OneClickAZ | Marketing Agent Studio

⬜ 1. History Audit

Pending

⬜ 2. AI Agents

Pending

⬜ 3. Visual Lab

Pending

⬜ 4. Publish

Pending

🔴 AUTO POST ARMED

⬆ Up Config

Service Account

No active profile

Drive Account

No active account

Processing...

1. Historical Content Audit

Data Source Settings

Configure the spreadsheet source directly inside Historical Content Audit.

Google account used for Sheets access (Service Account vault — System Admin)

Google Spreadsheet ID

Rows To Fetch

Target Tab

The Spreadsheet ID, selected tab, and row limit are saved locally and mirrored to the legacy global keys for Auto-Run compatibility.

Select Columns to Fetch

Row Filters (Data filter conditions)

No data loaded. Click ▶ Run.

⚙️ Topic Agent

⚙️ Topic Agent Settings

1. Identity & Positioning

Strategy Type *

Brand Name *

Industry Niche *

Agent Expertise

Background & Persona

2. Tone & Strategy

Uncompromising Traits

Target Relationship

Ultimate Goal *

Target Audience

Who the content is for. When filled, the Agent MUST use it verbatim as target_audience. Leave empty to let the Agent deduce one from the Ultimate Goal.

Progression Path

Strategic Focus Matrix

3. Docs & Custom Output

📚 Knowledge Base (KB)

+ Click to upload Brand Soul, Content Strategy, etc.

🔗 Pipeline Inputs (from Previous Agents)

Each enabled row is injected into the agent's prompt under the Goal label. Use this to pass context from previous steps or add custom strategic directives.
History Table — pulls past 7-day topic data from Tab 1 (for deduplication). | Manual — type any custom context, strategy override, or extra instructions for this run.

Agent Source Field Result Goal (inject into prompt)

🧩 Output Structure & Language

Output Language

Single Response No Markdown

Content Visual Output

content_visual_videos Add to Schema

Video Production Brief for Plot, Script & Storyboard agents

content_visual Add to Schema

Static / Social Design Brief for downstream design agents

Output Schema (JSON)

🔵 Topic Agent Output

① Json Output

② Parsed Schema Fields (Editable)

Run Topic Agent to generate fields...

③ Parsed Schema Custom

Chưa dựng — chạy Topic Agent để field tự lên theo khoá bản ghi thật.

📝 Copywriter Agent Output

⚙️ Copywriter Agent Settings

1. Writer & Tone

Brand Name *

🔗 Auto from Topic Agent

Auto

Writer Expertise *

Content Objectives *

Ultimate Goal (Optional)

Tone of Voice *

2. Format & Language

Content Length (Preset)

Hashtag

3. CTA & Forbidden Terms

CTA Instruction (optional)

Forbidden Terms

📚 Knowledge Base (KB)

KB Documents (.md / .txt)

+ Click to upload Brand Voice, Copy Frameworks, etc.

✍️ Signature — appended to end of all output (optional)

😊 Icon / Emoji — add emoji icons to output (optional)

🧩 Output Structure & Language Default: Content + Hashtag

Output Language

🇬🇧 English

🇻🇳 Vietnamese

🇨🇳 Chinese

🇪🇸 Spanish

🇸🇦 Arabic

Output Schema (JSON)

🔗 Pipeline Inputs (from Previous Agents)

Agent Source Field Result Goal (inject into prompt)

① Json Output

② Parsed Schema Fields (Editable)

No content generated yet. Run the agent to see results.

③ Parsed Schema Custom

Chưa dựng — chạy Copywriter Agent để field tự lên theo khoá bản ghi thật.

🎬 Video Script Agent

⚙️ Video Script Agent Settings

1. Writer & Tone

Brand Name

🔗 Auto

Auto

Writer Expertise (Optional)

If left empty, the system's default configuration will be used to ensure output quality.

