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 6529-Collections/6529seize-frontend --skill fe-polish-judgegit clone --depth 1 https://github.com/6529-Collections/6529seize-frontendWrote 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/6529-collections/6529seize-frontend/fe-polish-judge)<a href="https://agentmods.dev/skills/6529-collections/6529seize-frontend/fe-polish-judge"><img src="https://agentmods.dev/badge/skills/6529-collections/6529seize-frontend/fe-polish-judge/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/6529-collections/6529seize-frontend/fe-polish-judge"><img src="https://agentmods.dev/badge/skills/6529-collections/6529seize-frontend/fe-polish-judge.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 | $0.00117 | $0.01093 |
| Opus 5.5 | $0.00047 | $0.00437 |
| Sonnet 5.5 | $0.00023 | $0.00219 |
| Haiku 4.5 | $0.00012 | $0.00109 |
Grade A, and why
fe-polish-judge scanned grade A with 0 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 7d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
FE Polish Judge
Run the loop defined in ops/standards/frontend-polish-judge.md. Read that
standard and ops/standards/frontend-design-ui-ux.md first.
Workflow
-
Pin the run. Create
<run-dir>/run.jsonfromreferences/run-record.md(full or lite template). Record the audience, primary task, routes, in-scope states, viewports, locales, auth, N/A dimensions with reasons, references, and budget. Mark J13 applicable for any wallet, signature, transaction, address, ownership, delegation, consolidation, or fee surface. For changed/museum/networkroutes, use the Museum template ("profile": "museum"). -
Capture evidence. Render the candidate and capture the pinned matrix into
<run-dir>/evidence/with the browser tooling at hand. Set the browser context locale for non-en-UScells. Record task and recovery-path observations alongside the screenshots. For crypto flows, use simulated stages or a component harness; never sign a real transaction. Use the6529-agent-loginskill for signed-in surfaces. After the baseline, recapture only what the repair batch can affect. Museum runs use a production build, full-page captures at all three widths, and a contact sheet per viewport. -
Judge. Start a fresh-context judge agent (prefer a different model from the builder) with
references/judge-prompt.md,run.json, the evidence folder, and the previous round's output. Leave out the builder's narrative. Save its JSON as<run-dir>/rounds/NN.json(00-baseline.jsonfirst). For a Museum run, start three independent judges, one per lens (curatorial,visual,copy), each with its lens brief from the prompt, and save their outputs as<run-dir>/rounds/NN/<lens>.json. -
Score. Run:
node ops/skills/fe-polish-judge/scripts/score.mjs <run-dir>Act on
decision:invalid_judge_output: rerun the judge once witherrorsquoted. If it fails again, stop and report.incomplete: capture the missing evidence named inunverified, then rejudge.repair: fix the open required findings in user-impact order as one coherent batch, commit, setcandidate.commitinrun.jsonto the new head, then repeat from step 2. Whenbudget.recovery_batchis true, target the single most promising required finding.accepted: stop. Report the labeljudge-accepted (uncalibrated).threshold_not_reached: stop. Report thebestround and its unresolved findings.- If a prerequisite (fixture, environment, auth) is genuinely unavailable,
stop as
blockedwith the evidence.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 7d ago First seen · 87 lines · 117 tokens per session scan A c4cd2275e62f
fe-polish-judge is a skill published in the GitHub repository 6529-Collections/6529seize-frontend (10 stars, last pushed today), licensed Apache-2.0. It adds 117 tokens to every session and 1,093 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-12.
Other skills, from other repositories
anti-slop-design
Autonomous Principal Design Technologist. Cures vibe-coded software from AI frontend slop using curated token archetypes and 7-axis quality gating.
chakra-ui-builder
Build responsive, accessible UI components and layouts using Chakra UI v3, install or configure Chakra UI in new and existing projects, and design scalable themes using tokens, semantic tokens, recipes, and slot recipes. Use this skill whenever a user asks to build, create, or generate any UI component, page, form…
visual-ralph
Visual Ralph orchestration for frontend UI from generated references, static references, or live URL targets, using $ultragoal with built-in visual verdict and pixel-diff evidence until the implementation matches and leaves a reproducible design system.
frontend-ui-dark-ts
Build dark-themed React applications using Tailwind CSS with custom theming, glassmorphism effects, and Framer Motion animations. Use when creating dashboards, admin panels, or data-rich interfaces with a refined dark aesthetic.
frontend-visual-qa
Audits rendered UI and specific task/object-link handoffs; excludes ordinary source citations alone. Audits already-rendered UI and Markdown/Obsidian image layout — web app, deck/slide, dashboard, design-system, or Electron/native app — via real-browser/native-app journeys and a Playwright sweep. Use after UI…
prototype-web
A clickable, high-fidelity web product prototype with navigation, a hero section, feature cards, steps, social proof, and optional pricing. It is designed to resemble a finished landing page while remaining a prototype.