@leejungkiin/awkit 1.4.0 → 1.4.3

package/skills/short-maker/outputs/fitbite-promo/tts_script.txt

@@ -0,0 +1,11 @@
+SCENE 1:
+Tông giọng: 'Gặp khó khăn khi theo dõi dinh dưỡng hàng ngày?'
+
+SCENE 2:
+Tông giọng: 'FitBite Pro với AI thông minh sẽ tự động lên thực đơn chuẩn y khoa cho bạn.'
+
+SCENE 3:
+Tông giọng: 'Báo cáo chi tiết, giao diện tối ưu hoàn hảo cho cả iPad.'
+
+SCENE 4:
+Tông giọng: 'Tải ngay FitBite Pro trên App Store để thay đổi vóc dáng!'

package/skills/short-maker/scripts/google-flow-cli/.project-identity

@@ -0,0 +1,41 @@
+{
+  "_comments": {
+    "projectId": "Auto-generated. DO NOT change — used by Symphony for task scoping.",
+    "trello": "Fill in your Trello board/list/card names. Run \"awkit trello info\" to verify."
+  },
+  "projectName": "GoogleFlowCli",
+  "projectId": "googleflowcli",
+  "projectType": "mobile-firebase",
+  "bundleIdentifier": "com.company.googleflowcli",
+  "primaryLanguage": "en",
+  "techStack": {
+    "platform": "Mobile",
+    "backend": "Firebase"
+  },
+  "services": {
+    "firebase": {
+      "enabled": true,
+      "features": [
+        "analytics",
+        "crashlytics",
+        "remote-config",
+        "auth"
+      ]
+    }
+  },
+  "trello": {
+    "board": "Your Board Name",
+    "list": "Your List Name",
+    "card": "Your Card Name"
+  },
+  "projectStage": "development",
+  "codingStandards": {
+    "language": "en",
+    "namingConvention": "camelCase",
+    "indentation": "spaces-4",
+    "lineLength": 120
+  },
+  "projectGoals": [],
+  "createdDate": "2026-03-29",
+  "lastUpdated": "2026-03-29"
+}

package/skills/short-maker/scripts/google-flow-cli/.trae/rules/project_rules.md

@@ -0,0 +1,52 @@
+# Project Rules (gflow)
+
+## Cost Model
+
+- Image generation is free (0 credits).
+- Video generation costs 20 credits per video.
+
+## Workflow Optimization (Credits)
+
+- Default to images first: use image generation to validate prompts, style, character consistency, and scene composition.
+- Only generate videos after explicit approval that the image storyboard is acceptable.
+- When videos are required, minimize iterations:
+  - Keep prompts stable and only adjust one variable at a time.
+  - Prefer shorter runs (e.g., single segment) before multi-segment/extended workflows.
+  - Avoid unnecessary retries or batch runs unless requested.
+
+## Output Planning
+
+- For scripts/storyboards: generate one image per scene first, then move to video generation.
+- Always save outputs under `outputs/` with a predictable naming scheme (scene/seg numbering).
+
+## Voiceover (TTS)
+
+- Videos generated by Flow may come without voice. Treat voice as a separate pipeline step.
+- TTS is mandatory via LucyLab JSON-RPC only (use runtime config, avoid keeping local capture/log files in repo).
+- Do not use local TTS (e.g., macOS `say`) in this project.
+
+## TTS Tooling
+
+- **LucyLab batch TTS script**: `python lucylab_tts.py`
+  - Token handling:
+    - Use environment variable `LUCYLAB_BEARER` (do not store bearer tokens in repo files).
+  - Voice selection:
+    - Prefer explicit `--voice <id>[:name]` to avoid storing voice catalogs in repo.
+    - If a voice catalog is needed for AI selection, keep a compact library file with only `items[].{id,name,description,tag}`.
+  - Common usage:
+    - Plain text:
+      - `LUCYLAB_BEARER=... python lucylab_tts.py --out-dir outputs/tts-lucylab --voice <id>:<name> --text "..." `
+    - Script mode (extracts thoại from `SCENE ... '...')`:
+      - `LUCYLAB_BEARER=... python lucylab_tts.py --out-dir outputs/tts-lucylab --voice <id>:<name> --text-file script.txt`
+    - Build compact voice library from a captured log file:
+      - `python lucylab_tts.py --voice-json voice.json --export-voice-library voices_library.json`
+    - Build split voice library folder (load only one chunk/category into AI context):
+      - `python lucylab_tts.py --voice-json voice.json --export-voice-library outputs/voice_library`
+- **Audio to video merge** (ffmpeg): add the generated voiceover track to the final merged video after transitions.
+  - Output naming convention:
+    - Video without VO: `outputs/.../script-full.mp4`
+    - Video with VO: `outputs/.../script-full-vo.mp4`
+
+## Safety
+
+- Never commit or share bearer tokens (e.g., from `tts_curl.txt`). Use `LUCYLAB_BEARER` and keep local-only artifacts under `outputs/`.