Content Objectives *

Tone of Voice

CTA Instruction (optional)

Forbidden Terms

2. Video Specs

Video Length (Preset)

Scripts / Run

max 5

Content Framework 🎲 Monologue / Dialogue auto-apply the mode framework library · Custom = your own framework

Script Mode:

⏱ Sentence timing – s / line — over-limit auto-splits to the next character line (Dialogue/Custom only)

Monologue / Dialogue automatically apply the system framework library for that mode · Custom: enter your own framework below

Custom Framework (Custom mode only)

📚 Knowledge Base (KB)

+ Upload Brand Handbook, Visual Standards, etc.

3. Script Signature

🎙️ TTS Optimization — clean text for AI Voice generators

• Remove all Markdown: * - # "" `

• Preserve sentence breaks: . , ? !

• Sentence timing: set per Script Mode — see ⏱ in Content Framework (Dialogue/Custom)

✍️ Script Signature — appended to end of all scripts

🧩 Output Structure & Language Default: TTS-clean JSON

Output Format

JSON Mode (structured keys — only supported format)

Output Language

🇬🇧 English

🇻🇳 Vietnamese

🇨🇳 Chinese

🇪🇸 Spanish

🇸🇦 Arabic

Custom language:

Output Schema (JSON)

—

🔗 Pipeline Inputs (from Previous Agents)

Agent Source Field Result Goal (inject into prompt)

① Json Output

② Parsed Schema Fields (Editable)

All output fields will appear here after running.

③ Parsed Schema Custom — TTS ready

Hook + Script Content (no labels) will appear here.

📖 Brief Story Videos Agent

⚙️ Story Agent Settings

A. Brand Identity

Brand Name

Expert Screenwriter & Visual Storyteller transforming video content scripts into Brief story entries for social media video production for OneClickAZ.

Content Objectives (Optional)

For standalone use (without Video Script Agent). If filled, this content is added to Pipeline Inputs as supplementary guidance. If left empty, the agent relies entirely on Pipeline Inputs below.

B. Core Video Style — select 1 for fixed, 2+ for random per run

🎬 Scene Count & Duration Add to prompt

Min scenes:

Max scenes:

Voice-over/scene: ~ sec

AI produces AT LEAST Min scenes (never pads with empty entries if the real voice-over is shorter). Max scenes is an optional cap. Each Voice-over reads aloud within the time range.

Storyboard Direction Objective (auto-fills on style select, editable)

C. Master Character

1 character = fixed in prompt. 2+ = AI randomly selects one based on content & market context.

🔗 Pipeline Inputs (from Previous Agents)

Agent SourceFieldResultGoal (inject into prompt)

📚 Knowledge Base (KB)

Drop .md/.txt or browse

🧩 Output Structure & Language

Output Format

JSON — TTS-clean {"Brief_Video_Story": "..."} Raw Text — plain string, no JSON wrapper

Output Language

🇬🇧 English

🇻🇳 Vietnamese

🇨🇳 Chinese

🇪🇸 Spanish

🇸🇦 Arabic

🇰🇭 Khmer

🇹🇭 Thai

🇳🇬 Yoruba

Custom:

Toàn bộ nội dung (Voice-over + Visual) ra theo ngôn ngữ này — chỉ label cấu trúc giữ tiếng Anh.

Output Schema (JSON)

story_framework — Scriptwriting Framework

story_character — Master Character

Brief_Video_Story — Story Script (full scene entries)

story_objectives — Meaning & Message Objectives

① Json Output

② Parsed Schema Fields (Editable)

Output components will appear here after running.

③ Parsed Schema Custom

Chưa dựng — chạy Story Agent để field tự lên theo khoá bản ghi thật.

Parsed Scene Entries (dẫn xuất, chỉ đọc)

Scene entries will appear here after running.

🎬 Storyboard Agent

⚙️ Storyboard Agent Settings

System Role: You are a Cinematic Visual Director & Strategic Storyboard Expert (Universal Storyboard Director). Your mission: receive story scripts from ANY industry, analyze and transform every concept (complex, abstract, or technical) into vivid, physically-realistic Prompt commands (Visual Briefs) for AI image/video generation tools (Midjourney, Flux, Kling, Veo...).

