lavish-axi: For when a rich editor is not rich enough (4.2k stars)
What it does
All text below is quoted directly from the project README.
>HTML is the new markdown. Lavish is the new editor for your HTML artifacts.
How It Works:
- File-path identity - Sessions are keyed by the canonical HTML file path, so agents do not need opaque IDs.
- Portable artifacts - The artifact runs in a sandboxed iframe while Lavish injects a small SDK for annotations, snapshots, feedback controls, and render-time layout checks.
- Self-paint warning - lavish-axi <html-file>, export, and share run a render-free check for artifacts missing an explicit page background and return a one-line self_paint_warning.
- Open-time layout gate - The browser chrome masks an artifact only while the real in-iframe audit waits for fonts and final geometry.
- Layout issues inbox - Detection is passive. After fonts and finite animations settle, the injected SDK confirms severe failures from direct rendered evidence such as materially escaped meaningful content or required controls, clipped text fragments, viewport reachability, or near-total semantic occlusion.
>npx skills add kunchenguid/lavish-axi --skill lavish
Repository
kunchenguid/lavish-axi
Repository stats
|Metric|Value|
|:-|:-|
|Stars|4,175|
|Forks|381|
|Last release|lavish-axi-v0.1.89 (2026-10-15)|
|Languages|JavaScript 95.4%, HTML 2.8%, CSS 1.8%|
|Contributors|40|
|AI contributor accounts (Claude / Codex / Cursor / Copilot)|Copilot: 1, cursoragent: 13, claude: 13|
|AI co-authored commits|Claude: 13, Cursor: 13, Copilot: 1|
Main contributor
|Field|Value|
|:-|:-|
|Name|Kun Chen|
|GitHub|@kunchenguid|
|Commits to this repo|123 (top contributor)|
|Bio|Member of the Technical Community. Former L8 engineer at Meta, Microsoft, Atlassian.|
|Company / Location|Bellevue, WA|
|Website|linktr.ee|
|X / Twitter|@kunchenguid|
|Followers|6697|
|Public repos|78|
|On GitHub since|2013-01-10|
3 points · 1 comment
Mentions
@poteto Can I do a feature request of having something like kunchenguid/lavish-axi in grok bot itself so that I can select and direct my comments to specific sections of the artifact that is created by my @bot from my grok bot application without having to leave my chat screen?
@AgileDroids kunchenguid/lavish-axi this one? in my case that's svx / mdsvex.
lavish-axi: For when a rich editor is not rich enough (4.2k stars) What it does All text below is quoted directly from the project README. >HTML is the new markdown. Lavish is the new editor for your HTML artifacts. How It Works: - File-path identity - Sessions are keyed by the canonical HTML file path, so agents do not need opaque IDs. - Portable artifacts - The artifact runs in a sandboxed iframe while Lavish injects a small SDK for annotations, snapshots, feedback controls, and render-time layout checks. - Self-paint warning - lavish-axi <html-file>, export, and share run a render-free check for artifacts missing an explicit page background and return a one-line self_paint_warning. - Open-time layout gate - The browser chrome masks an artifact only while the real in-iframe audit waits for fonts and final geometry. - Layout issues inbox - Detection is passive. After fonts and finite animations settle, the injected SDK confirms severe failures from direct rendered evidence such as materially escaped meaningful content or required controls, clipped text fragments, viewport reachability, or near-total semantic occlusion. >npx skills add kunchenguid/lavish-axi --skill lavish Repository kunchenguid/lavish-axi Repository stats |Metric|Value| |:-|:-| |Stars|4,175| |Forks|381| |Last release|lavish-axi-v0.1.89 (2026-10-15)| |Languages|JavaScript 95.4%, HTML 2.8%, CSS 1.8%| |Contributors|40| |AI contributor accounts (Claude / Codex / Cursor / Copilot)|Copilot: 1, cursoragent: 13, claude: 13| |AI co-authored commits|Claude: 13, Cursor: 13, Copilot: 1| Main contributor |Field|Value| |:-|:-| |Name|Kun Chen| |GitHub|@kunchenguid| |Commits to this repo|123 (top contributor)| |Bio|Member of the Technical Community. Former L8 engineer at Meta, Microsoft, Atlassian.| |Company / Location|Bellevue, WA| |Website|linktr.ee| |X / Twitter|@kunchenguid| |Followers|6697| |Public repos|78| |On GitHub since|2013-01-10|
@solopribuilds I think the systems we use to approve ai code need to be better. I am using the lavish axi kunchenguid/lavish-axi which is much better than reading markdowns and approving decisions. but that would mostly evolve over time.
@trq212 @kunchenguid has a good one I have been using call Lavish. kunchenguid/lavish-axi It has been working great for me
@trq212 I have been using lavish and I found it very easy to add annotations and keep sending my feedback to the agent directly. Works really well and the UI is pretty simple kunchenguid/lavish-axi