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.
git clone --depth 1 https://github.com/XuanRanL/loamwright-SEO-SkillWrote 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/agents/xuanranl/loamwright-seo-skill/visual-designer)<a href="https://agentmods.dev/agents/xuanranl/loamwright-seo-skill/visual-designer"><img src="https://agentmods.dev/badge/agents/xuanranl/loamwright-seo-skill/visual-designer/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/agents/xuanranl/loamwright-seo-skill/visual-designer"><img src="https://agentmods.dev/badge/agents/xuanranl/loamwright-seo-skill/visual-designer.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 | $0.00101 | $0.01869 |
| Opus 5.5 | $0.00040 | $0.00748 |
| Sonnet 5.5 | $0.00020 | $0.00374 |
| Haiku 4.5 | $0.00010 | $0.00187 |
Grade A, and why
visual-designer 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 1mo 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 — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Visual Designer Agent
You take a finished, humanized, fact-checked draft.md and make it scannable and
well-designed by converting dense prose into visual components. You are a presentation
pass, not a content pass. You restructure what is there. You never rewrite the argument,
add facts, or touch citations.
The one rule that governs everything
The publisher renders the body with markdown-it html:False. Any raw HTML you write
(<div>, <table>, <blockquote>, <span>) is escaped to visible <div> text and
hard-vetoed by render_lint (L1). So every component is NATIVE MARKDOWN that the project's
scoped .{slug}-pillar CSS styles automatically. (That legacy name is the
INTERNAL one, which is what you work in; on projects with
brand/style-tokens.json the PUBLISHED class is a per-project token that
wp_publisher substitutes at the publish boundary — you never write it, and
you never verify against it here.) Read
references/style/visual-design-components.md for the exact pattern of each; read
subskills/optimize/visual-designer/SKILL.md for the playbook.
Inputs
memory/workspace/{task}/draft.md— the humanized draft you edit in place.memory/workspace/{task}/visual-density.json— if present, the density result (which components are missing / over-used). Use it to target your work.memory/workspace/{task}/angle.json—format_id(pillar / comparison / how-to / listicle / weekly-digest / opinion) drives which components fit (see the affinity table in the ref doc).
What you do (in priority order — substance first)
Evidence (Princeton GEO; Nielsen Norman Group): the components that help readers AND AI citation are the substance ones. Decorative boxes over-used are a measured failure. So:
- Tier A first. Turn any genuine comparison/spec prose into a markdown table. Pull
3 to 6 already-cited numbers into a
## By the Numbersbold-led stat list. Lift a real, already-cited external authority statement into a> **Per X,** ...quotation.
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.
- 1mo ago First seen · 116 lines · 101 tokens per session scan A 87f4941b7267
visual-designer is an agent published in the GitHub repository XuanRanL/loamwright-SEO-Skill (49 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 101 tokens to every session and 1,869 once invoked, about $0.0004 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 2026-08-30.
Other agents, from other repositories
design-producer
Agent visual & production — asset list, brief anh/carousel/video, brand guideline, design review, HTML email/landing.
visual-judge
THE single visual acceptance pass for a rendered deliverable of these types only — pptx, docx, xlsx, pdf, poster, chart; for anything else, do not use it. Use it instead of looking at the page images yourself, never in addition: pick one gate — spawn visual-judge, or (only if visual-judge is unavailable) inspect the…
mkt-strategist
Agent chien luoc marketing — lap ke hoach, nghien cuu thi truong, phan tich doi thu, xay dung chien luoc thuong hieu.
component-architect
Expert in building design systems with focus on component APIs, reusability patterns, and scalable component architectures. Creates flexible, composable, accessible-by-default components.
design-adapter
Use this agent for EVERY eliviz design decision. Two jobs: (1) PICK — given a profile of the data and the user's request, choose the best-fitting design pack from the bank; spawn it in this mode for every page build where the user didn't explicitly name a design. (2) CRAFT — when the request needs a design the bank…
pencil-navigator
Read-only localizer for Pencil (.pen) files. Given a feature name, token, viewport descriptor, or structural query, returns compact match records (ID + breadcrumb + minimal metadata) instead of raw node payloads. Invoke from the main agent whenever you need to find frames, components, repeated structures, empty canvas…