wb-flow 1.0.0-r01

package/templates/commands/wbTrack/wbTrack_template.md

@@ -0,0 +1,555 @@
+# /wbTrack — Session Tracking (v2: Universal Daily Session)
+
+
+<!-- HELP_GATE_START -->
+## Help intercept (handle FIRST — before any other action)
+
+**If `$ARGUMENTS` contains `--help`, `-h`, or `--h`** (case-insensitive, anywhere in the args), DO NOT execute the command's normal procedure. Instead:
+
+1. Output the **HELP BLOCK** below verbatim (rendered as markdown).
+2. Stop. Do not perform any file reads, writes, or other tool calls.
+3. Do not generate any reports under `.wb/workflows/reports/`.
+
+Otherwise, ignore this section and proceed to the rest of the template.
+
+### HELP BLOCK — `/wbTrack`
+
+# /wbTrack: Practical Guide 🛠️ (Claude version)
+
+## Quick Start
+
+```bash
+/wbTrack packages/wb-core    # start tracking (create or join today's file)
+/wbTest wb-core               # creates report + §N *(Your Model — HH:MM)* in session
+/wbAudit wb-core              # creates report + §N *(Your Model — HH:MM)* in session
+/wbTrack --stop               # §STOP for your model (file stays open for others)
+```
+
+## The Rule of Two Outputs
+
+When tracking is ON, every command produces **two things**:
+
+| Output | Location | Purpose |
+|---|---|---|
+| **Report file** | `reports/<date>/<type>/<type>_<target>_<date>.md` | Structured data, scanned tomorrow |
+| **§N in session** | `walkthroughs/<date>/session_<target>_<date>.md` | Narrative with commentary (shared by all models) |
+
+When tracking is OFF, only the report is created.
+
+## Create vs Append
+
+| Situation | Behavior |
+|---|---|
+| First model today runs `/wbTrack` | **Creates** the file + writes §0 |
+| Later model runs `/wbTrack` same scope | **Appends** contributor entry to existing file |
+| Any model runs `/wb*` while tracking | **Appends** §N tagged with model name + time |
+
+## Scoping Cheat Sheet
+
+| You type | Session file lives at | §0 reads |
+|---|---|---|
+| `/wbTrack packages/wb-core` | `wb-core/.wb/workflows/walkthroughs/<date>/session_wb-core_<date>.md` | wb-core code + reports |
+| `/wbTrack` | `core2/.wb/workflows/walkthroughs/<date>/session_core2_<date>.md` | All packages |
+
+## Common Mistakes
+
+❌ Expecting a `<model>/` subfolder under `walkthroughs/` → there is none in v2
+❌ Expecting reports to go to `walkthroughs/` → they always go to `reports/`
+❌ Using `/wbTrack` for a single command → overkill, just run the command
+❌ Running `/wbTrack --stop --finalize` mid-day → extracts partial derivatives (wait until end of day)
+
+## When NOT to Track
+
+- Quick one-off `/wbTest` or `/wbGit`
+- Commands you've run dozens of times
+- When you don't need commentary (just the facts)
+
+> For deeper reading: [`docs_claude/commands/wbTrack/wbTrack_practical_claude.md`](../../docs/docs_claude/commands/wbTrack/wbTrack_practical_claude.md) (or the `_eli5_`, `_expert_`, `_examples_` siblings).
+
+<!-- FLAGS_TABLE_START -->
+## Flags & shortcuts
+
+Both forms are equivalent — pass either:
+
+| Long form | Shortcut |
+|---|---|
+| `--finalize` | `-f` |
+| `--scope` | `-s` |
+
+`-h` / `--help` / `--h` (any command) prints this help block instead of executing.
+## Self-correct mode (dual-mode invocation)
+
+```
+/wbTrack <scope_folder>           # normal mode — produce a fresh output file
+/wbTrack <previous_output_file>   # self-correct mode — verify & repair the file in place
+```
+
+When the first arg is an existing output file from a prior `/wbTrack` run (detected by its first H1 — see this template's **Detection** section), the command runs in **verify-and-repair** mode: gap-fills missing fields, normalizes links, ticks done/valid checkboxes whose reports exist, never rewrites authored content. See [`../_shared/output_conventions.md`](../_shared/output_conventions.md) §3.
