How To Article from Video Demo - 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": "Convert MP4 video URL to a time-stamped transcription file",
    "taskDescription": "This task employs specialized audio processing to transform the speech within the provided MP4 video URL into a word-for-word text transcription, including timestamps indicating the exact start of each spoken segment.",
    "inputDescription": "One input is {VARIABLE1}, containing the URL of the MP4 video that needs to be transcribed. The system processes the video, extracting and converting its audio track into a textual transcription with timestamps for each spoken part.",
    "inputRequired": [
      "VARIABLE1"
    ],
    "outputDescription": "A text-based file that includes the complete transcription of the audio content in the MP4 video, with timestamps for each segment's start time. This output will be used in subsequent tasks for content analysis and structuring.",
    "outputName": "time_stamped_transcript",
    "promptInstruction": "no instruction"
  },
  {
    "taskID": 223,
    "semanticTitleOfTask": "Extract Six Key Steps from Video Transcript Using LLM",
    "taskDescription": "This task reads the time-stamped transcript and uses a language model to identify exactly six major steps, each with a short title, brief description, and a single relevant timestamp from the transcript.",
    "inputDescription": "Requires the complete time-stamped transcript with timing markers so the language model can accurately select representative timestamps for each step.",
    "inputRequired": [
      "time_stamped_transcript"
    ],
    "outputDescription": "Outputs a structured list of exactly six steps, each containing a concise title, a brief description, and one HH:MM:SS timestamp that best represents when that step begins.",
    "outputName": "six_steps_list",
    "promptInstruction": "Carefully read the entire transcript and identify exactly six major steps. For each step: 1) Provide a short title (up to seven words), 2) Write a concise 2-3 sentence description, 3) Select one timestamp (HH:MM:SS) that marks its start or demonstration. If the transcript has more or fewer than six steps, consolidate or subdivide to ensure six final steps. Label each step in sequence and end the response with 'END OF STEPS.'"
  },
  {
    "taskID": 202,
    "semanticTitleOfTask": "Extract Thumbnail Images from MP4 Video at Specific Timestamps",
    "taskDescription": "This task uses the MP4 video URL and up to six timestamps from a steps list to generate six distinct video frames. These frames are extracted as images at the specified timestamps, enabling visual documentation of each step.",
    "inputDescription": "This operation requires the MP4 video URL (VARIABLE1) and the six timestamps from the 'six_steps_list'. The system will extract a single frame at each specified timestamp. The resulting images form the basis for step-by-step visuals.",
    "inputRequired": [
      "VARIABLE1",
      "six_steps_list"
    ],
    "outputDescription": "This task returns six still images extracted from the provided MP4 video, each corresponding to one of the specified timestamps. These images will be used in subsequent steps to illustrate the tutorial.",
    "outputName": "extracted_images",
    "promptInstruction": "no instruction"
  },
  {
    "taskID": 223,
    "semanticTitleOfTask": "Generate JSON Mapping Each Step to Its Image URL",
    "taskDescription": "This task takes the six extracted image URLs and uses a language model to assemble them into a structured JSON object. Each step, from 1 to 6, is assigned its corresponding image URL, providing a clear mapping for the subsequent HTML assembly.",
    "inputDescription": "This operation requires the list of six image URLs from the previous task. The language model then creates a valid JSON object mapping each step number to its respective image URL, ensuring a consistent format for later use.",
    "inputRequired": [
      "extracted_images"
    ],
    "outputDescription": "This task outputs a concise JSON object with six keys, each a string representing the step number, and values containing the corresponding image URL. This final mapping object can be easily parsed by downstream tasks.",
    "outputName": "six_image_urls",
    "promptInstruction": "You are receiving six image URLs that correspond to six steps in a how-to tutorial. Create a structured JSON object that maps each step number (1-6) to its corresponding image URL. The output should be a valid JSON object with exactly six key-value pairs, where each key is the step number (as a string) and each value is the complete image URL. Format the JSON to be easily readable with proper indentation. Each URL should be enclosed in quotes and the output should be a complete, valid JSON object that could be parsed by any standard JSON parser. Do not include any explanation or additional text outside of the JSON structure itself."
