All templates By NappAI team Automatic YouTube Video Post
Automation that transforms YouTube videos into social media posts, including a summary of the content and AI-generated image.
+2
graph TD
%%{init: {'theme': 'mc','layout': 'elk'}}%%
CombineText-atea2[Combine Text]
style CombineText-atea2 stroke:#a170ff
ChatOutput-l0kif[<div><img alt="logo" src="/_astro/messages-square.BaSDmT6g.svg" style="height: 20px !important;width: 20px !important"/></div>Chat Output]
style ChatOutput-l0kif stroke:#a170ff
TextInput-m2egg[<div><img alt="logo" src="/_astro/type.Dy26vmDy.svg" style="height: 20px !important;width: 20px !important"/></div>VIDEO URL]
style TextInput-m2egg stroke:#a170ff
Prompt-wss2e[<div><img alt="logo" src="/_astro/square-terminal.BMOXc-nZ.svg" style="height: 20px !important;width: 20px !important"/></div>Blog Summary]
style Prompt-wss2e stroke:#a170ff
ParseData-mh1rk[<div><img alt="logo" src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>Extract transcript]
style ParseData-mh1rk stroke:#a170ff
OpenAIModel-v6p19[<div><img alt="logo" src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>OpenAI]
style OpenAIModel-v6p19 stroke:#a170ff
youtube_transcripts-wdsaq[YouTube Transcripts]
style youtube_transcripts-wdsaq stroke:#a170ff
Prompt-ek8pp[<div><img alt="logo" src="/_astro/square-terminal.BMOXc-nZ.svg" style="height: 20px !important;width: 20px !important"/></div>Image Prompt]
style Prompt-ek8pp stroke:#a170ff
DallEImageGenerator-tuo20[<div><img alt="logo" src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>DALL·E Image Generator]
style DallEImageGenerator-tuo20 stroke:#a170ff
JsCode-j6muq[Uri Builder]
style JsCode-j6muq stroke:#a170ff
ParseData-e7as5[<div><img alt="logo" src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>Parse]
style ParseData-e7as5 stroke:#a170ff
CombineText-atea2 -.- ChatOutput-l0kif
linkStyle 0 stroke:#a170ff
TextInput-m2egg -.- Prompt-wss2e
linkStyle 1 stroke:#a170ff
ParseData-mh1rk -.- Prompt-wss2e
linkStyle 2 stroke:#a170ff
Prompt-wss2e -.- OpenAIModel-v6p19
linkStyle 3 stroke:#a170ff
OpenAIModel-v6p19 -.- CombineText-atea2
linkStyle 4 stroke:#a170ff
TextInput-m2egg -.- youtube_transcripts-wdsaq
linkStyle 5 stroke:#a170ff
youtube_transcripts-wdsaq -.- ParseData-mh1rk
linkStyle 6 stroke:#a170ff
Prompt-ek8pp -.- DallEImageGenerator-tuo20
linkStyle 7 stroke:#a170ff
DallEImageGenerator-tuo20 -.- JsCode-j6muq
linkStyle 8 stroke:#a170ff
OpenAIModel-v6p19 -.- Prompt-ek8pp
linkStyle 9 stroke:#a170ff
JsCode-j6muq -.- ParseData-e7as5
linkStyle 10 stroke:#a170ff
ParseData-e7as5 -.- CombineText-atea2
linkStyle 11 stroke:#a170ff
Automatic YouTube Video Post
🧩 Overview
This workflow converts a YouTube video URL into a ready‑to‑publish social‑media post. It automatically extracts the video transcript, generates a concise summary and a compelling post, creates an accompanying image with AI, and assembles the final content for display in the Playground.
⚙️ Main Features
- Captures a YouTube URL and pulls the full transcript.
- Transforms raw transcript data into plain text for downstream processing.
- Uses GPT‑4o‑mini to produce a summary and a social‑media post with emojis and a call‑to‑action.
- Generates an image prompt from the summary and creates a vivid 1024×1024 image via DALL‑E 3.
- Converts the image to a Markdown‑formatted Base64 string.
- Concatenates text and image into a single message.
- Displays the complete post in the Playground chat interface.
🔄 Workflow Steps
| Component Name | Role in the Workflow | Key Inputs | Key Outputs |
|---|---|---|---|
| Video URL | Entry point for the user to provide a YouTube link. | URL string | Text |
| YouTube Transcripts | Retrieves the spoken content of the video. | URL string | Data (raw transcripts) |
| Parse (Extract Transcript) | Converts raw transcript data into plain text. | Data (transcripts) | Text (video transcription) |
| Blog Summary | Builds a prompt containing the transcription and the source URL. | Video transcription, Source URL | Prompt message |
| OpenAI Model | Generates a summary and a social‑media post. | Prompt message | Text (summary + post) |
| Image Prompt | Creates an image prompt based on the article summary. | Article summary (summary output) | Prompt message |
| DALL·E Image Generator | Produces a 1024×1024 vivid image from the prompt. | Image prompt | Binary (image data) |
| Uri Builder | Transforms the binary image into a Markdown‑encoded string. | Binary image | Markdown string |
| Parse (Markdown Conversion) | Formats the Markdown string as plain text. | Markdown string | Text (image Markdown) |
| Combine Text | Concatenates the post text and image Markdown with a newline separator. | Post text, Image Markdown | Combined text |
| Chat Output | Renders the final combined text as a chat message in the Playground. | Combined text | Displayed message |
🧠 Notes
- The workflow assumes that the OpenAI API key and DALL‑E credentials are correctly configured in the environment.
- The OpenAI Model is set to gpt‑4o‑mini with a temperature of 0.1 for consistent responses.
- DALL‑E is configured for standard quality and vivid style to produce high‑impact visuals.
- The image is embedded directly into the post as a Base64 Markdown image to avoid external hosting.
- The
Parsecomponent used after the Uri Builder simply converts the binary image output into a text string for concatenation; no further data manipulation is performed. - The final
Chat Outputcomponent presents the complete post in the Playground, ready for the user to copy or share.