+
+<!-- FLAGS_TABLE_END -->
+<!-- HELP_GATE_END -->
+
+<!-- FLAG_NORMALIZE_START -->
+## Flag normalization (apply BEFORE parsing args)
+
+Before processing `$ARGUMENTS`, normalize these short-form flags to their long equivalents:
+
+- `-f` → `--finalize`
+- `-s` → `--scope`
+
+The rest of this template documents only the long forms; the substitution above is the only place short forms are mentioned.
+<!-- FLAG_NORMALIZE_END -->
+
+
+
+> **Command #25** in the `/wb*` suite
+> **Purpose**: Toggle session tracking ON. Every `/wb*` command produces its normal report AND gets logged into a **universal daily session file** with strategic commentary.
+> **Companion**: `/wbTrack --stop` ends this model's tracking and writes a §STOP block.
+> **Read first:** [`../_shared/output_conventions.md`](../_shared/output_conventions.md) — every §N section appended to the session file MUST follow the relative-link rule (§1) and full-syntax-command rule (§2). Self-correct mode (§3) applies in a **limited** form: see the Detection block below — gap-fill only, never rewrites existing §N bodies.
+
+---
+
+## Convention Notes for Session Entries
+
+When writing each §N section to the session file:
+
+- **Relative links (§1):** Every `Files read`, `Files created/modified`, and `Report file` reference MUST be a relative markdown link computed from the session file's directory. Plain-text paths are forbidden.
+- **Full-syntax commands (§2):** The `# §N — \`/wbCommand target [flags]\`` heading and every entry in the `Recommended Next` table MUST use the full invocable form (e.g., `/wbTest packages/wb-core --scope=task-1`, not bare `/wbTest`).
+- **Limited self-correct on the session file itself:** Re-running `/wbTrack` on an existing track file (when the path is passed *as the argument*, not as the scope) triggers the Detection block below — verify-and-repair only. Re-running `/wbTrack` on the *scope folder* still appends a new contributor entry (§N) and never rewrites previous entries.
+
+---
+
+## ━━━ DETECTION (Self-Correct Mode) ━━━
+
+**Trigger:** `$ARGUMENTS` resolves to a file (not a folder) AND that file's first H1 matches:
+
+```
+# Track: <scope> — <YYYY-MM-DD>
+```
+
+When triggered, switch from "create / append contributor" mode to **verify-and-repair mode**. Behavior is governed by [`../_shared/output_conventions.md`](../_shared/output_conventions.md) §3, with these track-specific gap-fills.
+
+### What track self-correct CHECKS
+
+1. **§N order:** sections numbered §0, §1, §2, … with no gaps. Flag (do not renumber) if a §N is missing — could be intentional from another model.
+2. **Model-tag formatting:** every `# §N — ... *(<Model> — <HH:MM>)*` heading has both model name and time in italics+parens.
+3. **Footer present:** the `## 📂 Generated Files (<YYYYMMDD>)` block exists at the bottom with at least the Base Reference Files table.
+4. **Link integrity:** every relative markdown link in the file resolves to an actual on-disk file from the session file's directory. Broken links → list them under a `## ⚠️ Self-Correct Findings` section near the bottom (do NOT edit the original §N bodies — see "What it does NOT do" below).
+5. **Bare /wbX in `Recommended Next` tables:** expand to full-syntax `/wbX <target> [flags]` per §2 of output_conventions.
+6. **Plain-text file references in `Files read` / `Files created`:** convert to relative markdown links per §1.
+7. **Status freshness:** if the front-matter `> **Status:**` is `🟢 ACTIVE` but the most recent §N is > 24h old, change to `🟡 STALE` and note the lag in `## ⚠️ Self-Correct Findings`.
+
+### What track self-correct does NOT do
+
+- Does NOT rewrite the body text of any §N (preserve each model's voice and audit trail).
+- Does NOT add or delete §N sections.
+- Does NOT change the §0 strategic vision content.
+- Does NOT re-run `/wbStandup` or any sub-command (no new analysis).
