Skip to content

armgit / hypit

Public Imported from GitHub
0

Clone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-roll, ship 100 variants in one command, and get your 100M views.

docs(media): keep still video outside measured speech

kashorin3650b41b1771
diff --git a/packages/media-pipeline/README.md b/packages/media-pipeline/README.md
index 794773ba..d69d0331 100644
--- a/packages/media-pipeline/README.md
+++ b/packages/media-pipeline/README.md
@@ -77,8 +77,10 @@ pictures (every picture holds at least one frame, the rest go by weight with lef
remainders, earlier first), `bind-still-video-source` attaches each picture in authored order, and
encoding is one `render-still-video` Need. Pictures of different sizes are fitted into the first one's
frame and letterboxed. Inspection and normalization remain the same explicit steps used by imported
-or generated moving video, so the Blob can serve B-roll through Normalize, or continue into either
-estimated or measured A-roll semantics afterward.
+or generated moving video. The resulting Blob can carry a wordless Segment's media span, supply
+provisional visual media for estimated semantic timing, or satisfy a downstream input that expressly
+needs a video Artifact. Spoken performance and independent narration obtain their measured semantic
+time from their own A-roll media, not from this held picture.
The package also exposes one-picture Run Fragments. `still-video` takes `duration`, `clock`,
`layout` and `source-0`; `clip-time-still-video` takes `duration`, `clock` and `source`, supplying
diff --git a/packages/media-pipeline/src/manifest.ts b/packages/media-pipeline/src/manifest.ts
index 8aa9b3f9..c8ee5717 100644
--- a/packages/media-pipeline/src/manifest.ts
+++ b/packages/media-pipeline/src/manifest.ts
@@ -303,7 +303,7 @@ export const mediaPipelineMarkupSurfaces = [
{ name: "source", kind: "reference", required: false, accepts: [artifactTypes.blob],
summary: "Selects the one authored image held for the full video; write Still children instead for several images." },
{ name: "duration", kind: "literal", required: true,
- summary: "Sets the video's length in seconds, such as 6 or 2.5s; the author's decision, measured beforehand with hypit measure." },
+ summary: "Sets the video's length in seconds, such as 6 or 2.5s; choose it from the visual passage this held image is meant to carry." },
{ name: "clock", kind: "reference", required: true, accepts: [programSpaceTypes.clock],
summary: "Selects the frame clock used by the generated MP4." },
{ name: "guide", kind: "literal", required: false, values: ["clip-time"],
diff --git a/skills/hypit/references/production/media.md b/skills/hypit/references/production/media.md
index 3e55c7de..9c538275 100644
--- a/skills/hypit/references/production/media.md
+++ b/skills/hypit/references/production/media.md
@@ -114,7 +114,8 @@ time-bearing clip, use StillVideo:
StillVideo produces a video-only BlobArtifact. Multiple `media:Still` children divide the authored
duration by their optional weights. Normalization then makes that clip usable as prepared moving
-media. Choose this when the still should supply a Take's time or a video reference. Add
+media. Choose this when held pixels must themselves carry media time, such as a wordless Segment or
+provisional visual media paired with estimated semantic timing. Add
`guide="clip-time"` only when a short-lived diagnostic view benefits from a visible local timecode,
frame count and progress ruler; the guide is one display treatment and says nothing about later Film
placement.