create-vasvibe 2.4.0 → 2.5.0

package/template/schemas/workspace-registry.json

@@ -1,95 +1,94 @@
-{
-  "version": 1,
-  "description": "Curated registry for the Toolsmith agent. Maps project needs (tech stack / phase) to recommended skills + MCP servers, and encodes how to write MCP config for each AI tool platform. Toolsmith reads this as the DEFAULT catalog, then may augment via requirements + find-skills.",
-
-  "stackMappings": {
-    "_comment": "tech-stack / capability keyword -> recommended skills + mcp ids. Match loosely (case-insensitive substring) against project_overview tech stack & key features.",
-    "ui|frontend|react|next|nextjs|vue|svelte|tailwind|shadcn": { "skills": ["ui-ux-pro-max"], "mcp": ["shadcn"] },
-    "prototype|mockup|wireframe|demo": { "skills": ["ui-ux-pro-max", "theme-factory"], "mcp": ["shadcn"] },
-    "postgres|postgresql|prisma|sql|database": { "skills": [], "mcp": ["postgres"] },
-    "e2e|test|testing|playwright|qa": { "skills": [], "mcp": ["playwright"] },
-    "security|auth|oauth|pentest|owasp": { "skills": ["penetration-testing"], "mcp": [] },
-    "*": { "skills": ["find-skills"], "mcp": ["filesystem", "git", "fetch"] }
-  },
-
-  "mcpCatalog": {
-    "_comment": "Common, safe MCP servers. command/args are indicative starting points — Toolsmith MUST verify the current invocation against the live MCP registry (or each server's README) before applying, since package names drift. `runtime` notes whether it ships via npx (Node) or uvx (Python).",
-    "filesystem": {
-      "description": "Read/write files within allowed roots — grounds the agent in the workspace.",
-      "runtime": "npx",
-      "command": "npx",
-      "args": ["-y", "@modelcontextprotocol/server-filesystem", "."],
-      "env": {}
-    },
-    "git": {
-      "description": "Inspect git history, diffs, and status — context for code-aware agents.",
-      "runtime": "uvx",
-      "command": "uvx",
-      "args": ["mcp-server-git"],
-      "env": {}
-    },
-    "fetch": {
-      "description": "Fetch and convert web content (docs, references) into agent-readable text.",
-      "runtime": "uvx",
-      "command": "uvx",
-      "args": ["mcp-server-fetch"],
-      "env": {}
-    },
-    "postgres": {
-      "description": "Query a Postgres database (schema introspection, read queries). Set POSTGRES_URL.",
-      "runtime": "npx",
-      "command": "npx",
-      "args": ["-y", "@modelcontextprotocol/server-postgres", "${POSTGRES_URL}"],
-      "env": { "POSTGRES_URL": "${POSTGRES_URL}" }
-    },
-    "playwright": {
-      "description": "Drive a real browser for E2E testing & verification (Tester agent).",
-      "runtime": "npx",
-      "command": "npx",
-      "args": ["-y", "@playwright/mcp@latest"],
-      "env": {}
-    },
-    "shadcn": {
-      "description": "Search & pull shadcn/ui component examples — pairs with ui-ux-pro-max skill. Verify current invocation (e.g. `npx shadcn@latest mcp`) before applying.",
-      "runtime": "npx",
-      "command": "npx",
-      "args": ["-y", "shadcn@latest", "mcp"],
-      "env": {}
-    }
-  },
-
-  "platformTargets": {
-    "_comment": "How to write MCP config declaratively for each tool. `file` is project-relative. `topLevelKey` is the JSON key holding servers. `entryShape` documents the per-server object shape; substitute command/args/env from mcpCatalog.",
-    "claude-code": {
-      "detect": [".claude/", ".mcp.json"],
-      "file": ".mcp.json",
-      "topLevelKey": "mcpServers",
-      "entryShape": { "command": "<command>", "args": ["<args>"], "env": {} },
-      "skillsDir": ".claude/skills/"
-    },
-    "opencode": {
-      "detect": [".opencode/", "opencode.json"],
-      "file": "opencode.json",
-      "topLevelKey": "mcp",
-      "entryShape": { "type": "local", "command": ["<command>", "<args>"], "enabled": true, "environment": {} },
-      "skillsDir": ".opencode/skills/",
-      "note": "OpenCode merges command+args into a single `command` array and uses `environment` (not `env`). Include \"$schema\": \"https://opencode.ai/config.json\" at the root."
-    },
-    "antigravity": {
-      "detect": [".agents/"],
-      "file": ".agents/mcp.json",
-      "topLevelKey": "mcpServers",
-      "entryShape": { "command": "<command>", "args": ["<args>"], "env": {} },
-      "skillsDir": ".agents/skills/",
-      "note": "Antigravity uses an mcpServers-shaped config. Confirm the exact path your Antigravity build reads (workspace vs global) with the human; .agents/mcp.json is the project-local default."
-    },
-    "github-copilot": {
-      "detect": [".github/prompts/", ".vscode/"],
-      "file": ".vscode/mcp.json",
-      "topLevelKey": "servers",
-      "entryShape": { "command": "<command>", "args": ["<args>"], "env": {} },
-      "skillsDir": null,