+- Does NOT delete broken-link references — only flags them in the Findings section.
+
+### Where self-correct WRITES
+
+Append (or update if it exists) a **single section** at the bottom of the file, just above the `## 📂 Generated Files` footer:
+
+```markdown
+## ⚠️ Self-Correct Findings *(<Model> — <YYYY-MM-DD HH:MM>)*
+
+**Verified:** §N order, model-tag formatting, footer present (✅ all clean) OR list specific issues found.
+
+**Auto-repaired:**
+- bare `/wbX` → full-syntax in §<N> "Recommended Next" table (rows: …)
+- plain-text path `<old>` → relative link `[<text>](<rel>)` in §<N>
+
+**Flagged (NOT auto-repaired — needs human decision):**
+- Broken link in §<N>: `<bad-link>` → no file at resolved path `<resolved>`
+- §<N> missing model tag — last edited by ?
+
+> _Self-corrected: <YYYY-MM-DD HH:MM> by <Model>_
+```
+
+If a previous self-correct run already wrote this section, **update it in place** (do not stack multiple Findings sections — the file is meant to grow §N entries, not Findings entries).
+
+---
+
+---
+
+## How It Works
+
+`/wbTrack` is a **toggle**, not a one-shot command. It changes the model's behavior for the rest of the session.
+
+| State | What happens when you run a `/wb*` command |
+|---|---|
+| **Tracking OFF** (default) | Command creates its report in `reports/<date>/<type>/<type>_<target>_<date>.md` (universal daily file). Nothing else. |
+| **Tracking ON** (`/wbTrack` active) | Command creates its report **AND** appends a §N section to the **universal daily session file**. |
+
+**Key principle:** Commands always produce their normal reports. Tracking is an **addition**, never a replacement.
+
+### v2: One File Per Day Per Scope
+
+In v2, **all models share a single session file per day**. There is no `<model>/` subfolder under `tracks/`.
+
+| v1 (old) | v2 (current) |
+|---|---|
+| `tracks/<date>/<model_slug>/track_<target>_<timestamp>.md` | `tracks/<date>/track_<target>_<date>.md` |
+| One file per model per session | **One file per day per scope, all models contribute** |
+| 4 models = 4 files + 24 derivatives | 4 models = **1 file** + 6 derivatives |
+
+---
+
+## Scoping Rules
+
+### Syntax
+```bash
+/wbTrack <target>       # scoped to a specific package/app
+/wbTrack                # scoped to the monorepo root (core2)
+```
+
+### What the target arg controls
+
+| | `/wbTrack packages/wb-core` | `/wbTrack` (no args) |
+|---|---|---|
+| **§0 reads** | wb-core's source, reports, `.agents/` | Monorepo-wide state |
+| **Session file lives in** | `packages/wb-core/.wb/workflows/tracks/` | `.wb/workflows/tracks/` (core2 root) |
+| **Best for** | Focused session on one package | Cross-package session |
+
+### One session at a time (per model)
+
+If tracking is already active for this model:
+```
+/wbTrack packages/wb-dataviewer
+
+⚠️ Already tracking: core2
+   Run /wbTrack --stop first, then start a new session.
+```
+
+You cannot run two `/wbTrack` sessions in parallel from the same model. Stop the current one before starting another.
+
+### Multi-model sessions (v2 change)
+
+Multiple **different** models can track the **same scope on the same day**. They all contribute to the same file:
+
+```
+Model A (Opus 4.7 via Cline):    /wbTrack core2  → creates session file + writes §0
+Model B (Gemini 3.1 Pro via AG):  /wbTrack core2  → file exists → appends contributor entry
+Model C (DeepSeek V4 via OC):     /wbTrack core2  → file exists → appends contributor entry
+```
+
+But each model can only track **one scope at a time**. Model A tracking `core2` cannot also start tracking `wb-core` — it must `/wbTrack --stop` first.
