ████████░░ 80% (4/5 steps)
69
+ ```
70
+
71
+ ## Regras
72
+ - Header máximo 12 caracteres
73
+ - Máximo 4 opções em radio; máximo 8 em checkbox
74
+ - Incluir opção "Nenhuma/Pular" em checkbox quando pertinente
75
+ - Não usar checkbox para decisões que mudam arquitetura (use radio)
@@ -0,0 +1,144 @@
1
+ ---
2
+ name: design-hybrid-forge
3
+ description: Process skill that creates project-local hybrid design skills by fusing exactly two primary AIOSON design skills, with optional limited modifiers. When activated, guides you through pair selection, identity synthesis, crossover spec, skill generation, preview creation, metadata, and optional promotion.
4
+ activation: |
5
+ You are now running the design-hybrid-forge process. Begin by asking the user for the names of the two primary AIOSON design skills they want to combine, then ask whether they want optional modifiers. Use up to two modifiers by default, or up to three only when the active variation preset or the user explicitly enables advanced mode. Follow the phases described in this skill.
6
+ ---
7
+
8
+ # Skill: design-hybrid-forge
9
+
10
+ > Process skill. Creates new hybrid design skills by fusing DNA from two existing AIOSON design skills.
11
+ > Load this file first. Then load only the `references/` file you need for your current phase.
12
+
13
+ ## What this skill produces
14
+
15
+ A complete, production-quality hybrid design skill package for the current project:
16
+
17
+ ```
18
+ .aioson/installed-skills/{hybrid-name}/
19
+ SKILL.md
20
+ .skill-meta.json
21
+ references/
22
+ art-direction.md
23
+ design-tokens.md
24
+ components.md
25
+ patterns.md
26
+ dashboards.md
27
+ websites.md
28
+ motion.md
29
+ previews/
30
+ {hybrid-name}.html
31
+ {hybrid-name}-website.html
32
+ ```
33
+
34
+ If tool-native skill folders exist, the finished package may also be mirrored to:
35
+
36
+ ```
37
+ .claude/skills/{hybrid-name}/
38
+ .cursor/skills/{hybrid-name}/
39
+ .windsurf/skills/{hybrid-name}/
40
+ ```
41
+
42
+ ## When to load
43
+
44
+ Load this skill when:
45
+ - You want to create a new hybrid design skill from two base skills
46
+ - You are planning this week's hybrid (pair selection phase)
47
+ - You are executing the crossover (generation phase)
48
+ - You need to validate a hybrid before shipping
49
+ - You want an anti-sameness variation overlay for a more extravagant, classic, animated, or CSS-advanced result
50
+
51
+ Do NOT load this skill to apply a design skill to a project. Use the target hybrid's own SKILL.md for that.
52
+
53
+ ## Output modes
54
+
55
+ ### 1. Project-local installed skill (default)
56
+
57
+ Use this mode in normal AIOSON projects.
58
+
59
+ - Writes the new hybrid to `.aioson/installed-skills/{hybrid-name}/`
60
+ - Keeps the skill versioned with the project
61
+ - Generates local previews inside the skill folder
62
+ - Records author / generator metadata in `.skill-meta.json`
63
+ - Does not touch AIOSON core galleries or first-party skill folders
64
+
65
+ ### 2. Core promotion (optional)
66
+
67
+ Use this mode only when the user explicitly wants to propose the hybrid back to AIOSON core or marketplace curation.
68
+
69
+ - Starts from an already-generated project-local skill
70
+ - Prepares promotion artifacts for `.aioson/skills/design/{hybrid-name}/`
71
+ - Updates gallery / registry files only in the core repo
72
+ - Must stay a separate step from project-local generation
73
+
74
+ ## Process overview
75
+
76
+ | Phase | What happens | Produces |
77
+ |---|---|---|
78
+ | **1. Pair Selection** | Choose 2 primary skills with creative tension | Chosen pair + rationale |
79
+ | **2. Identity Synthesis** | Name, 3 pillars, accent fusion, substrate | Hybrid identity doc (in conversation) |
80
+ | **3. Crossover Spec** | Define DNA from A, DNA from B, optional modifiers, new elements | Crossover map (in conversation) |
81
+ | **4. Skill Generation** | Write SKILL.md + 7 reference files + metadata | Skill package |
82
+ | **5. Preview Generation** | Write dashboard HTML + landing HTML inside the skill package | 2 `.html` files |
83
+ | **6. Distribution** | Register in `AGENTS.md` and mirror to tool-native skill folders if present | Tool-ready copies |
84
+ | **7. Optional Promotion** | Prepare for core/gallery/marketplace only if requested | Promotion bundle |
85
+
86
+ Each phase must complete before the next begins. Do not skip phase 2 and 3 — they are what makes the hybrid coherent rather than a random blend.