package/skills/short-maker/scripts/google-flow-cli/CODEBASE.md

@@ -0,0 +1,67 @@
+# GoogleFlowCli — Codebase Map
+
+> **Project Type:** mobile-firebase | **Tech Stack:** Mobile + Firebase
+> **Last Synced:** 2026-03-29
+
+> [!NOTE]
+> AI PHẢI đọc file này TRƯỚC KHI dùng `grep_search` / `codebase_search` / `list_dir`.
+> Đọc → xác định target → search cụ thể. KHÔNG search "mù".
+
+---
+
+## 🗺️ Directory Structure
+
+```
+src/
+├── features/
+├── services/
+└── models/
+```
+
+---
+
+## 📁 File Index
+
+> Compact format: `path` → purpose (1 line mỗi file)
+
+### Core / Infrastructure
+
+| Path | Purpose |
+|------|---------|
+| `{{CORE_FILE_1}}` | {{PURPOSE_1}} |
+| `{{CORE_FILE_2}}` | {{PURPOSE_2}} |
+
+### Features
+
+| Path | Purpose |
+|------|---------|
+| `{{FEATURE_FILE_1}}` | {{PURPOSE_1}} |
+| `{{FEATURE_FILE_2}}` | {{PURPOSE_2}} |
+
+### UI / Views
+
+| Path | Purpose |
+|------|---------|
+| `{{VIEW_FILE_1}}` | {{PURPOSE_1}} |
+| `{{VIEW_FILE_2}}` | {{PURPOSE_2}} |
+
+> ⚠️ Thêm file mới vào section phù hợp. Dùng `/codebase-sync` để auto-update.
+
+---
+
+## 🔑 Key Files
+
+| File | Purpose |
+|------|---------|
+| `.project-identity` | AI project context |
+| `CODEBASE.md` | This file — codebase map for AI |
+
+---
+
+## 📝 Notes
+
+_Architecture decisions, gotchas, và context quan trọng._
+
+---
+
+*Auto-generated by `awkit init` — keep updated with `/codebase-sync`*

package/skills/short-maker/scripts/google-flow-cli/GoogleFlowCli.code-workspace

@@ -0,0 +1,29 @@
+{
+  "folders": [
+    {
+      "path": "."
+    }
+  ],
+  "settings": {
+    "editor.formatOnSave": true,
+    "editor.tabSize": 4,
+    "files.exclude": {
+      "**/.DS_Store": true,
+      "**/node_modules": true,
+      "**/.git": true,
+      "**/build": false,
+      "**/.gradle": false,
+      "**/DerivedData": false
+    },
+    "files.watcherExclude": {
+      "**/node_modules/**": true,
+      "**/build/**": true,
+      "**/DerivedData/**": true
+    }
+  },
+  "extensions": {
+    "recommendations": [
+      "github.copilot"
+    ]
+  }
+}