+
+### Different scopes are different files
+
+Different models tracking different scopes write to **different files** — this is fine:
+```
+Model A tracks core2:    → core2/.wb/workflows/tracks/20260429/track_core2_20260429.md
+Model B tracks wb-core:  → packages/wb-core/.wb/workflows/tracks/20260429/track_wb-core_20260429.md
+```
+
+### Cross-package work during a session
+
+If you're tracking `wb-core` and run `/wbTest packages/wb-dataviewer`:
+- The **test report** goes to `packages/wb-dataviewer/.wb/workflows/reports/.../tests/` (normal behavior — report always goes to the target package)
+- The **§N section** gets appended to the wb-core session file (because wb-core is the active tracking target)
+
+This is fine — the session file is a narrative of what **you did**, not where the reports land.
+
+---
+
+## 🚀 The /wbTrack Prompt (copy to model ↓)
+
+```
+━━━━━━━━━━━━━ /wbTrack v2 ━━━━━━━━━━━━━
+
+📁 TARGET: __TARGET_PATH__
+📅 DATE: __TODAY__
+🤖 MODEL: __YOUR_MODEL_NAME__
+🖥️ CLIENT: __YOUR_CLIENT__ (Cline / Antigravity / OpenCode / Gemini CLI / etc.)
+
+━━━ TRACKING MODE ACTIVATED ━━━
+
+From this point forward, you are in session tracking mode.
+
+1. LOCATE the universal session file at:
+   __TARGET_PATH__/.wb/workflows/tracks/__TODAY_YYYY__/__TODAY_MM__/__TODAY_DD__/track___TARGET_NAME_____TODAY_YYYY__/__TODAY_MM__/__TODAY_DD__.md
+
+   NOTE: `tracks/` is a SIBLING of `reports/`, NOT inside it.
+   NOTE: There is NO `<model_slug>/` subfolder — one file per day per scope.
+
+2. CHECK if the file already exists:
+
+   ┌─────────────────────────────────────────────────────────────┐
+   │ FILE DOES NOT EXIST → CREATE (you are the first model today)│
+   └─────────────────────────────────────────────────────────────┘
+
+   Create the file with this header:
+
+   # Track: __TARGET_NAME__ — __TODAY__
+
+   > **Target:** Monorepo root (`frontEnd/wbc-ui/__TARGET_PATH__/`)
+   > **Created by:** __YOUR_MODEL_NAME__ via __YOUR_CLIENT__
+   > **Started:** __TODAY__ __CURRENT_TIME__
+   > **Status:** 🟢 ACTIVE
+
+   ---
+
+   Then write §0 — your strategic vision BEFORE I give you any commands:
+
+   # §0 — Strategic Vision *(__YOUR_MODEL_NAME__ — __CURRENT_TIME__)*
+
+   ## Current State
+   [Read the package/app source code, .wb/workflows/ files, and past reports.
+   Summarize: What is the current state of this target? What was done before?
+   What is broken, pending, or deferred?]
+
+   ## Past Debt: `/wbStandup __TARGET_NAME__` output
+   **MANDATORY:** Execute `/wbStandup __TARGET_NAME__` as a sub-command RIGHT NOW.
+   Paste its FULL output below — do NOT summarize, do NOT reference a report file.
+   This gives subsequent models the complete context without needing to re-run standup.
+
+   ```
+   [PASTE FULL /wbStandup OUTPUT HERE]
+   ```
+
+   ## My Recommendation
+   [If you were working on this package today, what would YOU do?
+   Prioritize based on: blockers first, then technical debt, then new features.
+   List 5-8 commands in recommended order with reasoning.
+   **For each recommendation, specify 2-3 models** from `commands/model_recommendations.md`, ordered best → budget.]
+
+   ## Suggested Command Sequence
+
+   | Order | Command | Why | Recommended Models (ordered) | Alternatives |
+   |---|---|---|---|---|
+   | 1 | `/wbStandup __TARGET_NAME__` | Resume from last session | ⚡ Gemini 3 Flash / DeepSeek V4 Flash | `/wbContext` if no prior session |
+   | 2 | `...` | `...` | 💻 Model1 / Model2 / Model3 | `...` |
+   | ... | | | | |
+
+   ## First Command
+
+   **Recommended:** `/wbXxx __TARGET_NAME__ [flags]`
+   **Recommended Model:** `Model1 (1st) / Model2 (2nd)` — [why this model for this task]
+   **Why this first:** [1-2 sentences]
+   **Alternative:** `/wbYyy __TARGET_NAME__` — [why this is also valid]
+
+   > [!NOTE]
+   > This is my suggestion. You decide what to run. Tell me your first command.