Core Identity & Industry

Core Industry {{CORE_INDUSTRY}}

AI auto-selects Props matching industry (Vest/Finance, Apron/F&B...)

Brand Vibe {{BRAND_VIBE}}

Brand Name

Used for brand asset tag injection in prompts

Visual & Audio Parameters

🖼 Image Style Lock {{IMAGE_STYLE_LOCK}} Add to prompt

Appended to END of every Prompt Imgs line.

🎨 Background Lock {{BACKGROUND_LOCK}} Add to prompt

AI locks this background CONSISTENTLY across ALL scenes.

🎥 Camera Control {{CAMERA_CONTROL}} Add to prompt

When enabled, this directive takes PRIORITY in Prompt Videos for EVERY scene — overrides Camera Angle Logic selection.

👤 Master Character Override {{MASTER_CHARACTER}} Add to prompt

When enabled: AI uses EXACTLY this description as Master Character for ALL scenes — does not create or modify the character.

Voice Tone Profile

Select one or more. Multiple selections: AI picks randomly per scene based on character demographics.

🇺🇸 EN Male — Adult male, 28-35 yrs, neutral American English accent, confident and authoritative tone, moderate-to-fast pace, professional and trustworthy delivery 🇺🇸 EN Female — Adult female, early 20s-30s, warm and clear American English accent, friendly and engaging persona, upbeat energetic tone, natural conversational flow

✎ Custom voice profiles (can add multiple)

🎬 Scene Count & Duration Add to prompt

Min scenes:

Max scenes:

Voice-over/scene: ~ sec

If a pipeline script is provided, its real scene count is used (capped to Max if set). Otherwise AI produces within Min–Max. Each Voice-over reads aloud within the time range.

📚 Knowledge Base (KB)

KB Documents (.MD / .TXT) + Upload Brand Handbook, Visual Standards, etc.

Suggested: Brand Handbook (colors, audience, character casting) · Visual Standards (asset lists, UI specs)

🧩 Output Structure & Language

Scene Output Style — drives Scene Generation module suggestion

B — Image + Motion prompt → Module 2 (Combo Image → Video) A — Video prompt only → Module 1 (Text to Video) C — Image Start/End + Motion → Module 3 (First+Last Frame)

Parsed Display Components

Subject Block [Subject Block:]
Master Character anchor — verbatim repeat every scene Prompt Imgs [Start_Frame]
Camera + EN static description + Setting + Lighting + Style Lock Prompt Imgs (End_Frame) [End_Frame] Style C
System component — end state AFTER the scene motion completes (first/last-frame video) Prompt Videos [Motion_&_Result]
1 action trajectory (5-8s) · Camera motion + Subject → End state Voice-over [Voice-over:]
Verbatim script copy + [Character said] or [Background Chatter] Tone Voice [Tone Voice:]
English TTS profile · gender, accent, energy, pace

Custom Fields

Add extra output lines after Tone Voice. AI fills each per scene. Example: Hashtags → "Write 3 hashtags for this scene".

Output Language {{OUTPUT_LANG}}

Language for Voice-over dialogue. Structural labels (Subject Block, Scene, Prompt Imgs...) always English.

Output Format

Plain Text — Subject Block: ... 1. Scene 1... JSON Array — [{scene_number, prompt_image...}]

JSON: backend parses into 3 streams (Image / Video / Audio TTS)

🔗 Pipeline Inputs (from Previous Agents)

Agent SourceFieldResultGoal (inject into prompt)

🏷️ Asset Brand Logic — Injection Rules Map trigger keywords to image asset tags. Tagged phrases injected verbatim into Prompt Imgs.

Add to prompt –

✓ Trigger Keyword / Condition Asset Type Inject Tag (verbatim into Prompt Imgs)

Uncheck ✓ to disable a rule. No Brand rule appended to scenes without brand mention when enabled.

① Json Output

② Parsed Schema Fields (Editable)

No content parsed yet. Click Parse JSON after running.

③ Parsed Schema Custom

No content parsed yet. Click Parse JSON after running.

