TikTok Newsreader Promo Video - 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": 171,
    "semanticTitleOfTask": "Generate 45s News Anchor Voiceover Script with CTA",
    "taskDescription": "This task uses skill #171 to produce a concise voice-over script. It incorporates key selling points (90 seconds per day, guaranteed 100k views, no face needed), emphasizes the course’s benefits (1M–5M real member views, low cost, images provided), and ends with a clear call to action directing viewers to facelessflipping.com/a/30. The final read time should be approximately 45 seconds and maintain a fun, flirty-yet-professional news anchor tone.",
    "inputDescription": "Requires 3 user inputs: VARIABLE1 for key selling points, VARIABLE2 for tone/structure, and VARIABLE3 for CTA details/time constraint. These will form the basis for a ~45-second script covering what it is, what it does, user steps, and how to get it.",
    "inputRequired": [
      "VARIABLE1",
      "VARIABLE2",
      "VARIABLE3"
    ],
    "outputDescription": "A single text script of approximately 45 seconds in length when spoken, containing the specified selling points in a fun, semi-serious, flirty yet professional tone, with a strong CTA to facelessflipping.com/a/30.",
    "outputName": "45s_course_promo_script",
    "promptInstruction": "Write a ~45-second voice-over script for a flirty, professional, fun blonde female news anchor on TikTok, covering these points: low cost, 90 seconds/day, 100k views guaranteed or money back, no face needed, daily images, real members achieving 1M–5M views. Structure it in the order: what it is, what it does, user’s steps, and how to get it. Maintain an enthusiastic, newscaster tone, and end with a clear CTA to facelessflipping.com/a/30. Add optional timing markers at about 15-second intervals."
  },
  {
    "taskID": 170,
    "semanticTitleOfTask": "Turn Script Into Voice Over MP3",
    "taskDescription": "This task uses AI voice synthesis to convert the provided 45-second promotional script into a professional female voice-over MP3. The output is timed to fit exactly 45 seconds and highlights key course selling points with a friendly and engaging tone.",
    "inputDescription": "Requires a finalized 45-second promotional script (the output from the previous task), specifying the desired female voice style and tone (professional but flirty, enthusiastic, clear diction).",
    "inputRequired": [
      "45s_course_promo_script"
    ],
    "outputDescription": "Returns an MP3 file URL containing the fully generated voice-over audio, ready for use in subsequent video creation tasks.",
    "outputName": "voiceover_mp3_url",
    "promptInstruction": "Please generate a professional yet friendly and slightly flirty female voice-over of the provided 45-second script. Maintain an enthusiastic, confident delivery suitable for a news anchor. Speak clearly, emphasizing key benefits (90 seconds daily, guaranteed 100k views) and end with a strong call to action at facelessflipping.com/a/30. Ensure the final narration fits within 45 seconds."
  },
  {
    "taskID": 219,
    "semanticTitleOfTask": "Cut Wav/mp3 Audio into Multiple Pieces/Samples",
    "taskDescription": "This task uses Skill #219 to take a 45-second MP3 or WAV file and split it into four shorter segments. Each segment is determined by start/end timestamps, enabling separate audio clips for different video angles or other purposes.",
    "inputDescription": "Requires the MP3 URL from the previous task (voiceover_mp3_url) along with four pairs of start/end timestamps in seconds. The skill will then output an array of four distinct WAV file URLs.",
    "inputRequired": [
      "voiceover_mp3_url",
      "segment1_start",
      "segment1_end",
      "segment2_start",
      "segment2_end",
      "segment3_start",
      "segment3_end",
      "segment4_start",
      "segment4_end"
    ],
    "outputDescription": "A single output containing an array of four URLs for each WAV segment, corresponding to the given timestamps.",
    "outputName": "segmented_wav_urls",
    "promptInstruction": ""
  },
  {
    "taskID": 1981,
    "semanticTitleOfTask": "Get MP3 Transcription With Timings (Segment 1)",
    "taskDescription": "Generates an accurate transcript for the first audio segment, returning text with timestamps so the avatar’s lip-sync can be perfectly aligned.",
    "inputDescription": "Requires the array of four WAV URLs from the previous task. This task will use the first index for segment #1.",
    "inputRequired": [
      "segmented_wav_urls"
    ],
    "outputDescription": "A text transcript with timestamps for the first audio segment.",
    "outputName": "segment_1_transcription",
    "promptInstruction": ""
  },
  {
    "taskID": 1681,
    "semanticTitleOfTask": "Generate Avatar Video from MP3 & Transcription (Segment 1)",
    "taskDescription": "Creates an AI avatar video (front-facing angle) featuring a flirty yet professional blonde newsreader reading the first audio segment in 9:16 portrait orientation.",
    "inputDescription": "Takes the array of four WAV URLs (using index 0 for audio segment #1) and the transcription for segment #1 to produce a talking head avatar video.",
    "inputRequired": [
      "segmented_wav_urls",
      "segment_1_transcription"
    ],
    "outputDescription": "An MP4 video of the avatar delivering the first audio segment at a front-facing angle.",
    "outputName": "segment_1_video",
    "promptInstruction": "Generate a front-facing flirty yet professional blonde newsreader avatar with shoulder-length hair, blue eyes, white blouse (tasteful cleavage), subtle necklace, no rings, in a bright news studio background. Format is 9:16. Sync to provided audio (segment #1) and maintain an enthusiastic, engaging tone."