-      "note": "Copilot/VS Code uses the `servers` key (not `mcpServers`). Copilot has no skills dir — skills are Claude/OpenCode/Antigravity only."
-    }
-  }
-}
+{
+  "version": 1,
+  "description": "Curated registry for the Toolsmith agent. Maps project needs (tech stack / phase) to recommended skills + MCP servers, and encodes how to write MCP config for each AI tool platform. Toolsmith reads this as the DEFAULT catalog, then may augment via requirements + find-skills.",
+
+  "stackMappings": {
+    "_comment": "tech-stack / capability keyword -> recommended skills + mcp ids. Match loosely (case-insensitive substring) against project_overview tech stack & key features.",
+    "ui|frontend|react|next|nextjs|vue|svelte|tailwind|shadcn": { "skills": ["ui-ux-pro-max"], "mcp": ["shadcn"] },
+    "postgres|postgresql|prisma|sql|database": { "skills": [], "mcp": ["postgres"] },
+    "e2e|test|testing|playwright|qa": { "skills": [], "mcp": ["playwright"] },
+    "security|auth|oauth|pentest|owasp": { "skills": ["penetration-testing"], "mcp": [] },
+    "*": { "skills": ["find-skills"], "mcp": ["filesystem", "git", "fetch"] }
+  },
+
+  "mcpCatalog": {
+    "_comment": "Common, safe MCP servers. command/args are indicative starting points — Toolsmith MUST verify the current invocation against the live MCP registry (or each server's README) before applying, since package names drift. `runtime` notes whether it ships via npx (Node) or uvx (Python).",
+    "filesystem": {
+      "description": "Read/write files within allowed roots — grounds the agent in the workspace.",
+      "runtime": "npx",
+      "command": "npx",
+      "args": ["-y", "@modelcontextprotocol/server-filesystem", "."],
+      "env": {}
+    },
+    "git": {
+      "description": "Inspect git history, diffs, and status — context for code-aware agents.",
+      "runtime": "uvx",
+      "command": "uvx",
+      "args": ["mcp-server-git"],
+      "env": {}
+    },
+    "fetch": {
+      "description": "Fetch and convert web content (docs, references) into agent-readable text.",
+      "runtime": "uvx",
+      "command": "uvx",
+      "args": ["mcp-server-fetch"],
+      "env": {}
+    },
+    "postgres": {
+      "description": "Query a Postgres database (schema introspection, read queries). Set POSTGRES_URL.",
+      "runtime": "npx",
+      "command": "npx",
+      "args": ["-y", "@modelcontextprotocol/server-postgres", "${POSTGRES_URL}"],
+      "env": { "POSTGRES_URL": "${POSTGRES_URL}" }
+    },
+    "playwright": {
+      "description": "Drive a real browser for E2E testing & verification (Tester agent).",
+      "runtime": "npx",
+      "command": "npx",
+      "args": ["-y", "@playwright/mcp@latest"],
+      "env": {}
+    },
+    "shadcn": {
+      "description": "Search & pull shadcn/ui component examples — pairs with ui-ux-pro-max skill. Verify current invocation (e.g. `npx shadcn@latest mcp`) before applying.",
+      "runtime": "npx",
+      "command": "npx",
+      "args": ["-y", "shadcn@latest", "mcp"],
+      "env": {}
+    }
+  },
+
+  "platformTargets": {
+    "_comment": "How to write MCP config declaratively for each tool. `file` is project-relative. `topLevelKey` is the JSON key holding servers. `entryShape` documents the per-server object shape; substitute command/args/env from mcpCatalog.",
+    "claude-code": {
+      "detect": [".claude/", ".mcp.json"],
+      "file": ".mcp.json",
+      "topLevelKey": "mcpServers",
+      "entryShape": { "command": "<command>", "args": ["<args>"], "env": {} },
+      "skillsDir": ".claude/skills/"
+    },
+    "opencode": {
+      "detect": [".opencode/", "opencode.json"],
+      "file": "opencode.json",
+      "topLevelKey": "mcp",
+      "entryShape": { "type": "local", "command": ["<command>", "<args>"], "enabled": true, "environment": {} },
+      "skillsDir": ".opencode/skills/",
+      "note": "OpenCode merges command+args into a single `command` array and uses `environment` (not `env`). Include \"$schema\": \"https://opencode.ai/config.json\" at the root."
+    },
+    "antigravity": {
+      "detect": [".agents/"],
+      "file": ".agents/mcp.json",
+      "topLevelKey": "mcpServers",
+      "entryShape": { "command": "<command>", "args": ["<args>"], "env": {} },
+      "skillsDir": ".agents/skills/",
+      "note": "Antigravity uses an mcpServers-shaped config. Confirm the exact path your Antigravity build reads (workspace vs global) with the human; .agents/mcp.json is the project-local default."
+    },
+    "github-copilot": {
+      "detect": [".github/prompts/", ".vscode/"],
+      "file": ".vscode/mcp.json",
+      "topLevelKey": "servers",
+      "entryShape": { "command": "<command>", "args": ["<args>"], "env": {} },
+      "skillsDir": null,
+      "note": "Copilot/VS Code uses the `servers` key (not `mcpServers`). Copilot has no skills dir — skills are Claude/OpenCode/Antigravity only."
+    }
+  }
+}

