video-analysis

Video Analysis

Use When

• The user provides a local video file or video URL and asks to watch, inspect,

break down, deconstruct, analyze hooks, understand shots, capture spoken lines, or explain why a video works.

• Use this for video-level evidence beyond metadata. Do not answer actual

video-understanding requests from transcript guesses or general marketing knowledge.

Do Not Use When

• The task belongs to ideation, QA, or another released skill listed in the handoff section.
• Required inputs are missing and guessing would change the result.

Execution Boundary

• Analysis runs through the hosted video-analysis capability; discover the

model keys and request shape with postplus media schema --json.

• Supported local formats are .mp4, .m4v, .mov, and .webm.
• The analyze verb only accepts an already-hosted video reference. Upload the

local video first with postplus media-file upload, then pass that result's output.storageReference object verbatim as the Gemini file_reference. The file_reference must be that storage-reference object (bucket, storagePath, name, mimeType), not a URL and not output.data.download_url — the hosted boundary reads the bytes back from storage and materializes the Gemini file server-side. Upload is a separate generic verb, not part of media analyze.

• This boundary does not claim inline video bytes, compression, segmentation,

resumable upload, or file URI reuse. If the upload or the hosted analyze call fails, stop on that error.

Source And Path

• A direct local file can be analyzed immediately. For a URL, first download or

recover the local video, then upload it with media-file upload.

• Preserve sourceId, sourceUrl, videoFilePath, sourceMetadataPath or

dataset path, model, prompt version, and source basis so results can be joined back to source metadata.

• Keep downloaded videos when they are expensive to source. Keep analysis

Markdown files and manifests under a stable workspace path.

Analysis Scope

• The default analysis is a single output per source video.
• It covers practical short-form structure such as hook, pacing, shot beats,

VO/on-screen text, product timing, and creative strategy.

• It also asks for Visual & Brand Signals when visible: genre/mood, color

palette, lighting, camera language, editing rhythm, brand feeling, and best-fit creative use cases.

Output And Handoff

• media analyze returns the hosted provider response (the Gemini candidates)

verbatim. Read the analysis text from that response and write one natural Markdown file per source video into a stable workspace path; there is no batch runner or summary file.

• The analysis should cover useful video evidence such as shot beats, timeline,

VO/on-screen text, reusable content structure, and creative strategy when those are relevant. If the provider returns the analysis wrapped in JSON, unwrap it to readable Markdown in-context.

• Results should stay grounded in observable video evidence. Database fields,

catalog frontmatter, or search indexes belong to a separate ingestion step, not to the general video-analysis boundary.

Public Command Boundary

• Step 1 — upload the local video:

postplus media-file upload --input-file <video> --mime <video/mp4|video/quicktime|video/webm> --output <upload.json>. Read the output.storageReference object from the result (bucket, storagePath, name, mimeType, sizeBytes).

• Step 2 — author the Gemini request file: contents with a text prompt part

and a file_reference part set to that output.storageReference object (not a URL and not output.data.download_url), plus optional generationConfig.

• Step 3 — run the analysis:

postplus media analyze <model-key> --request <gemini-request.json> --output <result.json>. When the source video duration is known (e.g. from the local file before upload), pass --video-seconds <n> so the hosted boundary can route eligible short videos efficiently; omit it when the duration is unknown.

<!-- BEGIN GENERATED EXECUTION EXAMPLE -->

postplus media analyze video-analysis --request request.json --output result.json

<!-- END GENERATED EXECUTION EXAMPLE -->

• Discover the model keys and request shape with postplus media schema --json;

do not call provider APIs directly.

• If the CLI returns a quote-confirmation challenge, run postplus quote confirm --json --challenge-file <challenge.json> and retry with the returned token.
• Choose the smallest matching command from the user input and run it directly.
• Readiness diagnostics: postplus doctor --skill video-analysis.

If a command fails, report the exact error and stop. Do not bypass the failure by answering from metadata, base64-inlining video, readiness probing, or unowned fallbacks.