📺 YouTube SEO & Metadata Agent

⚙️ YouTube SEO Agent Settings

A. Channel & Brand Identity

Channel Name {{CHANNEL_NAME}}

Brand / Product Name {{BRAND_NAME}}

Industry / Niche {{CHANNEL_NICHE}}

Title Format Pattern {{TITLE_FORMAT}}

If left empty, the system's default configuration will be used to ensure output quality.

Title length: – chars

B. SEO Strategy & Visual Parameters

Language / Market {{SEO_LANG}}

Keyword Strategy {{KEYWORD_STRATEGY}} (multi-select — AI picks randomly if multiple)

Long-tail High-Intent ×

SEO description length: – chars

📚 Knowledge Base (KB)

KB Documents (.MD / .TXT) + Upload Channel Handbook, SEO Style Guide, etc.

Suggested: Channel Handbook (audience, tone, brand voice) · SEO Style Guide (keyword rules, title patterns, banned phrases)

🧩 Output Structure & Language

Output Language

English (US) ×

Multiple languages → AI randomly picks one per run.

Output Schema (JSON)

Each key is a field in the AI's JSON response. Uncheck to exclude from output.

Output Format

JSON Object — {"yt_title":"..."} Plain Text — yt_title: ...

JSON: backend auto-fills Title, Filename, Description, Thumbnail fields.

🔗 Pipeline Inputs (from Previous Agents)

Each enabled row is injected into this agent's prompt. Pull from Topic, Video Script, or Story Agent results — or add Manual context.

Agent Source Field Result Goal (inject into prompt)

E. Description Template

Opening Style {{OPENING_STYLE}}

If left empty, the system's default configuration will be used to ensure output quality.

CTA Text {{CTA_TEXT}} Add to prompt

Links Block {{LINKS_BLOCK}} Add to prompt

Appended at the bottom of every SEO description.

# Hashtag Configuration

Mode:

Count range: – hashtags

① Json Output

② Parsed Schema Fields (Editable)

YouTube Title (max 110 chars)

0 / 110

Filename (ASCII safe)

SEO Description

③ Parsed Schema Custom

Chưa dựng — chạy YouTube SEO Agent để field tự lên theo khoá bản ghi thật.

Post-Processing Content Agents (Signature / Reformat / Append)

🖼 Banner Social Prompt Agent

Visual AI Director

⚙️ Banner Social Prompt Agent Settings

📐 Visual Identity & Style

🧍 Brand Character Add to prompt

Locked across every run — kept identical in every generated banner so the same brand always shows the same character. Leave empty (or untick Add to prompt) if the brand uses no character.

🏭 Industry Preset — auto-fills Style, Typography & Background

🎨 Main Design Style Add to prompt

🔤 Typography Rules Add to prompt

🖼️ Background Rules Add to prompt

📐 Image Layout Composition Add to prompt

👆 CTA on Image Add to prompt

🏷️ Overlay & Output

💬 Slogan / Watermark On

If filled, AI will mandatorily overlay this text on the image.

🔵 Brand Logo Integration

When enabled: AI places the Logo at the position below.

📐 Aspect Ratio

Output Schema (JSON)

Output Language

Single Response No Markdown

📚 Knowledge Base (KB)

Upload reference documents (.md / .txt). System wraps them in <reference> tags when sending to AI.

📎 Click or drag & drop .md / .txt

e.g. Visual Standards, Brand Handbook

🔗 Pipeline Inputs (from Previous Agents)

Agent Source Field Result Goal (inject into prompt)

🏗️ Banner Template Fields (Output Structure · relabel per industry)

from: topic + hook ✎

from: sub_topic ✎

→ Module 1 Typo ✎

from: geo + audience ✎

from: assets + type ✎

→ Module 1 Style ✎

→ Module 2 Slogan ✎

⑤ Prompt Output

No results yet. Run Visual Agent to see output cards.

③ Parsed Schema Custom

Chưa dựng — chạy Banner Social Prompt Agent để field tự lên theo khoá bản ghi thật.

