Added Nov 3, 2026
figurehead
Your agent writes a concept and a spec, then fixed code renders it into an SVG hero image for your README.
- Ecosystem
- Agent Skills
- Skill category
- Design
- Creator / organization
- oficiallyAkshay(GitHub 用户)
What it is
figurehead is a README hero-image Skill by GitHub user oficiallyAkshay. The agent writes a concept statement and a JSON spec, then a deterministic Node tool renders it to SVG and checks the result.
Only the concept goes to the AI. The README says there's no credential: the concept is written by your own agent, and rendering and checking are pure code that call no model. The author says the same spec always renders byte-identical output, which we haven't rebuilt to confirm. The author also lists two related sibling tools, readmerlin and archify.
Who it's for:Developers who want a consistent hero image at the top of their open-source README.
Before you start: Install with npx skills or copy the Skill; you need Node 20+ and no model credentials of your own. It reads README, SKILL and manifest files, plus other project files and export names if you pass --repo, and writes .hero.json and an SVG locally. Your host AI writes the concept, and its costs are separate. MIT license. English only.
Based on the project's public documentation. Not tested by AI Inspiration World.