package/template/skills-lock.json

@@ -1,11 +1,11 @@
-{
-  "version": 1,
-  "skills": {
-    "find-skills": {
-      "source": "vercel-labs/skills",
-      "sourceType": "github",
-      "skillPath": "skills/find-skills/SKILL.md",
-      "computedHash": "9e1c8b3103f92fa8092568a44fe64858de7c5c9dc65ce4bea8f168080e889cfd"
-    }
-  }
-}
+{
+  "version": 1,
+  "skills": {
+    "find-skills": {
+      "source": "vercel-labs/skills",
+      "sourceType": "github",
+      "skillPath": "skills/find-skills/SKILL.md",
+      "computedHash": "9e1c8b3103f92fa8092568a44fe64858de7c5c9dc65ce4bea8f168080e889cfd"
+    }
+  }
+}

package/template/specifications/.gitkeep

@@ -1,3 +1 @@
-# Keeps the specifications/ folder in the repository.
-# Specification files generated by the Analyst agent (e.g. 000_spec_environment_setup.md)
-# ARE tracked and should be committed — they are the source of truth for development.
+# Place specification documents here. The contents of this folder are gitignored by default.

package/template/state/knowledge_base/.gitkeep

@@ -1 +1 @@
-# Keep this directory for storing agent bug-fixes patterns and learnings.
+# Keep this directory for storing agent bug-fixes patterns and learnings.

package/template/state/knowledge_base/requirements/.gitkeep

@@ -1 +1 @@
-# Keep this folder tracked by git
+# Keep this folder tracked by git

package/template/tests/.gitkeep

@@ -1 +1 @@
-# Place test files here. The contents of this folder are gitignored by default.
+# Place test files here. The contents of this folder are gitignored by default.

package/template/.agents/agents/logger/agent.json

@@ -1,43 +0,0 @@
-{
-  "name": "logger",
-  "description": "Change Logger. Records session changes into the appropriate task log.",
-  "hidden": true,
-  "config": {
-    "customAgent": {
-      "systemPromptSections": [
-        {
-          "title": "Agent System Instructions",
-          "content": "You are the logger Agent. Read and follow ALL instructions in agent/workflows/logger.md as your complete system prompt. That file is your source of truth."
-        }
-      ],
-      "toolNames": [
-        "send_message",
-        "find_by_name",
-        "grep_search",
-        "view_file",
-        "list_dir",
-        "read_url_content",
-        "search_web",
-        "schedule",
-        "multi_replace_file_content",
-        "replace_file_content",
-        "write_to_file",
-        "manage_task",
-        "define_subagent",
-        "invoke_subagent",
-        "manage_subagents"
-      ],
-      "systemPromptConfig": {
-        "includeSections": [
-          "user_information",
-          "mcp_servers",
-          "skills",
-          "subagent_reminder",
-          "messaging",
-          "artifacts",
-          "user_rules"
-        ]
-      }
-    }
-  }
-}