🎬 Thumbnail Prompt Agent

YouTube Thumbnail

⚙️ Thumbnail Prompt Agent Settings

🎨 Visual Style & Composition

🧍 Brand Character Add to prompt

Locked across every run — kept identical in every generated thumbnail so the same brand always shows the same character. Leave empty (or untick Add to prompt) if the brand uses no character.

🎨 Thumbnail Style Add to prompt

🔤 Text Overlay Rules Add to prompt

🖼️ Background & Scene Add to prompt

📐 Image Layout Composition Add to prompt

👆 CTA on Image Add to prompt

💬 Slogan / Watermark On

If filled, AI will mandatorily overlay this text on the thumbnail.

🔵 Brand Logo Integration

When enabled: AI places the Logo at the position below.

🧩 Output Structure & Language

📐 Aspect Ratio

Output Schema (JSON)

Output Language

Single Response No Markdown

📚 Knowledge Base (KB)

Reference documents for the thumbnail prompt (.md / .txt) — also filled by Fast Setup's KB step.

📎 Click or drag & drop .md / .txt

🔗 Pipeline Inputs (from Previous Agents)

Agent Source Field Result Goal (inject into prompt)

🎬 Thumbnail Prompt Output

No results yet. Run Thumbnail Agent to see output.

③ Parsed Schema Custom

Chưa dựng — chạy Thumbnail Prompt Agent để field tự lên theo khoá bản ghi thật.

3. Visual Lab

📁

Shared Drive Folder

No Folder Created

This module owns the Google Drive root folder setting, creates the project folder (named after the project) and, on every run, one run folder inside it where the production agents save their results.

Google Drive Root Folder URL

This root folder will be used as the parent for all generated shared folders.

No root folder configured.

🔗 Pipeline Inputs (from Previous Agents) — run folder base name

AgentFieldValue (live)Purpose

The first ticked row names this run's folder (<time>-<value>). Only your Agent/Field choice is saved — the value is pulled fresh from each run's results. Purpose is never sent to any AI prompt.

Open Folder

Agent Audio (Multi Agent TTS)

Each Audio Agent has its own runtime, pipeline input, and output schema

No Audio Agents yet. Click "+ Add Audio Agent" to add one.

🎨 Visual Generation Agents

Fal.ai / API

No Image Agents yet. Click "+ Add Image Agent" to add one.

🎥 Video Generation Agents

Fal.ai / API

No Video Agents yet. Click "+ Add Video Agent" to add one.

🎞 Scene Generation Agents

Storyboard → Media

⚙ Master Config (All Scenes)

Storyboard / Manual Prompt — paste scenes text or pull from Storyboard, then Parse

☁ Drive Folder — required before Run Not linked

Open

Auto-detects the Shared Drive Folder from Visual Generation. Each Run auto-creates its own dated scene folder inside Scene Videos/ — no manual folder creation needed.

🖼 Image Reference URLs max 10

🖼 End Frame Reference URLs Module 3 · max 10

🎬 Video Reference URLs Module 1 & 4 · max 10

Models & Media API (all scenes) — pick the API first, the model list syncs to it

Aspect Ratio (all scenes)

🖼 Image Params (all scenes)

🎬 Video Params (all scenes)

No Scene Agents yet. Run the Storyboard Agent, then click "⚡ Generate from Storyboard" — or add one manually.

🎬 Video Composer

Scenes → Final .mp4

Merge all completed scenes into a single video file, upload it to Drive, and optionally write the URL to Google Sheets.

1 · RESOURCES — declare where every asset comes from; position is set in section 2

Video / Image — three ways in, one timeline

1 · Upload from computer — click to choose, or drop files here

Each file is uploaded to Google Drive right away; only its Drive link is stored in this project — never the file itself.

