How To Article from Video Guide - Confirm & Copy Final Agent Code

Describe

Describe your task

Refine

Refine the plan

SubAgents

Review all agents

Deploy

Deploy your agent

Review Final Agent Copy

Expand

[
  {
    "taskID": 207,
    "semanticTitleOfTask": "Get Transcription From MP4 Video URL",
    "taskDescription": "This task takes an MP4 video URL as input and processes its audio track to generate a full text transcription with precise timestamps. It ensures each segment of speech is mapped accurately to its time range.",
    "inputDescription": "This task requires a single input, the MP4 video URL, provided as VARIABLE1. The skill will retrieve and analyze its audio content, performing speech-to-text conversion and mapping each spoken phrase to an exact timecode.",
    "inputRequired": ["VARIABLE1"],
    "outputDescription": "The output is a text-based transcription, structured by timestamps indicating when each segment of speech occurs in the video content. This facilitates easy reference and navigation within the video.",
    "outputName": "timestamped_transcription",
    "promptInstruction": ""
  },
  {
    "taskID": 223,
    "semanticTitleOfTask": "Parse Video Transcription Into Six Tutorial Steps",
    "taskDescription": "In this task, we use a powerful LLM to analyze the entire timestamped video transcription. The LLM identifies exactly six key steps in the tutorial, each with a unique title, relevant timestamp from the transcription, and a concise description.",
    "inputDescription": "This task requires the complete timestamped transcription from Subagent1 as a text input. The LLM uses this text to parse and select exactly six important steps, each requiring a stepNumber, a relevant timestamp, and a short descriptive summary. The input must also contain actual timestamps that appear in the transcription.",
    "inputRequired": ["timestamped_transcription"],
    "outputDescription": "The output is a well-structured JSON containing exactly six steps derived from the provided transcription. Each step includes a stepNumber (1-6), a short descriptive title, a relevant timestamp, and a concise explanation (1-2 sentences) outlining the action or concept covered in that step.",
    "outputName": "six_steps_summary",
    "promptInstruction": "Analyze the provided transcript. Return a JSON with exactly six steps, each containing: 1) stepNumber (1-6), 2) a short, action-oriented title, 3) a timestamp (HH:MM:SS) from the transcript, and 4) a concise description (1-2 sentences). Focus on the essential steps needed for the tutorial. Ensure the output is valid JSON. If the transcript has more or fewer than six steps, condense or expand so that exactly six steps are returned."
  },
  {
    "taskID": 301,
    "semanticTitleOfTask": "Gather Input for Screenshot Extraction",
    "taskDescription": "This task gathers and validates the MP4 URL and the six timestamps needed for screenshot extraction.",
    "inputDescription": "Takes two inputs: (1) the MP4 video URL [VARIABLE1], and (2) the six-step summary [six_steps_summary] which contains the timestamps.",
    "inputRequired": ["VARIABLE1", "six_steps_summary"],
    "outputDescription": "No direct output; the input is simply prepared and passed along to the next task.",
    "outputName": "",
    "promptInstruction": ""
  },
  {
    "taskID": 302,
    "semanticTitleOfTask": "Extract Six Screenshots from the Video",
    "taskDescription": "This task calls skill #202 to extract six screenshots from the MP4 video at the specified timestamps, then returns an array of six image URLs with standardized filenames.",
    "inputDescription": "Requires the validated MP4 URL [VARIABLE1] and six timestamps from [six_steps_summary]. Both are essential for generating the correct screenshots.",
    "inputRequired": ["VARIABLE1", "six_steps_summary"],
    "outputDescription": "An array of six image URLs, named screenshot_step1.png to screenshot_step6.png, each capturing a frame at the specified timestamp.",
    "outputName": "six_screenshots",
    "promptInstruction": "Please extract screenshots from the provided MP4 video at the following six timestamps: step1, step2, step3, step4, step5, step6. For each screenshot, name the output image 'screenshot_stepX.png' and capture the entire frame at exactly the specified time. Return a list of all six image URLs in sequence."
  },
  {
    "taskID": 223,
    "semanticTitleOfTask": "Generate publication-ready HTML article from multiple inputs",
    "taskDescription": "This single LLM-based step transforms the user instructions, raw transcription text, six-step summary, and screenshot URLs into a polished, self-contained HTML tutorial with six steps, each featuring a heading, description, and an embedded image.",
    "inputDescription": "Requires the user's initial instructions, the raw transcription of the video, a structured list of six steps (with titles, descriptions, and timestamps), six image URLs for screenshots, and a prompt guiding the LLM to build the final HTML article.",
    "inputRequired": [
      "VARIABLE1",
      "timestamped_transcription",
      "six_steps_summary",
      "six_screenshots"
    ],
    "outputDescription": "Outputs a single, self-contained HTML article containing exactly six steps, each with a heading, a short action-oriented description, and one embedded screenshot image. The article references key points from the transcription and is publication-ready.",
    "outputName": "final_article_html",
    "promptInstruction": "In this single final step, you will combine the provided user instructions, raw transcription, six-step summary, and screenshot URLs into one cohesive HTML tutorial. Include exactly six steps with headings, short descriptions, and the corresponding image for each step. Begin with a concise introduction, maintain a clear instructional tone, embed images with alt text, and ensure the final article is ready to publish without modification."
  }
]

Happy? Now Copy-Paste To Proceed...

BACK TO SUBAGENTS

COPY TO CLIPBOARD