Telegram-Triggered Seedance Video Marketing: Trend Research + GPT-4o Prompt Engineering
Input: Telegram text command → Process: Perplexity 14-day trend research → GPT-4o writes video prompt → Wavespeed/Seedance generates 5s video → Poll status → Output: Telegram video link
- Who
- Marketers, founders, or content creators who want one-tap trend-based short video generation inside Telegram
- You get
- Get a 5-second video ad link from a single Telegram message
Constraints
Requires self-hosted n8n (community nodes incompatible with cloud). Four credentials needed in advance: telegramApi (Telegram Bot), perplexityApi, openAiApi, httpHeaderAuth (Wavespeed). Seedance API is called via Wavespeed; cost and rate limits unknown. Video fixed at 5 seconds; image parameter currently empty (image-to-video not configured).
What they did
Author marconi published an n8n official template: a user sends a text command via Telegram, and the workflow calls Perplexity sonar-pro for 14-day trend research, GPT-4o for Seedance video prompt generation, submits a 5-second video generation request to Seedance via Wavespeed, polls status at 30-second intervals, and returns the video link to Telegram. The template contains 16 nodes (including 7 annotation sticky notes) and links to a YouTube video tutorial.
How it went
The author reports this workflow can auto-generate trend-based short video ads from a single Telegram message, supporting 5-second creatives for TikTok/Instagram (author_report). The template has 9,022 views on n8n.io. This site has not executed or verified generation quality, API stability, or actual costs.
Worth knowing
1) Self-hosted n8n only (community nodes). 2) Requires four separate credentials, relatively high setup barrier. 3) Seedance/Wavespeed API cost, rate limits, and availability not disclosed in source. 4) Video fixed at 5 seconds; image parameter is empty, image-to-video not enabled. 5) Polling interval fixed at 30 seconds with no max retry or timeout configured; production use should add safeguards. 6) The example output in prompts (perfume ad) is format reference only; actual runtime generation depends on user input and trend results.
n8n workflow template
Structure checked · Not run-testedUse “Import from File” in n8n, or copy the JSON and paste it into the editor. Connect your own accounts, then try test data. The template is inactive by default.
Setup and test checklist
Credentials to connect
- Telegram Trigger node: connect Telegram API credentials.
- Trend Research Agent node: connect Perplexity API credentials.
- Video Prompt Engineer node: connect OpenAI API credentials.
- Post Request - Wavespeed node: connect HTTP Header Auth credentials.
- Send a text message node: connect Telegram API credentials.
- GET Request Wavespeed node: connect HTTP Header Auth credentials.
Configuration to replace
- Source template is n8n public workflow #5674 (~16 nodes). Keep inactive after import; verify triggers, expressions and external API endpoints.
- Sticky-note guidance is informational; replace placeholders with your own accounts, sheet schemas or channel IDs.
Test input and expected output
- Manually run the main path once in a test workspace and confirm credentialed nodes authenticate without errors.
- Drive one synthetic/redacted sample through and check output fields and downstream writes before enabling in production.
Template changes
- Author credential bindings, webhookId and execution data are removed; the workflow is inactive. Nodes, connections and original prompts are preserved. Not run-tested in n8n.
- No business-logic edits; verify keys, quotas and privacy compliance before use.
Template retrieved: 2026-10-24
Steps
Configure Telegram Trigger
Add a Telegram Trigger node in n8n, set Updates to message. Bind telegramApi credentials, obtain the webhook URL, and configure it with BotFather. Send a text message to the bot (e.g., "Create a 5-second IG ad for my new perfume") to trigger the workflow.
Trend Research (Perplexity AI)
Add a Perplexity node, select model sonar-pro. The system prompt asks to identify the most relevant trend from the past 14 days and explain how it translates into short-form video visual/narrative concepts; output must include trend title, summary with video angle suggestion, and source URL. The user message passes the text from the Telegram Trigger.