+
+   ┌─────────────────────────────────────────────────────────────┐
+   │ FILE ALREADY EXISTS → APPEND (another model started today) │
+   └─────────────────────────────────────────────────────────────┘
+
+   Read the existing session file. Determine the next §N number.
+   Append a contributor entry:
+
+   ---
+
+   # §N — Contributor Entry *(__YOUR_MODEL_NAME__ via __YOUR_CLIENT__ — __CURRENT_TIME__)*
+
+   > **Model:** __YOUR_MODEL_NAME__
+   > **Client:** __YOUR_CLIENT__
+   > **Joined:** __TODAY__ __CURRENT_TIME__
+
+   ## My Assessment
+   [Read the existing session file + current source code state.
+   What has changed since the last entry? What do you see differently?
+   Agreement/disagreement with previous §0 or §N recommendations.]
+
+   ## My Recommendation
+   [If you were continuing this session, what would you do next?
+   How does your view differ from the previous contributor(s)?]
+
+   ## Suggested Next Steps
+
+   | Priority | Command | Why | Recommended Models (ordered) | Alternative |
+   |---|---|---|---|---|
+   | 🔴 1st | ... | ... | ... | ... |
+   | 🟡 2nd | ... | ... | ... | ... |
+
+   > [!NOTE]
+   > This is my perspective. Previous contributors may have different priorities.
+
+3. For EVERY /wb* command I give you:
+
+   a) Execute the command normally and create its report in the universal daily file:
+      __TARGET_PATH__/.wb/workflows/reports/__TODAY_YYYY__/__TODAY_MM__/__TODAY_DD__/<type>/<type>___TARGET_NAME_____TODAY_YYYY__/__TODAY_MM__/__TODAY_DD__.md
+
+   b) THEN append a §N section to the universal session file in this exact format:
+
+   # §N — `/wbCommandName target [flags]` *(__YOUR_MODEL_NAME__ — __CURRENT_TIME__)*
+
+   ## What the user did
+
+       /wbCommandName target [flags]
+
+   **What the user wants:** [1-2 sentences]
+
+   ## What happened
+
+   ### Files read (input)
+   [List files you read]
+
+   ### Files created/modified (output)
+   | File | Action | What it contains |
+   |---|---|---|
+
+   ### Report file
+   [Path to the report entry in reports/<date>/<type>/<type>_<target>_<date>.md — specify Entry #N]
+
+   ## Commentary
+
+   - **Outcome:** (What went well / What went wrong / Final result)
+   - **Insights:** (Lesson learned / Key finding / Self-correction note)
+   - **Decisions:** (Why this command now / Why this matters / What was NOT produced)
+   - **Status:** (Conclusion / Current state)
+   - **Next Step:** (Next step used / Next meaningful step)
+
+   > [!TIP/WARNING/IMPORTANT]
+   > [High-value strategic callout]
+
+   ## Recommended Next
+
+   | Priority | Command | Why | Recommended Models (ordered) | Alternative |
+   |---|---|---|---|---|
+   | 🔴 1st | `/wbXxx target` | [reason based on what just happened] | 💻 Model1 / Model2 | `/wbYyy target` |
+   | 🟡 2nd | `/wbZzz target` | [reason] | ⚡ Model1 / Model2 | — |
+   | 🟢 3rd | `/wbAaa target` | [reason] | 💻 Model1 / Model2 | — |
+
+4. You are NOT a passive logger. You are a strategic agent:
+   - Analyze outcomes: say WHY it matters, not just what happened
+   - Provide vision: compare what the AI saw vs. what is actually true
+   - Decide next steps: if a command reveals a critical issue, suggest course correction
+   - After EVERY section, suggest what comes next (the "Recommended Next" table)
+
+━━━ AUTO-APPEND FOOTER ━━━
+
+At the VERY END of the file (after "What's Next?"), you MUST append the `## 📂 Generated Files (__YYYYMMDD__)` cross-link footer. Do NOT use simple tables. You MUST use the rich "Tier 1" layout from `_shared/output_conventions.md` §5.
