navig-video

Render a channel's video formats (kinetic episodes, bulletins, motion cards) from its workspace of texts, plates and voice: timeline → frames → ffmpeg, with per-destination encode variants.

Works on its own — its own CLI, no navig required · v0.2.2 · not on PyPI yet (install from source below)

On its own (no navig needed)

pip install "git+https://github.com/navig-run/plugins#subdirectory=navig-video"
navig-video --help   # no navig needed

With navig

pip install "git+https://github.com/navig-run/plugins#subdirectory=navig-video"
navig video --help

Render a channel's video formats — kinetic episodes, bulletins, motion cards and per-destination encode variants — from its workspace of texts, plates and voice: timeline → frames → ffmpeg. It runs on its own as navig-video, and inside navig as navig video.

On its own

pip install navig-video                 # adds the `navig-video` command
navig-video status                      # workspace, ffmpeg, engines, verbs
navig-video doctor                      # deep checks — exit 1 names what is missing
pip install "navig-video[render]"       # + navig-audio, navig-media-core, navig-generate, Pillow, numpy

The workspace (media, copy, voice, beds) stays where it is — pass --workspace or set NAVIG_WORKSPACE. Media never enters this repo: only the rendering code lives here. ffmpeg is probed on PATH (never a pip dependency); keys resolve from the environment or the shared vault (navig-vault).

Inside navig

The same commands are navig video ….

Why it exists

The channel's render engines (word-timed kinetic, bulletin pages, per-destination encode) are a video domain of their own — audio owns voice and beats, generate owns stills and the frame engine, and this plugin owns the timelines that turn them into video. A render verb ships only at parity with the channel's existing script output (same frames, duration and loudness); until then the scripts stay the fallback.