Skills
The agentic intelligence layer. Every skill teaches Claude how to compose the code primitives correctly. Click any skill to read the full SKILL.md source.
How to source and select b-roll/stock clips for Zavis videos. Load this skill any time you're writing a script that needs visual material, reviewing a rendered video for clip-relevance issues, or improving the YouTube/Pexels sourcing pipeline. The single biggest reason Zavis videos fail visually is that the clip doesn't match what the narrator is saying — this skill is the discipline that prevents that.
Engineer videos for retention, not just to exist. Load this skill whenever you need to ensure a video EARNS its viewer second-by-second. The intent is never just "make a video" — it's "make a video people watch through to the end and share." This skill is the strategic mind behind every pacing, structure, and content decision. Pairs with zavis-hook-engineer (for the opening) and zavis-taste-dna (for the look).
Craft and review the first 3 seconds of any video. The hook decides whether someone keeps watching. Load this skill whenever writing the opening of a script, reviewing why a video isn't performing, or auditing a draft template. Triggers on phrases like "hook," "intro," "first 3 seconds," "stop the scroll," or any video-opening work.
Write voiceover narration for videos. Load this skill whenever writing the actual words a narrator will say, choosing a narrator profile, or reviewing if narration timing aligns with visuals. Pairs with zavis-story-writer (for the structure) and integrates with packages/pipeline/src/voiceover/profiles.ts. Updated for single-pass narration as the default.
How to craft the meta-prompts that produce Zavis-quality scripts, visual queries, and narration. Load this skill whenever starting from a raw user brief ("make me a video about X"), when rewriting a script that feels flat, or when an agent upstream needs to generate an edit-list, a beat sheet, or a set of YouTube queries. This is the "how do I talk to an LLM to get the right output" skill — it pairs with zavis-story-writer (structure), zavis-narration-director (phrasing), and zavis-clip-curator (query craft).
Construct narrative arcs for short-form and long-form videos. Load this skill whenever writing a script, planning scene order, choosing what to cut, or deciding the story spine of a video. Pairs with zavis-hook-engineer (for the opening) and zavis-narration-director (for the voiceover words themselves).
The visual taste DNA distilled from analyzing premium short-form video creators. Load this skill ALWAYS when authoring or reviewing any visual element of a Zavis video — typography, color, motion, transitions, texture, pacing. This skill encodes specific aesthetic rules from the @xandhq (X&) and other premium documentary-style accounts that have been pixel-level analyzed by the Zavis taste pipeline at taste.zavisinternaltools.in.
The YouTube Montage template — videos built from sourced YouTube clips intercut with title cards, narration, and synced captions. Load this skill whenever the user asks for a montage video, an "evolution of X" video, a "history of X" video, a state-of-the-industry video, a retrospective, a founder story built from archival footage, or any video where existing YouTube footage drives the visual layer. This is the authoritative entry point for the template — it links out to every specialist skill, every code module, every rule.
The top-level guide for the Zavis Video OS — a brand-aware video generation ecosystem built on Remotion. Load this skill whenever the user asks to make a video, work with templates, render content, or modify the Remotion primitives. This is the orientation skill — it tells Claude how the system is structured and which other skills to load next.
The self-analytical review pass that runs after every render. Load this skill whenever evaluating a rendered video, deciding whether to ship or iterate, or performing a vision-based critique. Works hand-in-hand with packages/pipeline/src/review/ which handles the deterministic checks. This skill is what you load to do the SUBJECTIVE / VISION-BASED checks that code can't do alone.