+
+Format required:
+```markdown
+---
+## 📂 Generated Files (__YYYYMMDD__)
+> Auto-appended per `_shared/output_conventions.md` §5. Same-level snapshot of top-level command outputs at write time.
+
+### 📚 Base Reference Files
+| Type | File | Description |
+|---|---|---|
+| Foundational | [context.md](../../../../../context.md) | Permanent Identity and Architecture (Source of Truth) |
+| Snapshot | [context_<scope>_<date>.md](../contexts/context_<scope>_<date>.md) | Daily snapshot used for current session context |
+| Foundational | [dev.md](../../../../../dev.md) | Permanent Development Commands and Status |
+
+### Global Files (`core2/` monorepo root)
+| Category | File | Source Command |
+|---|---|---|
+| Reports | [audit_core2_<date>.md](../../../../../../../../../../.wb/workflows/reports/<YYYY>/<MM>/<DD>/audits/audit_core2_<date>.md) | `/wbAudit core2/` |
+| Reports | [plan_core2_<date>.md](../../../../../../../../../../.wb/workflows/reports/<YYYY>/<MM>/<DD>/plans/plan_core2_<date>.md) | `/wbPlan core2/` |
+| Tracks | [track_core2_<date>.md](../../../../../../../../../../.wb/workflows/tracks/<YYYY>/<MM>/<DD>/track_core2_<date>.md) | `/wbTrack core2/` |
+
+<details>
+  <summary>📂 Sub-Package: [Active Package Name]</summary>
+
+| Category | File | Source Command |
+|---|---|---|
+| Reports | [audit_subpkg_<date>.md](../../../../../../../../../../apps/wb-core/subpkg/.wb/workflows/reports/<YYYY>/<MM>/<DD>/audits/audit_subpkg_<date>.md) | `/wbAudit` |
+
+</details>
+```
+```
+
+---
+
+## Variables to Replace
+
+| Variable | Description | Example |
+|---|---|---|
+| `__TARGET_PATH__` | Path to the package or app | `packages/wb-core` |
+| `__TODAY__` | Today's date (YYYY-MM-DD) | `2026-04-29` |
+| `__TODAY_YYYY__/__TODAY_MM__/__TODAY_DD__` | Today's date (no dashes) | `20260429` |
+| `__YOUR_MODEL_NAME__` | Model running the session | `DeepSeek V4 Pro` |
+| `__YOUR_CLIENT__` | AI client/tool being used | `Antigravity` / `Cline` / `OpenCode` / `Gemini CLI` |
+| `__MODEL_SLUG__` | Lowercase model name for report folder paths | `deepseek-v4-pro` |
+| `__TARGET_NAME__` | Short name for filenames | `wb-core` |
+| `__CURRENT_TIME__` | Current time (HH:MM) | `22:38` |
+
+---
+
+## Directory Structure
+
+```
+.wb/workflows/
+├── reports/                               ← command outputs (scanned by /wbStandup, /wbPlan, etc.)
+│   └── 20260429/
+│       ├── tests/deepseek-v4-pro/test_wb-core_unit_*.md        ← from /wbTest
+│       ├── plans/gemini-3.1-pro/plan_wb-core_*.md              ← from /wbPlan
+│       ├── audits/opus-4.7/audit_wb-core_*.md                  ← from /wbAudit
+│       └── standups/gemini-3.1-pro/standup_wb-core_*.md        ← from /wbStandup
+│
+└── tracks/                          ← session narratives (universal daily file)
+    └── 20260429/
+        ├── track_wb-core_20260429.md              ← ONE file, ALL models contribute
+        ├── track_wb-core_20260429_tips.md          ← derivative (all models' tips)
+        ├── track_wb-core_20260429_warnings.md
+        ├── track_wb-core_20260429_importants.md
+        ├── track_wb-core_20260429_commentaries.md
+        ├── track_wb-core_20260429_all_commands.md
+        └── track_wb-core_20260429_resume.md
+```
+
+### The Key Principle
+
+| Folder | Contains | Scanned by | Purpose |
+|---|---|---|---|
+| `reports/<date>/<type>/<type>_<target>_<date>.md` | Structured command outputs (universal daily file) | `/wbStandup`, `/wbContext`, `/wbPlan`, `/wbAudit` | Data for automation |
+| `tracks/<date>/` | Universal session narrative | Humans, future sessions, all models | Audit trail + learning |
+
+**Reports** = what happened (facts, scores, file lists).