package/skills/short-maker/scripts/google-flow-cli/README.md

@@ -0,0 +1,168 @@
+# gflow — CLI for Google Flow
+
+A command-line interface to [Google Flow](https://flow.google) (AI image & video generation), built using the same reverse-engineering approach as [tmc/nlm](https://github.com/tmc/nlm) for NotebookLM.
+
+Lets AI agents and scripts generate images/videos via Google Flow without the GUI.
+
+## Architecture
+
+```
+┌────────────────────────────────────┐
+│  CLI Layer (click)                 │  gflow/cli/main.py
+│  generate-image, generate-video,   │
+│  list, download, collections, raw  │
+├────────────────────────────────────┤
+│  API Client (FlowClient)          │  gflow/api/client.py
+│  High-level ops, polling, parsing  │
+├────────────────────────────────────┤
+│  BatchExecute Protocol            │  gflow/batchexecute/client.py
+│  RPC encoding, SAPISIDHASH,       │
+│  chunked response decoding         │
+├────────────────────────────────────┤
+│  Browser Auth                     │  gflow/auth/browser_auth.py
+│  Cookie extraction from Chrome,    │
+│  Selenium interactive login        │
+└────────────────────────────────────┘
+```
+
+This is the same layered architecture as `tmc/nlm`:
+- **Auth** extracts Google cookies from your browser (browser_cookie3 or Selenium)
+- **BatchExecute** encodes RPCs into Google's wire format and decodes responses
+- **API Client** wraps BatchExecute with typed methods for each Flow feature
+- **CLI** exposes everything as clean subcommands
+
+## Install
+
+```bash
+pip install -e .
+```
+
+Or with optional Selenium support for interactive login:
+```bash
+pip install -e ".[dev]"
+pip install selenium
+```
+
+## Quick Start
+
+```bash
+# 1. Authenticate (extracts cookies from your Chrome browser)
+gflow auth
+
+# 2. Generate an image
+gflow generate-image "a cat astronaut floating in space"
+
+# 3. Generate a video
+gflow generate-video "a timelapse of a flower blooming"
+
+# 4. List your assets
+gflow list
+
+# 5. Download an asset
+gflow download <asset-id> -o output.png
+```
+
+## Setup: Discovering RPC IDs
+
+Google Flow uses the same BatchExecute protocol as NotebookLM, but with different RPC endpoint IDs. You need to discover these by inspecting network traffic:
+
+1. Open [flow.google](https://flow.google) in Chrome
+2. Open DevTools → Network tab
+3. Filter requests by `batchexecute`
+4. Perform an action (e.g., generate an image)
+5. In the request payload, find the `rpcids` parameter — that's the RPC ID
+6. Update `gflow/api/rpc_ids.py` with the real ID
+
+Check which IDs are configured:
+```bash
+gflow rpc-ids
+```
+
+Use raw mode to test discovered IDs:
+```bash
+gflow raw "xYz123" --args '["my prompt", "16:9"]'
+```
+
+## Commands
+
+| Command | Description |
+|---------|-------------|
+| `gflow auth` | Authenticate with Google Flow |
+| `gflow auth --status` | Check auth status |
+| `gflow auth --clear` | Clear saved credentials |
+| `gflow generate-image PROMPT` | Generate images (Imagen 4) |
+| `gflow generate-video PROMPT` | Generate videos (Veo 3.1) |
+| `gflow list` | List assets in your library |
+| `gflow get ASSET_ID` | Get asset details |
+| `gflow download ASSET_ID` | Download an asset |
+| `gflow delete ASSET_ID` | Delete an asset |
+| `gflow collections list` | List collections |
+| `gflow collections create NAME` | Create a collection |
+| `gflow collections add COL_ID ASSET_ID` | Add asset to collection |
+| `gflow raw RPC_ID` | Execute raw RPC (discovery mode) |
+| `gflow rpc-ids` | Show configured RPC IDs |
+
+All commands support `--json` for machine-readable output (ideal for scripts/agents).
+
+## Environment Variables
+
+| Variable | Description |
+|----------|-------------|
+| `GFLOW_AUTH_TOKEN` | Auth token (overrides saved credentials) |
+| `GFLOW_COOKIES` | Cookie string (overrides saved credentials) |
+| `GFLOW_CHROME_PATH` | Path to Chrome executable |
+| `GFLOW_DEBUG` | Set to `true` for debug output |
+
+## For AI Agents / Scripts
+
+Every command supports `--json` output:
+
+```bash
+# Generate and get JSON response
+gflow generate-image "a logo" --json | jq '.[0].url'
+
+# List assets as JSON for processing
+gflow list --type image --json | jq '.[].id'
+
+# Pipeline: generate, wait, download
+ASSET_ID=$(gflow generate-video "ocean waves" --json | jq -r '.[0].id')
+gflow download "$ASSET_ID" -o waves.mp4
+```
+
+## How It Works (Same as tmc/nlm)
+
+1. **Browser Auth**: Extracts Google cookies from Chrome/Brave/Edge profiles using `browser_cookie3`, then fetches the XSRF token from the Flow page HTML
+2. **BatchExecute Protocol**: Encodes RPC calls into Google's `batchexecute` wire format — form-encoded POST with nested JSON arrays, SAPISIDHASH authorization header
+3. **Response Decoding**: Parses Google's chunked response format (byte-count prefixed JSON chunks with `wrb.fr` markers and multi-layer JSON encoding)
+4. **Retry Logic**: Exponential backoff for transient errors (429, 500, 502, 503, 504)
+
+## Project Structure
+
+```
+gflow-py/
+├── pyproject.toml              # Package config & dependencies
+├── README.md
+├── gflow/
+│   ├── __init__.py
+│   ├── auth/
+│   │   ├── __init__.py
+│   │   └── browser_auth.py     # Cookie extraction & Selenium login
+│   ├── batchexecute/
+│   │   ├── __init__.py
+│   │   └── client.py           # Google BatchExecute protocol
+│   ├── api/
+│   │   ├── __init__.py
+│   │   ├── client.py           # FlowClient (high-level API)
+│   │   ├── models.py           # Pydantic models (Asset, Collection, etc.)
+│   │   └── rpc_ids.py          # RPC endpoint IDs (fill these in!)
+│   └── cli/
+│       ├── __init__.py
+│       └── main.py             # Click CLI commands
+└── tests/
+    ├── __init__.py
+    └── test_batchexecute.py    # Unit tests
+```
+
+## License
+
+MIT

