agent-ui
Skill2 starsChanged 4 months agoMITView on GitHub
Use when building any UI piece inside packages/agent-app — the agents' React PWA. Establishes the Zustand-store-per-domain pattern, design-token consumption, and motion conventions that keep the dashboard fast, calm, and consistent with the Editorial / Letterpress Calm aesthetic.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- AngelescuCiprian/claude-skillstheir own version, first seen Aug 22, 2026
- nikunj812/claude-skilltheir own version, first seen Jul 1, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- AngelescuCiprian/claude-skills★ 0same name
- nikunj812/claude-skill★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository3 other files it gives its agents.
- mock-stateSkill
- webrtc-mockSkill
- widget-componentSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 24
Prefer CSS Modules (.module.css) for component styling.
- Very specific: line 28
Don't lift state with useState across more than one parent — promote to a store instead.
- Very specific: line 29
Don't prop-drill past two levels.
- Distinctive: line 30
Don't introduce a UI framework (MUI, Chakra, Mantine, shadcn).
- Distinctive: line 31
Don't use drop shadows, glass blur, glow effects, gradients on text, or rounded chips.
- Very specific: line 32
Don't animate width, height, padding, or margin — use transform + opacity.
- Distinctive: line 33
Don't add toast libraries.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
agent-ui
The agent dashboard is what support agents stare at for entire shifts. Information density, scannability under load, and zero needless motion matter more than novelty. The aesthetic (Editorial / Letterpress Calm) is the constraint — rules and small-caps labels do the structural work that cards-with-shadows do elsewhere.
When to use
- Touching any file in
packages/agent-app/src/ - Adding a new agent-facing UI piece (visitor monitor rows, chat pane, video tile, capacity indicators, settings) or wiring agent-side state
Do
- Functional components only. No class components, no HOCs.
- One Zustand store per domain, mounted at module scope:
useVisitorsStore,useConversationsStore,useAgentStore. Components subscribe via selector functions, not whole-store reads, so unrelated state changes don't re-render. - Derive — don't duplicate — state. Capacity, soft-lock status, "is on call" all derive from the source-of-truth stores via selectors.
- Consume design tokens from
@indianic/shared/design/tokens.css(imported once at app root). Reference them as CSS custom properties in CSS Modules or inlinestyle. Never re-declare colors/spacing. - Layout: rules with inset small-caps labels for section dividers (the memorable bit). Avoid card-on-card; flatten hierarchy.
- Motion via the
motionlibrary (framer-motion's successor). Use onlyease-outexponential easing fromtokens.motion.easing.outExpo/.out. No bounce, no elastic. - Tabular numerics for any cell that holds a count, timer, or capacity (e.g.
2/3 chats,00:38). Addfont-variant-numeric: tabular-numsonce at app root or per-component. - Prefer CSS Modules (
.module.css) for component styling. Tailwind core utilities are acceptable only for one-off layout primitives — never for color, spacing, or typography (those must come from tokens).
Don't
- Don't lift state with
useStateacross more than one parent — promote to a store instead. - Don't
prop-drillpast two levels. If the third level needs it, it belongs in a store. - Don't introduce a UI framework (MUI, Chakra, Mantine, shadcn). The aesthetic is custom and these undo it.
- Don't use drop shadows, glass blur, glow effects, gradients on text, or rounded chips. They are forbidden by the brief.
- Don't animate
width,height,padding, ormargin— usetransform+opacity. For accordions, animategrid-template-rows: 0fr → 1fr. - Don't add toast libraries; the dashboard's status surface is the row + the header pills. For Phase 2 a single-page shell with view state in a store is fine — no router needed.
One short example
// packages/agent-app/src/stores/visitors.ts
import { create } from 'zustand';
import type { VisitorId, VisitorIdentity, VisitorContext, VisitorStatus } from '@indianic/shared';
interface Row {
identity: VisitorIdentity;
context: VisitorContext;
status: VisitorStatus;
seenAt: number;
}
interface VisitorsState {
rows: Map<VisitorId, Row>;
upsert: (row: Row) => void;
remove: (id: VisitorId) => void;
}
export const useVisitorsStore = create<VisitorsState>((set) => ({
rows: new Map(),
upsert: (row) =>
set((s) => {
const next = new Map(s.rows);
next.set(row.identity.visitorId, row);
return { rows: next };
}),
remove: (id) =>
set((s) => {
const next = new Map(s.rows);
next.delete(id);
return { rows: next };
}),
}));
export const selectEngagedRows = (s: VisitorsState) =>
Array.from(s.rows.values()).filter((r) => r.status === 'engaged');Reference
- Aesthetic + tokens:
packages/shared/src/design/README.md - Event schema:
packages/shared/src/events/schema.ts - Phase doc:
dev-docs/phase-2-agent-dashboard-mockup.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