+**Walkthroughs** = why it happened (analysis, decisions, recommendations) — **from all models' perspectives**.
+
+---
+
+## Best Practice: Daily Session Pattern
+
+The recommended workflow is: **one universal session file per day per scope, contributed to by all models**.
+
+```
+Morning (Model A — Opus 4.7 via Cline):
+  /wbTrack                              ← creates file + writes §0
+
+Mid-day (Model B — Gemini 3.1 Pro via Antigravity):
+  /wbTrack                              ← file exists → appends contributor entry
+  /wbPlan wb-core                       ← report + §N (tagged as Gemini)
+  /wbTest wb-core                       ← report + §N (tagged as Gemini)
+  /wbTrack --stop                       ← §STOP for Gemini
+
+Evening (Model C — DeepSeek V4 via OpenCode):
+  /wbTrack                              ← file exists → appends contributor entry
+  /wbRefactor wb-core src/WBC.js        ← report + §N (tagged as DeepSeek)
+  /wbTrack --stop --finalize            ← §STOP for DeepSeek + extract derivatives
+  /wbGit                                ← commit everything
+```
+
+This gives you a **single daily narrative** with contributions from all models — like a relay journal. Tomorrow, `/wbStandup` scans `reports/` for the facts, and you can read the session file for the multi-model story.
+
+---
+
+## When to Use /wbTrack
+
+| Situation | Use /wbTrack? |
+|---|---|
+| Starting a full day's work with a model | ✅ Yes — run once at the start |
+| Long session with 5+ commands | ✅ Yes — you want a trail |
+| Learning a new package | ✅ Yes — commentary helps future you |
+| Documenting a workflow for the team | ✅ Yes — produces shareable session files |
+| Quick one-off command (e.g., `/wbTest`) | ❌ No — just run the command |
+| Running `/wbGit` at end of day | ❌ No — no need to track a commit |
+| Model already verified via `/wbCheck` | ✅ Optional — `/wbTrack` adds value independently |
+
+---
+
+## Relationship with Other Commands
+
+| Command | Relationship |
+|---|---|
+| `/wbCheck` | Independent. `/wbCheck` verifies the model. `/wbTrack` logs the session. You can use either or both. |
+| `/wbStandup` | Scans `reports/` — unaffected by tracking. Session files live in `tracks/`. |
+| `/wbPlan` | Creates plan in `reports/plans/`. If tracking is ON, §N documents the plan creation. |
+| `/wbTest` | Creates test report in `reports/tests/`. If tracking is ON, §N documents the test results. |
+| `/wbTrack --stop` | Ends tracking **for this model**. Optionally extracts derivatives with `--finalize`. |

package/templates/commands/wbTranslate/wbTranslate_template.md

@@ -0,0 +1,114 @@
+# /wbTranslate: Execution Template
+
+
+<!-- HELP_GATE_START -->
+## Help intercept (handle FIRST — before any other action)
+
+**If `$ARGUMENTS` contains `--help`, `-h`, or `--h`** (case-insensitive, anywhere in the args), DO NOT execute the command's normal procedure. Instead:
+
+1. Output the **HELP BLOCK** below verbatim (rendered as markdown).
+2. Stop. Do not perform any file reads, writes, or other tool calls.
+3. Do not generate any reports under `.wb/workflows/reports/`.
+
+Otherwise, ignore this section and proceed to the rest of the template.
+
+### HELP BLOCK — `/wbTranslate`
+
+## Two forms
+
+```
+/wbTranslate <file>                  # extract + translate all hardcoded strings
+/wbTranslate <file> --new-only       # only handle strings not yet keyed
+```
+
+## When to run
+
+- After the component's UI is stable (post-design, post-iteration).