  },
  {
    "taskID": 190,
    "semanticTitleOfTask": "Generate Final How-To HTML Article",
    "taskDescription": "This task uses Skill #190 to combine the transcript, six-step list, and images into a complete HTML article with an introduction, six steps (each referencing its own image), and an option for a concluding paragraph. The final output is a polished, publication-ready tutorial.",
    "inputDescription": "Takes three structured inputs: (1) the time-stamped transcript containing the full spoken instructions, (2) the structured six-step list with descriptive headings and short explanations, and (3) a list of six matching image URLs. Each input is used to create a cohesive, final HTML tutorial.",
    "inputRequired": [
      "time_stamped_transcript",
      "six_steps_list",
      "six_image_urls"
    ],
    "outputDescription": "Produces a single HTML file featuring an introduction, exactly six steps (each with its own heading, textual explanation, and embedded image), and an optional conclusion. The output is a fully formatted, ready-to-publish how-to article in HTML format.",
    "outputName": "final_html_article",
    "promptInstruction": "You will be creating a polished how-to article in HTML format. Use the provided transcript, six-step list, and image URLs to build a comprehensive, publication-ready tutorial. Include a clear, engaging title, a brief introduction (1–2 paragraphs), six numbered steps (each with a heading, concise text, and an embedded image tag), and an optional concluding paragraph. Use proper HTML structure throughout, ensuring each step’s image is placed directly after its explanation. The final output must be fully ready for publication."
  },
  {
    "taskID": 501,
    "semanticTitleOfTask": "Transcribe Video to Generate Time-Stamped Transcript",
    "taskDescription": "This task takes the input MP4 video URL and returns a time-stamped transcript in text format, providing the foundation for subsequent steps.",
    "inputDescription": "Requires the MP4 video URL from the user or a prior step that references the location of the video file.",
    "inputRequired": [
      "VARIABLE1"
    ],
    "outputDescription": "A comprehensive text transcript of the entire video, with timestamps included to identify relevant parts.",
    "outputName": "time_stamped_transcript",
    "promptInstruction": "no instruction"
  },
  {
    "taskID": 502,
    "semanticTitleOfTask": "Extract Exactly Six Steps from Transcript",
    "taskDescription": "This task uses a language model to analyze the time-stamped transcript and produce a structured list of six key steps, each with a relevant timestamp.",
    "inputDescription": "Requires the full time-stamped transcript from the earlier transcription task.",
    "inputRequired": [
      "time_stamped_transcript"
    ],
    "outputDescription": "A structured list of exactly six steps, each with a short title, a brief description, and a timestamp from the transcript.",
    "outputName": "six_steps_list",
    "promptInstruction": "You are analyzing a video transcript of a how-to tutorial. Identify exactly six key steps from this tutorial. Each step must have a concise, action-oriented title, a short description (2-3 sentences), and the most relevant timestamp from the transcript. Consolidate if there are more than six steps, or subdivide if there are fewer, ensuring you end up with exactly six steps."
  },
  {
    "taskID": 503,
    "semanticTitleOfTask": "Extract and Collate Six Video Thumbnails",
    "taskDescription": "Using the assigned timestamps from the six-step list, this task first extracts frame images from the MP4 video, then collates those images into a structured reference list using a language model.",
    "inputDescription": "Requires the MP4 video URL from the user and the six-step list with timestamps from the preceding task.",
    "inputRequired": [
      "VARIABLE1",
      "six_steps_list"
    ],
    "outputDescription": "Generates six distinct image files, one for each specified timestamp, along with a structured text reference that describes each image and its association with each step.",
    "outputName": "six_image_urls",
    "promptInstruction": "Create a structured list of the six image files that were extracted. For each image, provide the corresponding step number, the timestamp it was extracted from, and a brief description of the image contents. The output should be a clean and concise format."