package/template/.agents/workflows/build-prototype.md

@@ -1,32 +0,0 @@
----
-description: Bangun prototipe UI/UX cepat (frontend-only, data mock) untuk presentasi/menyepakati desain dengan klien — tanpa backend.
----
-
-Jalankan **Build Prototype** — prototipe UI/UX cepat untuk: **$ARGUMENTS**
-
-> **Tujuan:** menampilkan tampilan aplikasi secepatnya untuk presentasi ke klien / menyepakati UI-UX — BUKAN produk jadi. **Frontend saja: tanpa backend, tanpa API nyata, tanpa database.** Data di-mock/fixture. Ini artefak demo (throwaway) yang bisa diganti; jangan diperlakukan sebagai implementasi final.
-
-**DI LUAR RUANG LINGKUP (jangan dikerjakan):** business logic, persistensi data, integrasi API nyata, autentikasi nyata, dan test fungsional. Untuk implementasi sungguhan gunakan `/build-feature`.
-
-Langkah:
-0. **Provisioning (Toolsmith)** → pastikan skill + MCP untuk prototyping terpasang di tool aktif: skill `ui-ux-pro-max` + `theme-factory`, MCP `shadcn`. Sumber acuan `state/workspace-manifest.json` + `schemas/workspace-registry.json` (mapping `prototype`). Skip cepat jika sudah ter-apply.
-1. **Brief Prototipe (INTERAKTIF — JANGAN delegasikan ke subagent)** → klarifikasi cepat dengan human di thread utama: layar/alur apa yang mau ditunjukkan, urutan navigasi antar-layar, dan referensi visual (brand/warna/mood) jika ada. Ringkas — bukan wawancara kebutuhan penuh.
-   - *(Opsional, pra-kode)* Untuk menyepakati layout/hierarki lebih dulu tanpa menulis kode, boleh pakai mockup inline (skill/tool `visualize`) atau aset visual (Canva) **bila tool aktif menyediakannya** — ini opsional, bukan dependensi.
-2. **Arah Visual (pilih satu):**
-   - `state/knowledge_base/design-system/` **sudah ada** → pakai sebagai acuan, lewati langkah ini.
-   - Belum ada & butuh **kilat** → gunakan skill **`theme-factory`** untuk menetapkan tema global (palette + font) instan.
-   - Belum ada & mau **lebih rapi** → delegasikan ke **UX Designer** (mode `fast`) untuk palette + tipografi + token dasar via `ui-ux-pro-max`.
-3. **Frontend Engineer (mode PROTOTYPE)** → bangun UI di `codes/`:
-   - Frontend-only, **tanpa panggilan backend** — gunakan data **mock/fixture statis**.
-   - **Tooling (komplementer):** `theme-factory` untuk tema global (bila belum ada design-system) · `ui-ux-pro-max` untuk pola, layout & kualitas komponen · MCP `shadcn` untuk menarik komponen jadi bila tersedia. Ikuti `design-system/` bila ada.
-   - Bangun semua layar yang disepakati + **navigasi antar-layar yang clickable** agar bisa didemokan.
-   - Tampilkan variasi state (empty/loading/success) sebagai tampilan statis bila relevan — tanpa logika nyata.
-   - **Lewati unit test dan integrasi API.**
-4. **CHECKPOINT — Demo ke klien:** **Jalankan dev server** prototipe dan tampilkan ke human/klien; kumpulkan feedback UI/UX. **Iterasi langkah 3** sampai desain disepakati.
-5. **Output:** prototipe frontend yang bisa dijalankan untuk presentasi + catatan kesepakatan UI/UX (masukan untuk `design-system/` atau spec berikutnya).
-
-**Aturan orkestrasi (WAJIB):**
-- Kamu adalah **Orchestrator** di thread utama. Delegasikan tiap langkah ke subagent (Task tool Claude Code / agent invocation OpenCode).
-- **BERHENTI di CHECKPOINT** untuk feedback klien sebelum iterasi atau menyelesaikan.
-- Ini alur **di luar 4 fase** (throwaway/demo) — JANGAN buat spec, API Contract, atau backend. Jika klien setuju dan ingin lanjut ke produk, arahkan ke `/plan-feature` (fitur) atau `/plan-project` (proyek baru).
-- Definisi kanonik pipeline: `agent/workflows/orchestrator.md`. Model fase: `agent/workflows/_shared/phases.md`.

