Hands-On With LibTV's Agent and SkillHub
卡尔的AI沃茨 · wechat · 2026-07-14
This article tests LibTV's Agent and SkillHub. The author immediately found over 100 Skills categorized into professional film, commercials, short comics, music MVs, and other niches, with further subdivisions for visual and narrative styles.
The author believes its core value isn't just "video generation" but productizing the creative workflow: the Agent asks targeted questions based on requirements, automatically connects nodes on a left-side canvas, and allows post-generation edits to prompts, camera angles, and editing timelines in the storyboard, alongside auto-checks and revisions. The article provides multiple test examples, including martial arts film clips, projector animated ads, e-commerce voiceover replication, New Chinese aesthetic TVCs, game engine PVs, and bilingual subtitle outputs.
The latter half details how to create custom Skills by breaking rules into SKILL.md, style-bible.md, and reference-film-patterns.md. These files are generated using agents like Codex/Claude Code, with errors fed back into the Skill through iterative testing. The author also shares a useful tip: generating Cantonese subtitles before adding visuals yields more stable Cantonese voiceovers.
Related event: Liblib Launches LibTVAgent for End-to-End AI Video Creation(3 posts)→
More from Multimodal
- Reddit user chains Ideogram 4 and Krea2 to mimic bbox-based image positioning — v3lh0t05c0 · 2026-07-22
- Ultimate Face Fix: Open-Source Multi-Face Repair Node for ComfyUI — Merserk13 · 2026-07-22
- Getting Started with AI Video: Solving Consistency and Censorship — cynicalnewenglander · 2026-07-22
- Storyboard-first workflows are making AI dance videos and influencers more consistent — aftahi_ai · 2026-07-22
- Runpod MCP and Claude help spin up image and video generation workflows — 802high · 2026-07-22
- Midjourney prompt turns a bee into a glitching pixel explosion — michaelrabone · 2026-07-22