Open sourceJavaScript · ≈105k stars · MIT
Taste Skill
Skills that give coding agents design taste: art direction, reference images, browser checks.
Agent skills that give coding agents design taste. They use art direction, reference images and a browser check to avoid generic AI-looking frontends. Install them with npx skills or as a Claude Code or Codex plugin.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- Docsv2 picks a researched art direction, builds with credited real photos, checks the page in a real browser at three widths and fixes the top issues from its own screenshots.[1]
- DocsNeeds no API keys and installs nothing; it is namespaced as taste-skill: so it doesn't collide with other skills.[1]
- DocsGrounded in a study of 247 curated sites, and a separate route handles dashboards and app screens.[1]
- MetadataMIT, actively released (v2.1.5 on 2026-10-24), with only 5 open issues.[1]
Cons
- DocsA build takes roughly 30 to 45 minutes and a few hundred thousand tokens.[1]
- DocsRequires Node 22+, an installed Chrome, Edge or Chromium, and network access to find images.[1]
- User reportsEarlier companion skills moved out, and users hit 'No matching skills found' on install and complained that agents don't reliably follow the rules.[1][2][3]
- User reportsOne commenter says tasteskill.dev, the site its license points to, itself looks like slop.[1]
- Best for
- Agents producing landing pages or app screens that must not look like generic AI templates.
- Watch out for
- Token and time costs are high per page, so save it for work where design quality matters.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
Neighbours on the workbench
MDX · ≈307k stars · MIT
obra/superpowers
An agentic skills framework and software methodology that works.
Maintained by Jesse Vincent (obra)
Shell · ≈306k stars · MIT
mattpocock/skills
Small, composable agent skills for real engineering work.
Maintained by Matt Pocock (AI Hero)
JavaScript · ≈292k stars · MIT
affaan-m/ECC
Agent harness performance optimization system: skills, instincts, memory and security for coding agents.
Maintained by Affaan Mustafa
Connections
Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.
Maintained by · 1
- LeonxlnxOrganisation
Built on · 2
- Claude CodeProject
- CodexProject
Similar entries · 2
- ImpeccableProject
- UI/UX Pro Max SkillProject
Sources
Verified Oct 29, 2026