2 · Paste Drive links — one per line, added to the timeline automatically

3 · Pipeline input — pull a result from a previous agent, then add it to the timeline

Thumbnail — optional still image shown first; its duration and audio live in section 2 1 · Pipeline input

2 · Link Auto-pick the thumbnail when this field is still empty

Audio · Voice Auto-detected

1 · Pipeline input

2 · Link

or upload a file — it becomes a Drive link and lands in the field above

Auto karaoke subtitles (AI Whisper) — not supported yet

Volume 100%

Auto-detect the voice track from Agent Audio

Audio · Background Music

1 · Pipeline input

2 · Link

or upload a file — it becomes a Drive link and lands in the field above

Volume 30%

Auto fade-out at the end seconds

3 · LAYOUT & RENDER — name, destination, frame and how the clips are joined

1 · Video name

.mp4

Pull name from a previous agent

2 · Drive upload folder — read-only; this run's folder created by the Drive module

Pull folder from the Drive module

3 · Template frame 4 · Output resolution (target frame size) 5 · Output aspect ratio 6 · Transition between clips

Live View — updates in realtime while you edit; the finished video plays here after Compose.

2 · TIMELINE SETTINGS — drag a row to set the order; each row's controls depend on what it is

0 items

Drag a row to reorder · ↑/↓ also work · × removes a row · Fal URLs expire in ~10 min

Each row, line 1: ⠿ drag · # · preview · name · slot type · source · ⏱ duration in seconds (locked for video — the clip decides it) · ⏩ speed · audio. Line 2: 🔗 the clip's Drive link, full width. ⏩ Speed is applied to the render — picture via setpts, original audio via atempo (pitch preserved). A 10s clip at 1.25x renders as 8s.

Per-clip audio overrides are active — the voice track narrates ONLY the rows marked "TTS narration" in this timeline, in order (it pauses over other rows and resumes where it left off); rows marked "Keep own audio" keep their original sound.

0. img Thumbnail FIXED 🔗 first frame — its link is declared in section 1 ⏱ ⏩ —

✓ Apply saves this timeline and these settings as the project template. Without Apply the edits affect only the run you start now.

● Queued...

4. Auto Post Social — Multi-channel Publishing

Add Social Provider

Display Name *

Platform *

API Service *

API Key *

Stored in the backend vault (gitignored). Only the last 4 characters are shown later.

Not tested

Account (page/profile to post as)

Detected via Test Connection. Pick one before using this provider.

📱 Auto Post — 1️⃣ Source → 2️⃣ Channels → 3️⃣ Content → 4️⃣ Schedule & Post

Source sheet

Select a source sheet to load its columns.

Filter column

Filter value

2️⃣ CHANNELS — pick where to post

Loading providers...

Content mapping (legacy)

Caption →

Hashtags →

Image URL →

Video URL → (from Video Composer)

Schedule (legacy)

Post now Schedule: Max rows:

📋 POST QUEUE

Post queue is empty.

Post content preview

No preview yet.

5. Google Sheets Targets

Configure one or more Google Sheets below. For each sheet, scan the columns and map them to the desired Agent Output data.

Result Sent to Google Sheets

📁 Project Manager

Saved projects · schedules · run history

Credentials

System-wide store for connected accounts and 3rd-party APIs — these credentials are shared by every project in Content Studio.

🔑 All Credentials

	Name	Type	Source	Note	Live Links	Actions
Loading credentials…

Page 1 of 1

⚙️ API Credentials

LLM API Key (DeepSeek)

LLM Base URL

LLM Runtime

Language model API keys (DeepSeek, OpenAI, Gemini, custom) — used to run every text-generation agent.

No active model

No agent overrides assigned.

Audio Runtime API

Text-to-speech providers (Vbee, ElevenLabs) — used by the Video Composer for narration.

No active Audio Runtime

Media API

Image/video generation API (Fal.ai) — powers Image Agent, Scene Gen, Banner/Thumbnail.

No active Media API

Fal Relay URL (advanced — only if your network blocks fal.run)

Deploy docs/tools/fal-relay-worker.js to Cloudflare Workers, paste its URL here. Applies instantly (no Save needed).

🗄 Storage Bridge (ImgBB)

Image relay — turns local files / transient URLs into stable public URLs for Fal.ai. Used by: Banner/Thumbnail reference upload, Image Agent edit-model references, Scene Gen 📤 local-file upload. Saved with Save All Credentials.

ImgBB API Key Storage Bridge

Used to upload temporary public URLs for media workflows.

