Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
- agent-skill
- video
- promo
- marketing
A Claude Code skill that turns a feature description into a clean, abstract product illustration — as real hand-written SVG, not a picture of one.
Spotkit converts a feature description into a minimal, abstract product illustration and ships it as production-ready SVG the agent writes by hand — never image generation. Each output carries a concept and a metaphor, not just shapes, and the repo demos twelve illustrations (teams, categories, workspace access, API gateway, audit log, invoices and more) each rendered in light and dark mode from the same source. The bundled AGENTS.md tells the agent what to read and what to skip, so the skill picks itself up whenever you mention an illustration need. Works in Claude Code, in the agents directory other harnesses read, and as a zip upload for Claude.ai.
Empty-state art, feature explainers, docs diagrams and abstract product marketing visuals where you need editable vectors — not another raster image to redraw.
# Claude Code: git clone https://github.com/Devesh-Shirsath/spotkit.git ~/.claude/skills/spotkit # other harnesses (Codex etc.): git clone https://github.com/Devesh-Shirsath/spotkit.git ~/.agents/skills/spotkit # Claude.ai: Settings -> Capabilities -> Skills -> upload a zip of this repo
MIT. Hand-written SVG means slower generation than an image model and no photorealism — it's for abstract product illustration, not hero photography. Restart Claude Code after installing so the skill registers.
Repository
Devesh-Shirsath/spotkitCharted
Guizang's agent skill turns your repo's recent updates into a finished promo video — real components, code-rendered animation, an original code-written soundtrack — no video editor involved.
An agent skill that geolocates a photo with nothing readable in it — OSM geometry, elevation skylines and satellite tiles turn “where was this taken?” into a measured answer with an error radius.
A skill that makes AI coding agents map the architecture and declare the blast radius before they touch a line of code — rendered as one standalone HTML page.