  },
  {
    "taskID": 1982,
    "semanticTitleOfTask": "Get MP3 Transcription With Timings (Segment 2)",
    "taskDescription": "Generates an accurate transcript for the second audio segment, returning text with timestamps for precise lip-sync alignment.",
    "inputDescription": "Requires the array of four WAV URLs from the segment cutting task. This task will use the second index for segment #2.",
    "inputRequired": [
      "segmented_wav_urls"
    ],
    "outputDescription": "A text transcript with timestamps for the second audio segment.",
    "outputName": "segment_2_transcription",
    "promptInstruction": ""
  },
  {
    "taskID": 1682,
    "semanticTitleOfTask": "Generate Avatar Video from MP3 & Transcription (Segment 2)",
    "taskDescription": "Creates an AI avatar video (slight left angle) featuring a flirty yet professional blonde newsreader reading the second audio segment in 9:16 portrait orientation.",
    "inputDescription": "Takes the array of four WAV URLs (using index 1 for audio segment #2) and its transcription to produce a talking head avatar video.",
    "inputRequired": [
      "segmented_wav_urls",
      "segment_2_transcription"
    ],
    "outputDescription": "An MP4 video of the avatar delivering the second audio segment, slight left angle (30°).",
    "outputName": "segment_2_video",
    "promptInstruction": "Generate a slight-left-angle (30°) flirty yet professional blonde newsreader avatar with shoulder-length hair, blue eyes, white blouse (tasteful cleavage), subtle necklace, no rings, in a bright news studio background. Format 9:16. Sync to the provided audio (segment #2) and maintain an enthusiastic tone."
  },
  {
    "taskID": 1983,
    "semanticTitleOfTask": "Get MP3 Transcription With Timings (Segment 3)",
    "taskDescription": "Generates an accurate transcript for the third audio segment, returning text with timestamps for precise lip-sync alignment.",
    "inputDescription": "Requires the array of four WAV URLs from the segment cutting task. This task will use the third index for segment #3.",
    "inputRequired": [
      "segmented_wav_urls"
    ],
    "outputDescription": "A text transcript with timestamps for the third audio segment.",
    "outputName": "segment_3_transcription",
    "promptInstruction": ""
  },
  {
    "taskID": 1683,
    "semanticTitleOfTask": "Generate Avatar Video from MP3 & Transcription (Segment 3)",
    "taskDescription": "Creates an AI avatar video (slight right angle) featuring a flirty yet professional blonde newsreader reading the third audio segment in 9:16 portrait orientation.",
    "inputDescription": "Takes the array of four WAV URLs (using index 2 for audio segment #3) and its transcription to produce a talking head avatar video.",
    "inputRequired": [
      "segmented_wav_urls",
      "segment_3_transcription"
    ],
    "outputDescription": "An MP4 video of the avatar delivering the third audio segment, slight right angle (30°).",
    "outputName": "segment_3_video",
    "promptInstruction": "Generate a slight-right-angle (30°) flirty yet professional blonde newsreader avatar with shoulder-length hair, blue eyes, white blouse (tasteful cleavage), subtle necklace, no rings, in a bright news studio background. Format 9:16. Sync to the provided audio (segment #3) and maintain an enthusiastic tone."
  },
  {
    "taskID": 1984,
    "semanticTitleOfTask": "Get MP3 Transcription With Timings (Segment 4)",
    "taskDescription": "Generates an accurate transcript for the fourth audio segment, returning text with timestamps for precise lip-sync alignment.",
    "inputDescription": "Requires the array of four WAV URLs from the segment cutting task. This task will use the fourth index for segment #4.",
    "inputRequired": [
      "segmented_wav_urls"
    ],
    "outputDescription": "A text transcript with timestamps for the fourth audio segment.",
    "outputName": "segment_4_transcription",
    "promptInstruction": ""
  },
  {
    "taskID": 1684,
    "semanticTitleOfTask": "Generate Avatar Video from MP3 & Transcription (Segment 4)",
    "taskDescription": "Creates an AI avatar video (three-quarter angle) featuring a flirty yet professional blonde newsreader reading the fourth audio segment in 9:16 portrait orientation.",
    "inputDescription": "Takes the array of four WAV URLs (using index 3 for audio segment #4) and its transcription to produce a talking head avatar video.",
    "inputRequired": [
      "segmented_wav_urls",
      "segment_4_transcription"
    ],
    "outputDescription": "An MP4 video of the avatar delivering the fourth audio segment, three-quarter view.",
    "outputName": "segment_4_video",
    "promptInstruction": "Generate a three-quarter-angle flirty yet professional blonde newsreader avatar with shoulder-length hair, blue eyes, white blouse (tasteful cleavage), subtle necklace, no rings, in a bright news studio background. Format 9:16. Sync to the provided audio (segment #4) and maintain an enthusiastic tone."