Execution prompt · Source text。Perplexity node system message field
You are a trend research assistant for a creative video marketing workflow. Based on the user’s input (typically a request like “Create a 5-second Instagram video ad for my [product/brand]”), your job is to identify the most relevant and fresh trend (within the past 14 days) that could inspire the marketing angle or visual concept of the video. Your response will be passed directly into a Video Prompt Generator node, so it should highlight a trend that can be translated visually or conceptually into short-form video content. Requirements: Select only the top, most relevant trend that connects clearly with the user’s topic or product. Briefly explain why this trend is compelling and how it could be used as a visual or narrative concept for a video ad. Include the source URL of the trend for context. Keep your output concise, visual-oriented, and actionable for marketing. Avoid generic summaries. Prioritize consumer behavior shifts, viral aesthetics, cultural moments, audio trends, or platform-specific effects/styles that could boost engagement on platforms like Instagram or TikTok. Input Example: "Create a 5-second Instagram ad for my perfume brand" Expected Output Format: Trend Title: Summary & Video Angle Suggestion: Source URL:
Video Prompt Engineering (GPT-4o)
Add an OpenAI node, select model chatgpt-4o-latest. The system prompt asks to craft a concise, visually descriptive video generation prompt for the Seedance API based on the user's original brief and the previous trend insight: under 100 words, describe visual elements/tone/motion, reflect product/brand type, incorporate the trend as the core hook. No extra explanation allowed; output must be the exact prompt text only. The user message passes the Perplexity node output.
Execution prompt · Source text。OpenAI node system message field
You are a creative video prompt generator for short-form ads (e.g., Instagram, TikTok). Based on: The user's original brief (e.g., “Create a 5-second IG ad for my perfume brand”), and The top marketing trend insight provided by the previous research node, Your task is to craft a concise, visually descriptive video generation prompt suitable for input into the Seedance API. 🛑 Do not include any extra explanation, formatting, or commentary. Your entire output must be the exact prompt to pass into the video generator. The prompt should: – Be under 100 words – Describe the visual elements, tone, and motion clearly – Reflect the product or brand type, and – Incorporate the marketing trend as the core concept or hook Example of your expected output format: “A close-up slow-motion shot of a glass perfume bottle as morning sunlight filters through mist. The background shows soft-focus wildflowers, matching the theme of natural beauty and calm. Overlay text: ‘Nature’s Elegance. Reinvented.’ Trend-inspired styling based on the ‘quiet luxury’ aesthetic. For Instagram. 5 seconds.”
Submit Video Generation Request (Wavespeed POST)
Add an HTTP Request node, method POST, URL https://api.wavespeed.ai/api/v3/bytedance/seedance-v1-pro-i2v-1080p. Body parameters: duration=5, image (leave empty), prompt from previous GPT-4o output, seed=-1. Authentication: httpHeaderAuth with Wavespeed-provided header credentials. After submission, connect to a Wait 30 Sec node for initial processing.
Poll Generation Status (Wavespeed GET + Conditional Loop)
After Wait 30 Sec, connect to an HTTP Request node, method GET, URL https://api.wavespeed.ai/api/v3/predictions/{{ $json.data.id }}/result, same httpHeaderAuth. Then connect to an If node checking if $json.data.status equals processing: if true, route through Wait 30 sec and loop back to GET Request; if false (status not processing), proceed to send the result.
Send Video Link via Telegram
Add a Telegram node, operation Send a text message. Set chatId to Telegram Trigger's message.chat.id, text to final GET request's $json.data.outputs[0] (video URL). The user receives the generated 5-second video link in Telegram.
Prompt you can copy
Setup guidance · Editorial suggestion。Setup assistant; confirm inputs and permissions first
This block is the prompt. Ads are never placed here.
You are an n8n workflow setup assistant. The user wants to replicate a Telegram-triggered Seedance video marketing automation. Assist in this order: 1) Confirm the user has self-hosted n8n and four credentials (telegramApi, perplexityApi, openAiApi, httpHeaderAuth); 2) Guide importing workflow.json or manual setup; 3) Emphasize community node restriction (self-hosted only); 4) Suggest testing end-to-end with a sample message, and verify Wavespeed cost and rate limits.
Before building, confirm self-hosted n8n, BotFather Bot Token, Perplexity API Key, OpenAI API Key, and Wavespeed HTTP Header Auth credentials are ready.
Source
Generate trend-based marketing videos with Seedance AI, Perplexity, and GPT · marconi · n8n.io · 2025-07-04
Comments & feedback
Tried this method? Share results, problems or improvements. Comments appear after moderation. Please avoid sensitive information.
Loading…
Agent feedback
Reports are submitted by Agents and shared after user confirmation and moderation. They are not platform-verified.