package/template/.agents/workflows/explorer.md

@@ -1,78 +0,0 @@
-# Explorer Agent — Workflow Detail
-
-**Role:** Senior Software Architect & Code Analyst  
-**Trigger:** `/explore-codebase` atau dipanggil Orchestrator sebelum `/build-feature` pada project existing.  
-**Input:** Codebase yang sudah ada di folder `codes/`  
-**Output:** `project_overview.md`, `state/knowledge_base/codebase/inventory.md`, `state/context.json`
-
----
-
-## Kapan Menggunakan Explorer?
-
-Gunakan Explorer (bukan Discovery + Initiator) ketika:
-- Kamu **bergabung ke project yang sudah berjalan** dan perlu memahami kondisinya sebelum mengerjakan fitur baru.
-- Ada **handover codebase** dari tim lain dan tidak ada dokumentasi overview.
-- Kamu ingin **memverifikasi** apakah `project_overview.md` yang ada masih akurat dengan kode terkini.
-- Sebelum menjalankan `/build-feature` atau `/start-fix` pada project existing yang belum punya state.
-
----
-
-## Alur Kerja
-
-```
-codes/ exists?
-     │
-     ├─ YES → Explorer baca kode → tulis project_overview.md + inventory + context.json
-     │              │
-     │              └─ Human review → lanjut ke /build-feature atau /start-fix
-     │
-     └─ NO  → Arahkan ke /plan-project (project baru, mulai dari Discovery)
-```
-
----
-
-## Dokumen yang Dihasilkan
-
-| Dokumen | Path | Deskripsi |
-|---------|------|-----------|
-| Project Overview | `project_overview.md` | Tech stack, fitur, design system — disintesis dari kode |
-| Codebase Inventory | `state/knowledge_base/codebase/inventory.md` | Daftar API endpoints, data models, file penting |
-| Context State | `state/context.json` | Status project saat ini untuk kontinuitas antar-session |
-
----
-
-## Panduan Membaca Kode per Stack
-
-### Node.js / TypeScript (Next.js, Express, NestJS)
-- Root: `package.json`, `tsconfig.json`, `.env.example`
-- Routes: `src/routes/`, `src/app/api/`, `pages/api/`, `src/controllers/`
-- Models: `prisma/schema.prisma`, `src/models/`, `src/entities/`
-- Config: `src/config/`, `next.config.js`, `nest-cli.json`
-
-### Python (FastAPI, Django, Flask)
-- Root: `pyproject.toml`, `requirements.txt`, `setup.py`
-- Routes: `app/routers/`, `app/views.py`, `urls.py`
-- Models: `app/models.py`, `alembic/versions/`
-- Config: `app/config.py`, `settings.py`
-
-### Go
-- Root: `go.mod`, `go.sum`
-- Routes: `internal/handler/`, `internal/api/`, `cmd/`
-- Models: `internal/domain/`, `internal/model/`
-- DB: `internal/repository/`, migration files
-
-### Mobile (Flutter, React Native)
-- Root: `pubspec.yaml` atau `package.json`
-- Screens: `lib/screens/`, `src/screens/`
-- State: `lib/providers/`, `lib/bloc/`, `src/store/`
-- API: `lib/services/`, `src/api/`
-
----
-
-## Output Quality Rules
-
-1. **Jangan berasumsi** — setiap klaim HARUS dari file yang dibaca. Tandai dengan `[confirmed]`, `[inferred]`, atau `[unknown]`.
-2. **Inventarisasi endpoints lengkap** — jangan skip; ini akan dipakai Analyst untuk API Contract.
-3. **Tandai incomplete features** — gunakan `[~]` untuk partial, `[ ]` untuk planned/TODO.
-4. **Catat technical debt** — tulis di `notes` field `context.json` agar agent berikutnya tahu.
-5. **Jangan timpa state yang masih valid** — jika `state/context.json` sudah ada dan masih relevan, merge informasi baru ke dalamnya.

