Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add k-kohey/axe --skill fix-uigit clone --depth 1 https://github.com/k-kohey/axeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/k-kohey/axe/fix-ui)<a href="https://agentmods.dev/skills/k-kohey/axe/fix-ui"><img src="https://agentmods.dev/badge/skills/k-kohey/axe/fix-ui/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/k-kohey/axe/fix-ui"><img src="https://agentmods.dev/badge/skills/k-kohey/axe/fix-ui.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00054 | $0.00762 |
| Opus 5.5 | $0.00022 | $0.00305 |
| Sonnet 5.5 | $0.00011 | $0.00152 |
| Haiku 4.5 | $0.00005 | $0.00076 |
Grade C, and why
fix-ui scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://raw.githubusercontent.com/k-kohey/axe/main/install.sh | sh Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL https://raw.githubusercontent.com/k-kohey/axe/main/install.sh | sh How it starts
The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design-Driven UI Fix Loop
Iteratively modify a SwiftUI View until its preview matches a target design image.
Inputs
$0: Path to the SwiftUI source file (.swift)$1: Path to the design image (PNG, JPG, etc.)
Procedure
1. Resolve inputs
$0 may contain only a View name (e.g. ContentView) instead of a file path. Resolve it to a .swift path (e.g. TodoApp/ContentView.swift) using Glob.
2. Read the design target
Read the design image at $1 using the Read tool to understand the target appearance.
3. Read the source code
Read the SwiftUI source file to understand the current implementation.
4. Capture the current preview
axe preview report "$0" --output <output.png>
axe preview report is preferred because it waits for rendering to complete and retries on failure.
If the file has multiple #Preview blocks, --output file.png will fail (it requires exactly one preview). In that case:
- Use a directory as
--outputto capture all, then pick the relevant one - Or fall back to
axe preview "$0" --preview <title|index>(oneshot, no render wait)
Read the captured image to see the current appearance.
5. Compare and identify differences
Compare the current preview against the design image. Focus on:
- Layout structure (VStack/HStack/ZStack arrangement)
- Spacing and padding
- Font sizes and weights
- Colors and backgrounds
- Corner radius and borders
- Image sizing and aspect ratios
- Text content and alignment
6. Edit the source code
Make targeted edits to the SwiftUI source file to address the identified differences. Prioritize the most visually impactful changes first.
7. Re-capture and verify
Capture a new preview after the edits. Use --reuse-build to skip rebuilding since only the View source changed:
axe preview report "$0" --output <output.png> --reuse-build
Read the new image and compare against the design.
8. Iterate
Repeat steps 5-7 until the preview closely matches the design image. Stop after at most 5 iterations. If the design still does not match, report remaining differences to the user.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 94 lines · 54 tokens per session scan C 626830e53558
fix-ui is a skill published in the GitHub repository k-kohey/axe (99 stars, last pushed 22d ago), licensed MIT. It adds 54 tokens to every session and 762 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-01.
fix-ui compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| fix-uik-kohey/axe | Skill | 99 | 22d ago | MIT | C | 54 | Claude Code |
| ios-expertpersonamanagmentlayer/pcl | Skill | 44 | 11d ago | Apache-2.0 | A | 64 | Claude Code |
| ttb-skill-native-swiftui-componentstqtuan1201/TTBaseUIKit | Skill | 23 | 3mo ago | MIT | A | 55 | Claude Code, Codex |
| ax-design-refKasempiternal/axiom-v2 | Skill | 4 | 8mo ago | MIT | A | 64 | Claude Code |
| ax-designKasempiternal/axiom-v2 | Skill | 4 | 8mo ago | MIT | A | 41 | Claude Code |
| swiftui-liquid-glassthiennc-tesoglobal/ios-skills | Skill | 3 | 1mo ago | none | A | 60 | Claude Code |
| swiftui-responsive-layoutthiennc-tesoglobal/ios-skills | Skill | 3 | 1mo ago | none | A | 76 | Claude Code |