87
+
88
+ ## Input contract
89
+
90
+ ```
91
+ primary_a: {skill-name} # e.g. "cognitive-core-ui"
92
+ primary_b: {skill-name} # e.g. "glassmorphism-ui"
93
+ modifiers: {optional 0..2} # e.g. ["bold-editorial-ui"]
94
+ variation_overlay: {optional} # selected via conversation or `aioson design-hybrid:options`
95
+ modifier_policy: {optional} # "up_to_2_modifiers" by default, "up_to_3_modifiers" in advanced mode
96
+ name_suggestion: {optional} # e.g. "aurora-command-ui" or leave blank
97
+ target_domain: {optional} # e.g. "SOC platform" — narrows expression modes
98
+ author_name: {optional} # e.g. "Jaime" or "ACME Design Team"
99
+ generator_model: {optional} # fill if the runtime/tool exposes it
100
+ ```
101
+
102
+ ## Output contract
103
+
104
+ The hybrid must satisfy ALL of the following:
105
+ - 8 files (SKILL.md + 7 references) with full content — no placeholders
106
+ - 1 metadata file: `.skill-meta.json`
107
+ - 2 HTML previews: dashboard (operational) + landing page (marketing)
108
+ - A name that is original and not a concatenation of the parent names
109
+ - A new accent that is NOT identical to either parent's accent
110
+ - A substrate rule that is clearly stated and non-negotiable
111
+ - At minimum 5 expression modes in art-direction.md
112
+ - At minimum 20 components in components.md
113
+ - Exactly 2 primary parents
114
+ - At most 2 optional modifiers with limited scope by default
115
+ - Up to 3 modifiers only when advanced mode is explicitly enabled
116
+ - A single coherent visual system that can be selected as one `design_skill`
117
+ - Metadata that records authorship and generator/model details when available
118
+ - When a variation overlay exists, the generated skill and previews must visibly express it
119
+ - When a temporary variation preset exists, it must be archived or removed from active context after successful generation
120
+
121
+ ## References available
122
+
123
+ | File | Load when |
124
+ |---|---|
125
+ | `references/pair-compatibility.md` | Choosing which two skills to combine this week |
126
+ | `references/crossover-protocol.md` | Running phases 2 and 3 (identity + crossover spec) |
127
+ | `references/variation-library.md` | Choosing a variation overlay or anti-sameness directions |
128
+ | `references/output-contract.md` | Running phases 4 and 5 (file generation) |
129
+ | `references/naming-registry.md` | Naming the hybrid and checking for conflicts |
130
+ | `references/quality-gates.md` | Validating the hybrid before shipping (distribution / promotion gates) |
131
+
132
+ ## Non-negotiable rules
133
+
134
+ 1. Exactly 2 primary parents are required. Never generate a hybrid with 3 or 4 co-equal parents.
135
+ 2. Optional modifiers are capped at 2 by default. A 3rd modifier is allowed only in advanced mode and still cannot own substrate or structure.
136
+ 3. The hybrid must have its own identity — not "A with B colors" but a third thing.
137
+ 4. The crossover spec must be explicit: what comes from each primary parent, what modifiers influence, and what is new.
138
+ 5. The accent must be a genuine fusion — not parent A's accent, not parent B's. A new value or gradient pair.
139
+ 6. The substrate rule is always the first decision. One primary parent wins the background model.
140
+ 7. The hybrid's SKILL.md must explicitly name its parents in a `## Hybrid DNA` section.
141
+ 8. Never combine two primary skills from the same family (e.g. cognitive-core + premium-command = too similar).
142
+ 9. Every hybrid ships with both previews and a `.skill-meta.json`. No preview = not done.
143
+ 10. Project-local generation goes to `.aioson/installed-skills/` by default. Promotion to core is a separate, explicit step.
144
+ 11. `design-hybrid:options` creates a temporary preset in `.aioson/context/design-variation-preset.md`; after successful generation, archive or remove the active preset and preserve the history snapshot.