Video To Claude
Sends a video/GIF to Gemini's video model, gets back a frame-accurate JSON build spec.
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 16, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 2447178
Couldn't run it live (no GEMINI_API_KEY, google-genai not installed) but the repo ships real prior outputs (gallery-specs/*.json) from its own --micro pipeline — e.g. a credit-card reveal button spec with exact hex colors, a 600ms cubic-bezier(0.34,1.56,0.64,1) flip animation, and the non-obvious 'swap text at the 90-degree mark so the user never sees it change' detail — the kind of frame-level accuracy a baseline text-only pass simply cannot invent.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 4/5
What Video To Claude does
Uploads a YouTube URL, local video, GIF, or tweet video to Gemini's multimodal API with a prompt tuned to extract implementable specs (exact hex colors, pixel dimensions, animation timing in ms, cubic-bezier easing, state transitions), then hands the structured JSON to Claude to build. Triggers when a user shares a video/GIF and asks to clone a UI, reverse engineer a product demo, or extract tutorial build steps; a --micro flag adds a 'delight details' checklist for subtle micro-interaction effects (ring draws, shimmer, overshoot easing).
How to install Video To Claude
git clone https://github.com/rohunvora/video-to-claude ~/.claude/skills/video-to-claude
pip install google-genai
# then set GEMINI_API_KEY in environment (free tier available)
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Video To Claude
-
/video-to-claudeSends a video/GIF to Gemini's video model, gets back a frame-accurate JSON build spec.
It also activates on plain-language prompts like these:
-
Here's a GIF of a button animation, extract the exact hex colors and timing -
Clone this micro-interaction from the tweet video I just linked -
Reverse engineer this product demo video into implementable build specs
Frequently asked questions
- Is the Video To Claude skill free?
- Yes. The skill itself is free from rohunvora/video-to-claude. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Video To Claude work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 16, 2026. Verdict: Works with setup. Couldn't run it live (no GEMINI_API_KEY, google-genai not installed) but the repo ships real prior outputs (gallery-specs/*.json) from its own --micro pipeline — e.g. a credit-card reveal button spec with exact hex colors, a 600ms cubic-bezier(0.34,1.56,0.64,1) flip animation, and the non-obvious 'swap text at the 90-degree mark so the user never sees it change' detail — the kind of frame-level accuracy a baseline text-only pass simply cannot invent.
- What is the Video To Claude SkillProof Score?
- 8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 4/5.
- How do I install Video To Claude?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Video To Claude with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.