  },
  {
    "taskID": 504,
    "semanticTitleOfTask": "Assemble Final HTML Article",
    "taskDescription": "This task takes the transcription, six-step list, and six images, then compiles them into a polished, publish-ready HTML tutorial article.",
    "inputDescription": "Requires the final outputs from the preceding tasks: the time-stamped transcript, the six-step list, and the list of six extracted images.",
    "inputRequired": [
      "time_stamped_transcript",
      "six_steps_list",
      "six_image_urls"
    ],
    "outputDescription": "A fully formatted HTML document that includes an introduction, a section for each of the six steps (with headings, descriptive text, and embedded images), and a conclusion if appropriate.",
    "outputName": "final_html_article",
    "promptInstruction": "Create a polished, professional how-to article using the provided transcript, six steps, and six images. The article should have a clear introduction, exactly six steps, and one pertinent image per step. Use clear language and format everything in semantic HTML, ready for immediate publication."
  },
  {
    "taskID": 601,
    "semanticTitleOfTask": "SEO Analysis and Readability Scoring for Final HTML",
    "taskDescription": "This task reviews the final HTML article content, analyzing it for SEO best practices (keyword usage, headings, meta tags) and evaluating readability scores. It outputs a structured summary highlighting areas for optimization.",
    "inputDescription": "Takes in the final HTML article created by the earlier subagent, which includes textual content and embedded images. The input must be the string or file reference to the HTML. This data is used to evaluate SEO factors and readability metrics.",
    "inputRequired": [
      "final_html_article"
    ],
    "outputDescription": "Outputs a structured SEO and readability analysis, detailing keyword density, recommended headings, meta descriptions, and any potential issues with text clarity or grammar. The results guide subsequent optimization steps.",
    "outputName": "seo_readability_analysis",
    "promptInstruction": "Use the final HTML content to evaluate search engine optimization best practices, including headings usage, keyword density, meta tag presence, and clarity of language. Provide a structured analysis with major findings and recommended improvements."
  },
  {
    "taskID": 602,
    "semanticTitleOfTask": "Improve Content for SEO and Clarity",
    "taskDescription": "Following the analysis from Task 601, this task refines the existing HTML text. It adjusts headings, ensures keyword usage, updates or inserts relevant meta tags, and clarifies confusing passages to improve SEO ranking and overall readability.",
    "inputDescription": "Receives the initial SEO/readability analysis from Task 601, as well as the original HTML article. Uses these inputs to systematically improve the text for search engines and human readers.",
    "inputRequired": [
      "seo_readability_analysis",
      "final_html_article"
    ],
    "outputDescription": "A revised version of the HTML article that includes recommended SEO improvements, improved headings, enhanced keyword placement, and clarified text, ensuring stronger visibility and SERP performance.",
    "outputName": "optimized_html",
    "promptInstruction": "Review the provided HTML article and the SEO analysis results to rewrite and reorganize the content. Insert relevant keywords naturally, revise headings, and ensure the tone remains coherent. Output a clean, SEO-optimized HTML version ready for publication."
  },
  {
    "taskID": 603,
    "semanticTitleOfTask": "Generate Social Media Sharing Snippet",
    "taskDescription": "This task extracts a concise, engaging snippet from the optimized HTML. The snippet highlights the key steps and includes relevant keywords for maximum outreach when shared on social media platforms.",
    "inputDescription": "Takes as input the optimized HTML produced by Task 602. Parses the content to identify the most compelling summary or highlight. The snippet should represent the article effectively and entice potential readers.",
    "inputRequired": [
      "optimized_html"
    ],
    "outputDescription": "A short, punchy text snippet (80-160 characters) suitable for posting on social media. It offers a quick overview of the article’s content and value, prompting interest in the full article.",
    "outputName": "social_sharing_snippet",
    "promptInstruction": "Review the optimized HTML article and extract a concise summary that is compelling for social media audiences. Keep it between 80 and 160 characters, and mention the core utility of the how-to article to intrigue readers."
  }
]

Happy? Now Copy-Paste To Proceed...

BACK TO SUBAGENTS

COPY TO CLIPBOARD