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 agentmods add skills/yonatangross/orchestkit/multi-surface-rendernpx skills add yonatangross/orchestkit --skill multi-surface-rendergit clone --depth 1 https://github.com/yonatangross/orchestkitWrote 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/yonatangross/orchestkit/multi-surface-render)<a href="https://agentmods.dev/skills/yonatangross/orchestkit/multi-surface-render"><img src="https://agentmods.dev/badge/skills/yonatangross/orchestkit/multi-surface-render.svg" alt="Measured on agentmods" 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 | $0.00098 | $0.03931 |
| Opus 5 | $0.00049 | $0.01965 |
| Sonnet 5 | $0.00020 | $0.00786 |
| Haiku 4.5 | $0.00010 | $0.00393 |
Grade A, and why
multi-surface-render 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 yesterday.
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 — 386 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Multi-Surface Rendering with json-render
Define once, render everywhere. A single json-render catalog and spec can produce React web UIs, PDF reports, HTML emails, Remotion demo videos, and OG images — each surface gets its own registry that maps catalog types to platform-native components.
Quick Reference
| Category | Rules | Impact | When to Use |
|---|---|---|---|
| Target Selection | 1 | HIGH | Choosing which renderer for your use case |
| React Renderer | 1 | MEDIUM | Web apps, SPAs, dashboards |
| PDF & Email Renderer | 1 | HIGH | Reports, documents, notifications |
| Video & Image Renderer | 1 | MEDIUM | Demo videos, OG images, social cards |
| Registry Mapping | 1 | HIGH | Platform-specific component implementations |
Total: 5 rules across 5 categories
How Multi-Surface Rendering Works
- One catalog — Zod-typed component definitions shared across all surfaces
- One spec — flat-tree JSON/YAML describing the UI structure
- Many registries — each surface maps catalog types to its own component implementations
- Many renderers — each package renders the spec using its registry
The catalog is the contract. The spec is the data. The registry is the platform-specific implementation.
Quick Start — Same Catalog, Different Renderers
Shared Catalog (used by all surfaces)
import { defineCatalog } from '@json-render/core'
import { schema } from '@json-render/react/schema'
import { z } from 'zod'
export const catalog = defineCatalog(schema, {
components: {
Heading: {
props: z.object({
text: z.string(),
level: z.enum(['h1', 'h2', 'h3']),
}),
children: false,
},
Paragraph: {
props: z.object({ text: z.string() }),
children: false,
},
StatCard: {
props: z.object({
label: z.string(),
value: z.string(),
trend: z.enum(['up', 'down', 'flat']).optional(),
}),
children: false,
},
},
})
What ships with it
13 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.
- references/ork-delta.md 5.4 KB
- references/upstream-email.md 5.8 KB
- references/upstream-image.md 2.8 KB
- references/upstream-pdf.md 3.7 KB
- references/upstream-remotion.md 2.9 KB
- rules/_sections.md 1.4 KB
- rules/_template.md 339 B
- rules/pdf-email-renderer.md 3.6 KB
- rules/react-renderer.md 2.6 KB
- rules/registry-mapping.md 3.9 KB
- rules/target-selection.md 2.8 KB
- rules/video-image-renderer.md 3.7 KB
- test-cases.json 3.5 KB
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.
- yesterday First seen · 386 lines · 98 tokens per session scan A 042fa56271c8
multi-surface-render is a skill published in the GitHub repository yonatangross/orchestkit (225 stars, last pushed yesterday), licensed MIT. It adds 98 tokens to every session and 3,931 once invoked, about $0.0005 per session on Opus 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 2026-09-03.
Other skills, from other repositories
design-brain
The design knowledge (设计大脑装备) for the remotion-director builder (乙). Loads the full design equipment — the motivational framing, §0 心法, §1 visual conceit, §2 the 3-step process (narrative → texture → engine plan, order is a hard rule), §3 build discipline, §4 render self-check (you-the-designer signing off on real…
critic-loop
The blind-select + critic-loop (盲选 + 甲乙环) stage of the remotion-director pipeline — how to pick the most promising base among N draws, then run the design-blind aesthetic critic (甲) against the builder (乙), round after round until it converges, with the orchestrator ferrying verdicts verbatim. Loads the two…
c-video-edit
Programmatic video creation using Remotion (React-based) and Editly (JSON-based declarative). Create, render, and compose videos from code.
decor-cli
Use decor-cli to decorate screenshots, images, and videos with backgrounds, gradients, templates, annotations, arrows, shapes, counters, and spotlight effects from CLI or MCP.
layout
Skill for the Layout area of paddock. 55 symbols across 15 files.
frontend-design
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI).…