package/skills/short-maker/scripts/google-flow-cli/docs/specs/PROJECT.md

@@ -0,0 +1,12 @@
+# Project Vision: GoogleFlowCli
+
+GoogleFlowCli (gflow) is a streamlined CLI-based workflow and pipeline for generating AI videos systematically, optimizing for cost (credits) and consistency.
+
+## Target Audience
+Video creators, AI filmmakers, and developers needing a structured bulk generation framework with separate audio pipelines.
+
+## Success Metrics
+- Stable, consistent video outputs based on predictable image storyboards.
+- High cost efficiency by validating free image scenes before running 20-credit video generations.
+- Automated, seamless merging of generated high-quality voiceovers using the LucyLab pipeline.
+- Secure token handling and artifact management.

package/skills/short-maker/scripts/google-flow-cli/docs/specs/REQUIREMENTS.md

@@ -0,0 +1,22 @@
+# Requirements
+
+## 1. Image-First Storyboarding
+- Generates zero-cost images for each scene/segment before generating videos.
+- Used to validate prompts, style, character consistency, and scene composition.
+
+## 2. Video Generation Workflow
+- Creates videos strictly upon explicit approval of the image storyboard.
+- Short iteration loops (single segments) favored to avoid batch errors.
+
+## 3. Voiceover (TTS) Pipeline
+- Mandatory use of `lucylab_tts.py` for TTS generation (JSON-RPC).
+- Local macOS TTS (`say`) is strictly forbidden.
+- Requires `LUCYLAB_BEARER` environment variable for token safety.
+
+## 4. A/V Composition
+- Post-process merging of generated video and audio using `ffmpeg`.
+- Standardized output naming convention: `outputs/.../script-full-vo.mp4`.
+
+## 5. Artifact Management
+- Outputs generated sequentially under `outputs/` folder.
+- Voice library caching strictly regulated to `voices_library.json`.

package/skills/short-maker/scripts/google-flow-cli/docs/specs/ROADMAP.md

@@ -0,0 +1,16 @@
+# Roadmap
+
+## Phase 1: Environment & Tooling Verification
+- [ ] Initialize project specifications and Antigravity tracking (Symphony, Brain).
+- [ ] Verify local Python dependencies and `lucylab_tts.py` tool access.
+- [ ] Verify FFmpeg installation and access.
+
+## Phase 2: TTS & Pipeline Groundwork
+- [ ] Establish environment variables integration (`LUCYLAB_BEARER`).
+- [ ] Test compact JSON voice library retrieval.
+- [ ] Test standalone script rendering (`script.txt` -> `outputs/tts-lucylab`).
+
+## Phase 3: Visual & Final Compositing Generation
+- [ ] Image storyboard extraction and generation workflow.
+- [ ] Video segment generation loop.
+- [ ] Master A/V compositing via FFmpeg.

package/skills/short-maker/scripts/google-flow-cli/docs/specs/TECH-SPEC.md

@@ -0,0 +1,13 @@
+# Technical Specifications
+
+## Architecture Decisions
+- **AD-1: Cost Model Gatekeeping**
+  - Image validation step must pre-empt any video API call. Video generation takes place only upon explicit approval.
+- **AD-2: Strict Output Segregation**
+  - Scripts and generated artifacts reside purely within `outputs/`.
+- **AD-3: Modular Voice Library**
+  - Use compact `voices_library.json` catalog containing only `{id,name,description,tag}` for context efficiency.
+
+## Constraints & Security
+- **Config & Tokens**: Bearer tokens are strictly ENV-based (`LUCYLAB_BEARER`). No static token commits or log files exposing tokens in the repo.
+- **Tooling Platform**: Python for automation scripts (`lucylab_tts.py`); `ffmpeg` required as system dependency for A/V composition.

package/skills/short-maker/scripts/google-flow-cli/gflow/__init__.py

@@ -0,0 +1,3 @@
+"""gflow - A command-line interface to Google Flow (AI image & video generation)."""
+
+__version__ = "0.1.0"

package/skills/short-maker/scripts/google-flow-cli/gflow/api/__init__.py

@@ -0,0 +1,19 @@
+"""Google Flow API client."""
+
+from .client import FlowClient, FlowAPIError, FlowRecaptchaError
+from .models import (
+    Asset,
+    AssetType,
+    GenerateImageRequest,
+    GenerateVideoRequest,
+)
+
+__all__ = [
+    "FlowClient",
+    "FlowAPIError",
+    "FlowRecaptchaError",
+    "Asset",
+    "AssetType",
+    "GenerateImageRequest",
+    "GenerateVideoRequest",
+]