+- Before shipping a user-facing release that needs FR/AR coverage.
+- When adding new user-facing strings to an already-translated component (use `--new-only`).
+
+## When *not* to run
+
+- Mid-iteration, when copy is still changing. You'll re-translate every iteration.
+- On a component without i18n setup. Command will refuse.
+- On a component already fully translated (use `--new-only` instead).
+- On developer-facing components that won't be shipped to users.
+
+## Prerequisites
+
+- vue-i18n (or equivalent) configured in the app.
+- `src/i18n/{en,fr,ar}.json` (or equivalent) exist.
+- The component imports `useI18n` or has access to `$t`.
+
+If any prerequisite is missing, the command refuses with setup steps.
+
+## Reading the output
+
+- **Extracted strings list** — exactly what will become i18n keys.
+- **Skipped strings list** — what was deliberately not extracted (CSS classes, events, URLs).
+- **Ambiguous strings** — requires your disambiguation before proceeding.
+- **Three JSON additions** — en, fr, ar. Each with cultural awareness.
+- **RTL note** — Arabic-specific layout flags.
+
+## The translation review step
+
+AI-generated translations are drafts. Review especially:
+
+- **Tone**: formal vs. informal "you" (vous/tu, anta/anti).
+- **Brand voice**: action verbs may differ ("Save" → "Enregistrer" vs. "Sauvegarder").
+- **RTL layout**: Arabic strings may overflow; layout may need adjustment.
+- **Date/number formatting**: not handled by `/wbTranslate`; uses i18n built-ins.
+
+If you don't speak FR/AR, defer to native speakers before shipping. AI translations of common UI strings are decent; AI translations of nuanced messaging are guesses.
+
+## When /wbTranslate is the wrong command
+
+- Pluralization rules → vue-i18n's plural syntax; AI can't fully infer.
+- Locale-aware formatting (dates, currency) → use Intl APIs, not strings.
+- Server-rendered text → translate server-side.
+- A11y attributes (`aria-label` etc.) → these *are* user-facing; `/wbTranslate` should handle them. If it skips them, re-prompt.
+
+> For deeper reading: [`docs_claude/commands/wbTranslate/wbTranslate_practical_claude.md`](../../docs/docs_claude/commands/wbTranslate/wbTranslate_practical_claude.md) (or the `_eli5_`, `_expert_`, `_examples_` siblings).
+
+<!-- FLAGS_TABLE_START -->
+## Flags & shortcuts
+
+Both forms are equivalent — pass either:
+
+| Long form | Shortcut |
+|---|---|
+| `--new-only` | `-n` |
+
+`-h` / `--help` / `--h` (any command) prints this help block instead of executing.
+<!-- FLAGS_TABLE_END -->
+<!-- HELP_GATE_END -->
+
+<!-- FLAG_NORMALIZE_START -->
+## Flag normalization (apply BEFORE parsing args)
+
+Before processing `$ARGUMENTS`, normalize these short-form flags to their long equivalents:
+
+- `-n` → `--new-only`
+
+The rest of this template documents only the long forms; the substitution above is the only place short forms are mentioned.
+<!-- FLAG_NORMALIZE_END -->
+
+
+
+**ROLE:** The Linguist
+**TARGET:** The provided UI component or translation directory.
+
+## ━━━ OBJECTIVE ━━━
+Your job is to ensure the monorepo remains fully multilingual (EN, FR, AR). You must extract hardcoded strings and generate localized JSON files.
+
+## ━━━ PHASE 1: EXTRACTION ━━━
+1. Scan the target UI component (e.g., `.vue` files).
+2. Identify any hardcoded human-readable strings (e.g., `<button>Submit</button>`).
+3. Replace them with the appropriate i18n translation key (e.g., `<button>{{ $t('buttons.submit') }}</button>`).
+
+## ━━━ PHASE 2: TRANSLATION GENERATION ━━━
+1. Take the extracted strings and generate perfect context-aware translations for French (`fr.json`) and Arabic (`ar.json`).
+2. Ensure RTL (Right-to-Left) UI considerations are respected for Arabic.
+
+## ━━━ PHASE 3: REPORTING ━━━
+Summarize the translation keys added.