agentark-ai/agentark/skills

Claude Code and agent skill by agentark-ai/agentark

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Install

npx vskill@latest install agentark-ai/agentark/skills

SKILL.md

ArkOrbit L0 Widget Catalog

ArkOrbit widgets run in the browser Orbit canvas. User-authored files live under the orbit's L2 folder and firmware modules live in this L0 catalog.

Write user widgets as browser JavaScript modules under mod/<name>/index.js, assets under assets/, and JSON state under data/. Widget modules must export render(el, ctx = {}); the Orbit canvas imports them and mounts them automatically.

When using structured ArkOrbit file operations, every write operation must include the complete file content. Never send a write with only a path.

For edits to existing widgets, use the structured ArkOrbit operation path with a small exact find/replace edit instead of re-emitting the whole file.

Security policy:

  • Treat orbit code as browser-only display code.
  • Never write OAuth tokens, API keys, cookies, bearer headers, passwords, session material, private keys, or credential-like secrets into orbit HTML, JavaScript, CSS, JSON, or assets.
  • If live authenticated data is needed, retrieve it through authorized server-side AgentArk tools first, then write only safe rendered summaries or non-secret derived data into the orbit.
  • Keep external scripts out of generated orbit pages unless the user explicitly asks for that dependency and it is necessary for the artifact.
  • Prefer the widget context helpers (ctx.importMod(path) and ctx.resolveText(path)) over direct control-plane API calls from inside widgets.

Available firmware modules:

  • app-shell/index.js: renders compact declarative mini-app specs from a widget registry entry's spec object. Prefer this for minimal customer-facing widgets, dashboards, trackers, plans, and checklists when the spec can use app-specific summary/content, metrics, sections, rows/items, views/tabs, checklist items, refresh actions, source/fetch bindings, and visual direction to produce a useful first screen. Include visual.design_type (hero-card, glass-card, dashboard-grid, profile-panel, checklist-board, or feed-panel) so the renderer can match the requested app type with domain-specific title, labels, palette, and hierarchy instead of showing a generic card. Default to Orbit-native dark/glass surfaces in the same visual family as the surrounding space-agent UI unless the user explicitly asks for light. Keep the initial surface concise and designed; put deeper detail behind tabs or compact sections. Full deployable/backend apps belong in main chat/app_deploy, not Orbit. Use custom JavaScript for compact Orbit widgets that need rendering, public API fetching, parsing, simulation, state, animation, or behavior the declarative shell cannot express.
  • markdown/index.js: renders default orbit introduction content.
  • iframe-html/index.js: renders a self-contained HTML fragment.
  • chart/index.js: exports barChart(el, values, options).
  • table/index.js: exports table(el, columns, rows).
  • todo/index.js: renders a small local todo list.
  • Public HTTPS data can be fetched from widgets through the render context: ctx.fetchText(url), ctx.fetchJson(url), or ctx.fetchPublic(url). Prefer these helpers over direct browser fetch() for news, RSS, pricing, market data, or other public feeds because the Orbit host routes them through AgentArk and avoids common CORS failures. Do not use them for private hosts, authenticated APIs, or secrets.
  • For general latest-news widgets, do not default to Reddit, X/Twitter, forum posts, or social-media search unless the user explicitly asks for that source. Prefer public news/RSS/search feeds from news providers or aggregators, label the source in the UI, and show a clear error if a public source is unavailable. Do not use JSONP or script-tag injection for news data.

For ordinary widget requests, do not rewrite index.html. Build minimal customer-facing Orbit widgets with an Orbit-native dark/glass visual system, strong typography, concise copy, and a clear focal point. Before writing files, reject designs that read like generic admin cards, reports, fact sheets, pale brochures, oversized text dumps, low-contrast surfaces, or large app implementations. Use a compact data/widgets.json app-shell registry spec only when it can produce that result; otherwise write a compact custom browser widget module. Longer code is acceptable when it exists for real public API handling, parsing, state, animation, or interaction inside a small widget. Detailed apps belong in main chat/app_deploy. Use surgical edits for existing modules unless the user explicitly asks for supporting assets or data files.

If the user asks to add back a widget, reuse mod/<name>/index.js by restoring its data/widgets.json registry entry when that module still exists. If the module was deleted, recreate it from the user's request and conversation context.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
92
Repository forks
19
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/agentark-ai/agentark/skills/badge)](https://verified-skill.com/skills/agentark-ai/agentark/skills)

Something wrong with this skill or its page? Tell us.