What this is
Caption News Feed turns one verified English fact into a sound-off vertical story: a highlighted hook, source tag, evidence card, question freeze, ledger turn, and measured verdict. The visual grammar uses self-drawn diagrams or rights-cleared local media rather than generic news footage.
Inputs and outputs
Provide one source URL, the exact claim wording, a source note, an explicit subject, and a 15–60 second scene brief. The Skill produces an editable project, a 1080×1920 MP4, cover image, burned captions, captions.srt, WebVTT, measured layout.json, and a render report.
A/B examples
Theme A is a 28-second source-led explainer about NASA’s measured lunar drift. Theme B is a 28-second source-led explainer comparing NASA’s Venus rotation and orbit. Both use English narration, captions, and self-drawn diagrams; their facts apply to these named examples only.
Runtime and dependencies
Use Node.js 22+, FFmpeg/FFprobe, HyperFrames 0.8.181 with headless Chrome, and local Kokoro narration when voice is enabled. Only self-drawn, public-domain, CC0, or properly licensed visual media belongs in a project.
Source and license
The grammar references LemoLab Caption News at pinned commit 413a6bd6d78ebd3bdd5abaec9538a26867f890df, under MIT. The Feed engine and package terms are recorded in LICENSE and THIRD_PARTY_NOTICES.md.
Access boundary
This is a Members Skill. The Skill package is delivered privately to active members; the public page and technical examples do not grant member delivery or rights to any source material.