  },
  {
    "taskID": 173,
    "semanticTitleOfTask": "Concatenate Four MP4 Angles into a Single Video",
    "taskDescription": "This task merges four separate 9:16 MP4 video segments, each featuring a different camera angle of the avatar newsreader, into one continuous ~40-second final video. It ensures sequential playback and preserves video quality.",
    "inputDescription": "Requires four MP4 video outputs from the previous steps (segment_1_video, segment_2_video, segment_3_video, segment_4_video). They should be concatenated in the order 1,2,3,4.",
    "inputRequired": [
      "segment_1_video",
      "segment_2_video",
      "segment_3_video",
      "segment_4_video"
    ],
    "outputDescription": "A single MP4 file containing all angles combined into one seamless video (around 40 seconds in length), retaining the 9:16 aspect ratio.",
    "outputName": "combined_newsreader_angles",
    "promptInstruction": ""
  },
  {
    "taskID": 223,
    "semanticTitleOfTask": "Verify Video Concatenation with LLM",
    "taskDescription": "This optional step uses the LLM to confirm that the concatenated video meets the required quality and format standards, such as the correct sequence of segments, aspect ratio consistency, smooth transitions, appropriate duration, and uniform video quality.",
    "inputDescription": "One MP4 URL of the newly combined video from the previous task. The LLM will review its essential properties and confirm whether it meets the specified criteria.",
    "inputRequired": [
      "combined_newsreader_angles"
    ],
    "outputDescription": "A confirmation message indicating whether the concatenated video meets the requirements or any issues identified that need to be resolved.",
    "outputName": "verification_result",
    "promptInstruction": "Please verify that the concatenated video meets the following criteria: includes all four camera angle segments sequentially, maintains 9:16 aspect ratio, is approximately 40 seconds in total duration, transitions between segments are smooth, and video quality is consistent across all segments. If any issues are detected, please specify them clearly."
  },
  {
    "taskID": 190,
    "semanticTitleOfTask": "Format Animation Prompt from Key Bullet Points",
    "taskDescription": "This task uses a powerful LLM to transform the provided bullet points, CTA, and style notes into a cohesive prompt. The output will be used to generate a 5-second vertical animation that showcases the key messages in an engaging, professional, yet energetic style.",
    "inputDescription": "Requires bullet_points (e.g., 'Only 90 seconds daily', 'Guaranteed 100k views', '1M+ potential', 'No face needed', etc.), cta_url (facelessflipping.com/a/30), and style_notes (e.g., 'modern typography').",
    "inputRequired": [
      "bullet_points",
      "cta_url",
      "style_notes"
    ],
    "outputDescription": "A single text prompt describing how the animation should display each bullet point in a dynamic, visually appealing vertical format, concluding with the CTA.",
    "outputName": "formatted_animation_prompt",
    "promptInstruction": "Create a detailed animation prompt that will generate a 5-second vertical video (9:16) featuring modern typography and smooth transitions. Present these bullet points briefly but clearly: 'Only 90 seconds daily', 'Guaranteed 100k views or money back', 'No face needed', 'Many reaching 1M+ views', 'Recent success: 5M views achieved'. End with a bold call-to-action showing 'facelessflipping.com/a/30'. Maintain professional yet energetic styling with clean backgrounds that keep text legible. Consider subtle geometric shapes or motion graphics to enhance visual appeal without overwhelming the message."
  },
  {
    "taskID": 169,
    "semanticTitleOfTask": "Generate 5-Second Animated Video from Formatted Prompt",
    "taskDescription": "This skill creates the final 5-second vertical MP4 animation, taking the formatted prompt as input and producing a dynamic, text-driven clip that highlights the bullet points and ends with the CTA.",
    "inputDescription": "Requires the finalized animation prompt detailing how the bullet points should appear, style guidelines, and the CTA. The output is a 9:16 MP4 file lasting 5 seconds.",
    "inputRequired": [
      "formatted_animation_prompt"
    ],
    "outputDescription": "A 5-second MP4 animation (vertical 9:16) that rapidly displays each bullet point, transitions smoothly, and ends with the CTA on screen.",
    "outputName": "five_second_animated_clip",
    "promptInstruction": ""
  }
]

Happy? Now Copy-Paste To Proceed...

BACK TO SUBAGENTS

COPY TO CLIPBOARD