Google Service Account JSON 🔑

🔗 Google Account Connect

One place for every Google connection — Service Account (share-based) or OAuth2 (personal account, recommended for Drive uploads).

No account selected.

Google Service Account

Vault-based setup for Sheets access. Active profile mirrors to legacy runtime key.

No active Service Account profile

Google Drive Folder URL 📁

Paste the link of the shared Drive folder (e.g. 01.ONECLICKAZ AI DESIGN)

🔗 DRIVE OAUTH2 (RECOMMENDED)

Not Connected

Use OAuth2 with personal Google accounts to avoid storage quota bottlenecks during uploads.

No account selected.

Callback URL:
http://localhost:3931/api/oauth/drive/callback

📱 SOCIAL POST APIs

3rd-party posting APIs (Blotato — more platforms later). Keys are stored on the backend only, always shown masked, and are consumed by the Auto Post module per project.

Loading providers...

🤖 AI Agent Control (MCP) off

Enable Agent Bridge — THIS tab executes commands sent from Claude Desktop / Claude Code (keep the tab open).

claude mcp add oneclickaz -- node backend/mcp/oneclickaz-mcp.js

📖 AI operation guide (what Claude can do)

1. Historical Content Audit

Data Source Settings

⚙️ Topic Agent

1. Identity & Positioning

2. Tone & Strategy

3. Docs & Custom Output

🧩 Output Structure & Language

🔵 Topic Agent Output

📝 Copywriter Agent Output

1. Writer & Tone

2. Format & Language

3. CTA & Forbidden Terms

📚 Knowledge Base (KB)

🧩 Output Structure & Language Default: Content + Hashtag

🎬 Video Script Agent

1. Writer & Tone

2. Video Specs

3. Script Signature

🧩 Output Structure & Language Default: TTS-clean JSON

📖 Brief Story Videos Agent

A. Brand Identity

B. Core Video Style — select 1 for fixed, 2+ for random per run

C. Master Character

🔗 Pipeline Inputs (from Previous Agents)

📚 Knowledge Base (KB)

🧩 Output Structure & Language

🎬 Storyboard Agent

Core Identity & Industry

Visual & Audio Parameters

📚 Knowledge Base (KB)

🧩 Output Structure & Language

🔗 Pipeline Inputs (from Previous Agents)

📺 YouTube SEO & Metadata Agent

A. Channel & Brand Identity

B. SEO Strategy & Visual Parameters

📚 Knowledge Base (KB)

🧩 Output Structure & Language

E. Description Template

Post-Processing Content Agents (Signature / Reformat / Append)

🖼 Banner Social Prompt Agent

📐 Visual Identity & Style

🏷️ Overlay & Output

📚 Knowledge Base (KB)

🎬 Thumbnail Prompt Agent

🎨 Visual Style & Composition

🧩 Output Structure & Language

📚 Knowledge Base (KB)

3. Visual Lab

Shared Drive Folder

Agent Audio (Multi Agent TTS)

🎨 Visual Generation Agents

🎥 Video Generation Agents

🎞 Scene Generation Agents

⚙ Master Config (All Scenes)

🎬 Video Composer

4. Auto Post Social — Multi-channel Publishing

Add Social Provider

📱 Auto Post — 1️⃣ Source → 2️⃣ Channels → 3️⃣ Content → 4️⃣ Schedule & Post

Post content preview

5. Google Sheets Targets

Result Sent to Google Sheets

📁 Project Manager

Credentials

🔑 All Credentials

⚙️ API Credentials

LLM Runtime

Audio Runtime API

Media API

🗄 Storage Bridge (ImgBB)

🔗 Google Account Connect

Google Service Account

🔗 DRIVE OAUTH2 (RECOMMENDED)

📱 SOCIAL POST APIs

👁 Agent — Full System Prompt Preview

📖 Agent — Guide To Use

Historical Content Audit Guide Hướng Dẫn Historical Content Audit

👁 Storyboard Agent — Full System Prompt Preview

Setup LLM API

Saved APIs

Assign Active LLM To Agents