How do I a planning-only Manager worker that converts one verified presenter video into an implementation-ready, semantically complete 62-scene production bundle, including exactly eight authorized Higgsfield Kling 3.0 1080p cinematic B-roll specifications and detailed Remotion contracts for every scene

the agent that answers this

Visual Treatment Planner — Higgsfield + Remotion. A planning-only Manager worker that converts one verified presenter video into an implementation-ready, semantically complete 62-scene production bundle, including exactly eight authorized Higgsfield Kling 3.0 1080p cinematic B-roll specifications and detailed Remotion contracts for every scene.

• Step 1tool

  Inspect the bound presenter_video through its authorized local media adapter without copying it. Record immutable digest, duration_ms, codec, dimensions, fps, audio presence, and source custody receipt.

  Your model fills this step
• Step 2tool

  Preflight ffprobe and a fully local word-timestamp speech-to-text route. Install only supported local dependencies; planning performs no paid generation.

  Your model fills this step
• Step 3

  Transcribe the complete source locally with word-level timestamps and confidence. Produce transcript.json, transcript.srt, and transcript.txt; mark uncertainty and never invent speech.

  Your model fills this step
• Step 4decision

  Validate transcript coverage, duration, word ordering, and timestamp alignment against the source probe. Refuse to plan if the transcript cannot be made internally consistent.

  Decision step
• Step 5

  Create exactly 62 contiguous semantic scenes spanning the entire presenter source exactly once with no gaps or overlaps. Every scene must include scene_id, start_ms, end_ms, an exact narration_anchor quote plus onset_ms, claim, visual_purpose, and presenter_treatment.

  Your model fills this step
• Step 6decision

  Assign exactly one lane to every scene. Exactly 8 scenes must use generated_video and specify authorized Higgsfield Kling 3.0 at 1920x1080; all other scenes use sourced_video, sourced_image, remotion_scene, or presenter_only. Each generated scene must include eligibility rationale, exact prompt, negative prompt, duration, camera, factual exclusions, approval requirement, and a no-generation fallback.

  Decision step
• Step 7

  Using the required Higgsfield generation procedure, write exactly eight implementation-ready Kling 3.0 1080p cinematic B-roll request specifications for the eight generated_video scenes. Do not generate media in this planning workflow.

  Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus...

  --- version: 0.3.0 name: higgsfield-generate description: | Generate images and videos via Higgsfield AI through 30+ models including Nano Banana 2, Soul V2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, plus Marketing Studio for…

  Full skill: Higgsfield Generate
• Step 8

  Using required Remotion best practices, write a semantic compositor contract for all 62 scenes. Every scene must contain scene_id, start_ms, end_ms, narration_anchor, claim, visual_purpose, lane, animation_family, layout, headline, supporting_copy, labels[], data_series[], units, source_refs[], evidence_asset, unique_motion_signature, presenter_treatment, transition, crop_safe_regions, and fallback. Charts require named datasets, axes, units, and sources; comparisons require both sides and values; checklists require real copy; maps require named locations/routes; timelines require dated milestones. Use at least 10 animation families, never repeat a family in adjacent scenes, and use a family at most twice in any six-scene window. Forbid unlabeled bars/cards, placeholder copy, captions, debug labels, beat/scene identifiers, or meaningless decorative graphics.

  Best practices for Remotion - Video creation in React

  --- name: remotion-best-practices description: Best practices for Remotion - Video creation in React metadata: tags: remotion, video, react, animation, composition --- ## When to use Use this skills whenever you are dealing with Remotion…

  Full skill: Remotion Best Practices
• Step 9decision

  Adversarially QA and repair the plan. Machine-verify exactly 62 contiguous scenes, exactly 8 generated_video requests, all required semantic fields, exact cross-manifest IDs and timestamps, at least 10 animation families, animation variety constraints, readable meaningful copy, evidence integrity, relative paths, and explicit fallbacks. Reject generic or unlabeled graphics.

  Decision step
• Step 10

  Assemble a portable project bundle using relative internal paths and immutable artifact identities: production-plan.md, narrative-beats.json, shot-manifest.json, asset-manifest.json, generated-video-requests.json, compositor-scenes.json, transcript JSON/SRT/TXT, presenter-source.json, provenance, QA, and asset scaffolding. Never include or copy the original presenter media. Package project-bundle.zip and record its SHA-256 as a validated final output.

  Best practices for Remotion - Video creation in React

  ↳ same skill as step 8 (Remotion Best Practices)

  Full skill: Remotion Best Practices
• Step 11decision

  Independently reopen the produced ZIP and machine-verify its SHA-256, required files, exactly 62 complete semantic scene records, exactly 8 Kling 3.0 1080p requests, contiguous full-duration coverage, valid relative paths, and absence of embedded presenter media. Deliver only after all checks pass.

  Decision step

common questions

1. v7Aug 26manual

   Manager L1: require executable Higgsfield and Remotion procedures, semantic scene data, 62-scene coverage, eight authorized Kling requests, and an independent bundle gate.

2. v6Aug 24generated

   added range-safe presenter-video inspection and transcription

3. v5Aug 20manual

   reconciled active outcome capability input authority

4. v4Aug 17manual

   Add the canonical V1 outcome capability, exact presenter_video input identity, and validated project_bundle ZIP handoff.

5. v3Aug 16manual

   Make the planner provider-neutral and emit typed, immutable orchestration handoff artifacts.

6. v2Aug 6manual

   Lock to local planning-only: no media generation/rendering, no Runway/HeyGen credentials; local-tools preflight replaces the paid-service preflight.

7. v1Aug 6generated

   auto-generated from "From one raw presenter video, auto-transcribe it, map narrative beats, assign ea"

Agents are alive: every change is a version, and a run can propose improvements that get reviewed and applied.

related agents

• dailybuilder

  How do I every morning, read the latest Implexa Boardroom Debate output and turn it into a prioritized action plan split across website, dashboard, and marketing inputs

  Boardroom Debate → Daily Action Plan

• weeklybuilder

  How do I draft the friday "what i shipped this week" thread with real metrics for X and LinkedIn

  build-in-public weekly thread

• weekdaybuilder

  How do I scan reddit, hacker news, X, and changelogs for product mentions, competitor moves, and buying-intent threads worth a reply

  competitor and mentions digest

• dailybuilder

  How do I find relevant Hacker News threads every day and draft comments to reply with?

  Daily Hacker News comment drafts