package/template/.agents/workflows/logger.md

@@ -1,117 +0,0 @@
-# Logger Agent — Workflow Detail
-
-**Role:** Technical Change Recorder & Analyst  
-**Trigger:** Dipanggil kapan saja oleh human — *"catat perubahan ini"*, *"log apa yang baru saja terjadi"*, atau cukup *"logger"*.  
-**Input:** Deskripsi perubahan dari human, dan/atau analisa sesi / kode secara mandiri  
-**Output:** Entry baru di `task/[TASK-ID]/dev_log.md` + baris baru di Status Log task detail file
-
----
-
-## Kapan Memanggil Logger?
-
-| Situasi | Cara panggil |
-|---------|--------------|
-| Baru selesai fix bug secara manual | *"Logger, catat: fix bug validasi di form login"* |
-| Selesai coding satu sesi, ingin rekap tercatat | *"Logger, analisa apa yang baru saja dikerjakan dan catat"* |
-| Mau pastikan enhancement tercatat sebelum tutup sesi | *"Logger"* (tanpa argumen — logger akan tanya / analisa sendiri) |
-| Ingin log lintas beberapa task | *"Logger, tadi ada 3 perubahan: ..."* |
-
----
-
-## Tipe Perubahan & Klasifikasinya
-
-```
-Sesuatu yang salah/rusak diperbaiki?
-    └─ YES → fix
-
-Kemampuan baru ditambahkan?
-    └─ YES → enhancement
-
-Kode diubah tanpa ubah perilaku?
-    └─ YES → refactor
-
-Config, dependency, atau dokumentasi?
-    └─ YES → chore
-```
-
-**Aturan tie-breaker:** Jika satu perubahan bisa fix DAN enhancement (mis. memperbaiki cara feature bekerja sambil menambah field baru), buat **dua entry terpisah** — satu fix, satu enhancement.
-
----
-
-## Struktur Dokumen yang Diisi
-
-### 1. Task Detail File — Status Log
-**Path:** `task/[TASK-ID]_[nama-task]/[TASK-ID]_[nama-task].md`
-
-Logger menambahkan baris baru di tabel Status Log yang sudah ada:
-```markdown
-| [timestamp] | logger | [tipe] | [deskripsi 1 baris] |
-```
-
-### 2. Dev Log File
-**Path:** `task/[TASK-ID]_[nama-task]/dev_log.md`
-
-Logger membuat file ini jika belum ada, atau menambah entry di Revision History:
-```markdown
-## [timestamp] — [tipe]: [judul]
-
-**Type:** fix / enhancement / refactor / chore
-**Changed by:** [human / agent]
-
-### What Changed
-[Narasi singkat]
-
-### Files Affected
-- `path/file` — apa yang berubah
-
-### Notes
-[Catatan teknis jika ada]
-```
-
-### 3. Fallback — Untracked Changes
-Jika tidak ada task yang cocok, tulis ke:
-**Path:** `task/untracked/[YYYY-MM-DD]_untracked_changes.md`
-
-```markdown
-# Untracked Changes — [YYYY-MM-DD]
-
-## [HH:MM] — [tipe]: [judul]
-[Deskripsi perubahan]
-**Suggested Task:** [nama task yang mungkin relevan, atau "belum ada task terdaftar"]
-```
-
----
-
-## Alur Keputusan Logger
-
-```
-Human panggil logger
-        │
-        ├─ Ada deskripsi eksplisit? ──→ Gunakan sebagai basis
-        │
-        ├─ Tidak ada info? ──────────→ Tanya 1-2 pertanyaan ringkas
-        │
-        └─ Minta analisa mandiri? ───→ Periksa file yang berubah / konteks sesi
-                    │
-                    ▼
-        Klasifikasi tipe perubahan
-                    │
-                    ▼
-        Cari task yang cocok di task_list.md
-                    │
-                    ├─ Ketemu → tulis log di task tersebut
-                    ├─ Lintas task → tulis di semua task relevan
-                    └─ Tidak ketemu → tulis di task/untracked/
-                    │
-                    ▼
-        Konfirmasi ringkasan ke human
-```
-
----
-
-## Aturan Kualitas Log
-
-1. **Deskripsi konkret** — bukan *"fix bug"* tapi *"fix: email tidak divalidasi saat domain kosong"*
-2. **Satu entry per perubahan logis** — 3 perubahan berbeda = 3 entry
-3. **Tidak mengarang** — jika tidak yakin apa yang berubah, tanya dulu
-4. **Timestamp real** — selalu gunakan waktu saat ini, bukan placeholder

