Produce a researched explainer as a rendered vertical video

Why use this

1. A short explainer needs its research, spoken argument and visual timing to agree. A polished render can conceal a weak claim, an unreadable caption or an illustration that suggests something the underlying evidence does not support.

2. This job could connect production assets in a review package. Claim references and scene intentions beside the render help the creator distinguish factual corrections from pacing or design changes before publication.

Why an agent / simpler alternative

Useful judgment

Translate an explanation into timed scenes and identify where narration needs visual support or qualification.

Simpler alternative

Use a slide-video template when the script and visual sequence are already settled.

Unsuitable when

The topic requires specialist verification that no qualified reviewer can provide.

Proposed inputs

1. Name

   Topic and audience brief

   Owner

   Creator

   Minimum scope

   Learning objective, audience knowledge, desired length and excluded claims

   Quality check

   Agree what the viewer should understand after watching.

2. Name

   Approved evidence and assets

   Owner

   Research owner

   Minimum scope

   Source notes, permitted images and voice preferences

   Quality check

   Track claim support and distinguish illustrative imagery from factual depiction.

Access and setup — proposed or unknown

Proposed minimum

1. Use a local production copy and approved assets; render review files separately from publishing destinations.

Compatibility unknowns

1. Unknown: Remotion dependencies, narration providers, caption tooling and runtime compatibility.

External actions requiring approval

1. Installing software, purchasing generated media, using a person’s voice or publishing the video.

Expected deliverable and review checks

Proposed artifact

Vertical review render with scene list, claim references, caption transcript and asset provenance.

Critical checks

1. Every factual caption agrees with the script and its evidence.

2. Text stays readable inside the vertical frame and narration finishes before the final cut.

Missing output evidence

No new workflow run or output inspection is performed for this editorial layer. The linked implementation evidence retains the original source-specific outcome and artifact limits.

Failure signals and proposed responses

1. Signal

   A generated illustration includes a plausible but inaccurate diagram.

   Why it matters

   Viewers may remember the visual as factual evidence.

   Proposed response

   Replace it with a verified diagram or label it clearly as conceptual illustration.

2. Signal

   Caption timing follows an earlier narration version.

   Why it matters

   Words appear late or contradict the current voice track.

   Proposed response

   Regenerate or realign the affected captions and inspect the rendered sequence.

Effort and cost — unknown

Setup work

1. Build a claim-to-scene map and agree visual and voice review criteria.

Ongoing review

1. Check factual changes, caption timing and render artifacts after revisions.

Unknowns

1. Unknown: model and voice charges, rendering resources and editorial review effort.

First check — proposed, unexecuted

Status

proposed_unexecuted

Synthetic fixture

A synthetic three-scene explainer contains one unsupported statistic, a revised narration line and a caption outside the safe frame.

Acceptance checks

1. The statistic is held for evidence.

2. The timing and frame problems are identified before any proposed release.

Stop conditions

1. A new factual claim, paid asset or voice permission remains unresolved.

Limits

1. A completed video file would establish rendering only, not factual accuracy, rights clearance or audience effectiveness.