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 rojim666/SztuCode --skill component-instancegit clone --depth 1 https://github.com/rojim666/SztuCodeWrote 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/rojim666/sztucode/component-instance)<a href="https://agentmods.dev/skills/rojim666/sztucode/component-instance"><img src="https://agentmods.dev/badge/skills/rojim666/sztucode/component-instance/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/rojim666/sztucode/component-instance"><img src="https://agentmods.dev/badge/skills/rojim666/sztucode/component-instance.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 | $0.00040 | $0.03116 |
| Opus 5 | $0.00020 | $0.01558 |
| Sonnet 5 | $0.00008 | $0.00623 |
| Haiku 4.5 | $0.00004 | $0.00312 |
Grade A, and why
component-instance 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 today.
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 — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Components and Instances
COMPONENTorCOMPONENT_SETnodes are reusable (symbols).- Insert instances with
type: "ref"pointing to component/componentSet ID. - For Component/ComponentSet: call
batch_readwith the ID to getcomponentPropertyDefinitions, thenbatch_editto update instance properties. - Instance overrides:
- Root properties: set directly on the
refobject - Descendant properties: use
descendantsmap —{descendants: {"childId": {content: "New"}}} - Nested instances: slash-separated paths —
instanceId/nestedInstanceId/childId - Replace subtree: include
typein descendant override - "Delete" descendant: override
visible: false
- Root properties: set directly on the
- When using
descendants, paths can access multi-level descendant nodes — use paths indescendantskeys, DO NOT create multiple levels ofdescendantsobjects. - Prefer updating the component over individual instances for shared changes.
- ID formats: rendered tree uses semicolons (
instanceId;childId), batch_edit uses binding + nodeID (card+"childId"). Fall back to semicolon ID if binding fails. - Reuse existing components instead of creating duplicates.
- Instead of duplicating the same component multiple times with small tweaks, try to make them more generic so instances can reuse in more places.
- Cannot reference components across files — copy them over.
- Place reusable components on the side, next to the main design.
- Overrides are applied only to the overridden object — changes will NOT be inherited to all children.
- When parsing designs, treat "component" broadly — some are formal symbols, others are ad-hoc groupings visually behaving like components (sometimes prefixed "component/").
Composition Contract
Before inserting a component, inspect its root sizing, exposed properties, visible controls, and nested instances. A component is reusable only when its semantics fit the role in this screen.
- No duplicate affordances: inspect composed components for built-in status bars, back buttons, search controls, titles, close actions, tabs, or bottom navigation before adding siblings with the same role. Keep exactly one primary affordance per action unless the user explicitly asks for both. "It is part of the native component" does not justify two back buttons or two status bars.
- Give every repeated and nested instance distinct, scenario-derived text and icons: update the text and icon slots within each relevant nested instance to reflect its real purpose and never keeps a component default. For menu rows, tabs, list rows, table columns, and cards, replace generic labels such as "菜单" or "默认选项", placeholder sequences, and repeated default or empty-box icons with a coherent set inferred from the page scenario; otherwise, it is a generation defect.
- Operations: inspect nested descendants with
batch_readusing sufficientreadDepthandproperties: ["mainComponent"]. Override text through its TEXT component property orU(instance+"childId", {content: "..."}), and swap icons throughINSTANCE_SWAPorU(path, {mainComponent: "matchingIconId"}).
- Operations: inspect nested descendants with
- Semantic fit over coverage: do not instance a Table, Picker, Card, or other component merely because it was returned by search. Use it only when its interaction and content model match the requested module.
- Preserve composite structure: reuse the smallest complete component root; never use internal rows, columns, cells, items, or slots as standalone modules. Tables/DataGrids need headers, at least two columns, and data rows; otherwise use a list or fallback.
- Adaptive sizing: use
fill_containeron parent-driven axes andhug_contentson content-driven axes. Preserve verified intrinsic/fixed variants, and update relevant nested wrappers when their fixed sizing causes clipping or artificial gaps.
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.
- today First seen · 196 lines · 40 tokens per session scan A 06e1fce6c742
component-instance is a skill published in the GitHub repository rojim666/SztuCode (64 stars, last pushed today), licensed MIT. It adds 40 tokens to every session and 3,116 once invoked, about $0.0002 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 2026-09-12.
Other skills, from other repositories
figma
Complete guide for the Figma plugin — REST API access, real-time selection monitoring via CDP, and the figma-app interactive UI. Read this IMMEDIATELY when the user asks to work with Figma.
frontend-design
Use this skill for work a user will see rendered: websites, landing pages, dashboards, web applications, interface components, style overhauls, and requests to make an interface look better. It classifies the surface as expressive, conventional, or governed by an existing design system, then commits to palette…
design-system
DESIGN.md integration for Forge — ensures visual consistency across all UI tasks through standardized design specifications.
design-handoff
Use when a design is ready for engineering — produce an implementation-ready spec covering layout, tokens, states, responsive behavior, edge cases, motion, and accessibility.
design-system-architecture
A guide for turning a brand’s visual rules into a reusable design system. Design tokens are shared values such as colours, spacing, and font sizes; components are reusable interface building blocks.
design-taste-frontend
Anti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-flight check.