Let the voice reference ride every hop that speaks

The canvas boards still described the 1.0 node

Add last_frame_guide: opt-in last-pixel AddGuide

Add master_audio_file: lock every hop to one take

Merge media-3x3 into v2

Merge cache-key-batch into v2

Anchor on the photograph, and let a restart actually be a chain start

Three reference videos and three voices, and the clips keep their sound

Add pin_mech: choose the hop-2+ pin mechanism

Close two more bindings the fingerprint could not see

Video input size is an area budget, not a short edge

Open at 1200x900; the 4:3 request was about ratio, not scale

The 4:3 default never applied to anything

Decode the reference clip at the size it will actually be used at

The canvas is a short edge, not an area; and prose reaches the scanner

The Starter previewed its seam chart and saved its contact sheet

The reference clip can say what it is for

render_from, and a cache that stops re-rendering hops nothing touched

1.0.0: the writer on the node, and the docs to match

trim windows on every media input, and a per-reference pixel budget

soundtrack: put it in the panel, where the other media already lives

soundtrack: a music bed under the chain, ducked under the dialogue

0.4.5: the hop cache stops shelling out to ffmpeg

0.4.1: fix what shipped broken, and record what the first real session measured

0.4.0: anchor tone mode, dry run, contact sheet, seam report

Document cache_hops: how to re-roll one hop without re-rendering the rest

Hand Tie Clips 0.3.3