package/template/.agents/workflows/plan-feature.md

@@ -1,30 +0,0 @@
----
-description: Perencanaan terfokus untuk SATU fitur — hasilkan spesifikasi lengkap (low-entropy) tanpa merencanakan ulang seluruh proyek.
----
-
-Jalankan **Perencanaan Fitur** (sub-set Fase 1, ruang lingkup satu fitur) untuk: **$ARGUMENTS**
-
-> Berbeda dengan `/plan-project` yang menghasilkan SELURUH blueprint proyek, `/plan-feature` mengasumsikan blueprint proyek sudah ada dan hanya menghasilkan **spesifikasi satu fitur** yang siap dikerjakan. Output akhirnya identik dengan yang diproduksi Analyst: file spec lengkap di `specifications/` sesuai `schemas/specification.template.md`.
-
-**PRECONDITION CHECK (langkah 0):**
-- Pastikan `project_overview.md` ada dan "Tech Stack" + "UI Guidelines" terisi.
-- Pastikan `specifications/000_spec_environment_setup.md` sudah ada.
-- **Jika salah satu belum ada:** proyek belum direncanakan. **BERHENTI** dan arahkan human menjalankan `/plan-project` dulu. Jangan menambal blueprint proyek di sini.
-
-Langkah:
-1. **Feature Discovery (INTERAKTIF — JANGAN delegasikan ke subagent)** → kamu sendiri di thread utama mewawancarai human, khusus untuk fitur ini: tujuan & nilai, aktor/role, ruang lingkup, **apa yang di luar ruang lingkup (Non-Goals)**, aturan bisnis, dan kriteria sukses. Ringkas, fokus fitur — bukan wawancara proyek penuh. Ikuti gaya tanya-jawab `agent/workflows/discovery.md` namun terbatas pada fitur.
-2. **GATE:** Human sign-off ruang lingkup & Non-Goals fitur.
-3. **Delta acuan (KONDISIONAL — hanya jika fitur menuntutnya):**
-   - Jika fitur butuh entitas/kolom data BARU → delegasikan ke **Data Architect** untuk menambah delta di `state/knowledge_base/data-model/` (bukan redesign).
-   - Jika fitur butuh komponen/pola UI BARU → delegasikan ke **UX Designer** untuk menambah delta di `state/knowledge_base/design-system/`.
-   - Jika `depth=deep` dan fitur menyentuh data sensitif → koordinasi **Security (Mode S)** untuk aturan yang relevan.
-   - Jika tidak ada kebutuhan baru, lewati langkah ini.
-4. **Analyst** → tulis spesifikasi fitur lengkap di `specifications/NNN_...md` (nomor urut berikutnya), mengikuti `schemas/specification.template.md` dan **Low-Entropy Gate** di `agent/workflows/analyst.md`. Terapkan atomicity (1 file = 1 user story; pecah bila perlu). **API Contract wajib final & deterministik.**
-5. **GATE — Spec disetujui:** Human set `Human Approval Status: approved`. API Contract final sebelum fitur boleh masuk `/build-feature`.
-
-**Aturan orkestrasi (WAJIB):**
-- Kamu adalah **Orchestrator** di thread utama. Delegasikan tiap langkah ke subagent yang disebut menggunakan Task tool (Claude Code) / agent invocation (OpenCode).
-- **BERHENTI di setiap GATE** dan minta persetujuan human secara eksplisit sebelum lanjut.
-- Definisi kanonik pipeline: baca `agent/workflows/orchestrator.md`. Model fase & kepemilikan dokumen: `agent/workflows/_shared/phases.md`.
-- Hormati `WORK_DEPTH` dari `project_overview.md` (override dengan `depth=` di argumen jika ada).
-- Jika muncul perubahan di tengah pipeline, ikuti `agent/workflows/_shared/change-management.md` (No Silent Changes).