Skip to content
cccskills
FreePublished on GitHub

no-emojis-in-ui

Emoji and Unicode-glyph removal for Lucide-only UI (Pattern Patrol P6). Use when running a P6 patrol, a UI edit meets an emoji or icon glyph in TSX, or asked to enforce Lucide-only icons.

The English summary is being prepared. Showing the original description.

Run score Not tested

See how to install

Files included (2)

  • SKILL.md4.9 KB
  • agents/openai.yaml209 B

SKILL.md (original)

Review the instructions the agent will receive before you install.

no-emojis-in-ui

Enforce one rule: user-visible UI uses Lucide icons, never emoji or Unicode icon glyphs. Follow the Pattern Patrol constitution and keep each mutation batch to 15 files or fewer.

Scope: our own interface chrome only. Emoji a person chooses for their own content (page and callout icons, text they type, imported emoji) are allowed — Arman 2026-10-08. Never remove or flag those, including role="img" emoji spans in features/spaces.

Detect

Run the registry detector over .tsx files:

rg -n -P '[\x{1F000}-\x{1FAFF}\x{2600}-\x{26FF}\x{2700}-\x{27BF}\x{2205}]' --glob '*.tsx'

Inspect every match in rendering context. Classify it as:

  • Finding: JSX text, a rendered string, a chip/title/button label, seed or sample data rendered by the app, product-authored text copied/exported for the user, or an icon value that reaches the UI.
  • False positive: non-rendered comment/documentation, console-only diagnostic text, a parser constant accepting glyphs as input, or an internal sentinel value that never reaches rendered or exported content.
  • Unresolved: reachability or rendering cannot be proven. Report it; never silently clear it.

Rendered demos, admin tools, debug panels, and sample apps are UI findings. User-authored or external content is not ours to rewrite; report the rendering path only if the product itself injects the glyph.

Route approval; never stop at the scan

Every verified finding takes exactly one route:

  1. Auto-approved: it passes every gate below; fix it immediately.
  2. Manual approval: the problem is certain and the repair is known, but an auto-approval gate fails; propose the exact repair to Arman.
  3. Unresolved: certainty or a safe repair is missing; keep it open with the missing evidence.

An empty auto-approved set is not completion. Tier R forbids unapproved mutation; it still requires a plain-English approval proposal for every certain, safe, worthwhile repair.

Auto-approval gates

Auto-approve only when all are true:

  • The glyph is authored by us and visibly functions as UI iconography.
  • One exact Lucide icon preserves its meaning, or an adjacent Lucide icon already communicates the same meaning and the glyph is redundant.
  • Copy, handler behavior, state, semantic colors, DOM interaction, ARIA text, desktop/mobile layout, and light/dark behavior stay unchanged.
  • The repair adds no wrapper/card/chrome, changes no chunk boundary, widens no type, adds no suppression, and touches no generated file.
  • The result remains clear without relying on icon shape alone.

Anything ambiguous routes to manual approval. Never auto-approve edits to user content, parser behavior, terminal protocols, branded art, complex ASCII or Unicode diagrams, or copy whose meaning changes when the glyph disappears.

Approved transformations

Choose the semantic Lucide icon already used by the component when available. Otherwise import it directly from lucide-react.

UI meaningNormal repair
success / completeCheck or CheckCircle2
warningTriangleAlert
private / lockedLock
publicGlobe2
keyword / keyKeyRound
documentFileText
removed / truncatedScissors
new gameGamepad2
reset / refreshRotateCcw
celebrationPartyPopper
purely decorative glyphdelete it; preserve the text

This is a semantic map, not a blind replacement table. Preserve icon size, alignment, and shrink-0 where wrapping text needs it. Mark decorative icons aria-hidden="true"; an icon-only control keeps a real accessible label. Delete a glyph instead of adding a second icon when a sibling Lucide icon already conveys the same state. Never replace an emoji with a Lucide icon banned by matrx/no-banned-lucide-icons; delete redundant decoration or route the choice to manual approval.

Verify and certify

For each batch:

  1. Re-run the detector on every touched file and confirm only documented false positives remain.
  2. Run pnpm type-check, git diff --check, and the relevant repo gates.
  3. Inspect changed surfaces at desktop and mobile widths in light and dark themes. Exercise every changed control.
  4. Spawn a second adversarial agent with: "assume this broke something; find it." It reruns the gates, audits false-positive classes, and returns CERTIFIED or REJECTED.
  5. Fix or fully revert a rejected batch. Commit a certified batch by path and push to main; the release train ships it. Never run ./scripts/release.sh unless Arman asked you to.

Update .matrx/PATROL_SIGHTINGS.md and .matrx/patrol-reports/no-emojis-in-ui.md with the findings, fixes, approval route, and certifier verdict.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Admin debug system wiring for the floating AdminIndicator's live debug data and Copy Full Context. Use when adding debug visibility to a route or feature, wiring useDebugContext or the debug panel, capturing console errors, or enabling the copy-context workflow for a page.

The English summary is being prepared. Showing the original description.

armanisadeghi/ai-matrx4Updated Oct 23, 2026

Compact two-icon Copy / Copy-for-AI controls (components/agent-copy). Use when adding copy buttons to a row, card, list, or record; merging duplicate Copy/JSON/AI controls; continuing the copy rollout; or writing a Copy-for-AI payload. NOT for markdown content actions (use rich-document-actions).

The English summary is being prepared. Showing the original description.

armanisadeghi/ai-matrx4Updated Oct 23, 2026

Disclosing a surface's existing fixed AI jobs in the shell's top Agents menu. Use when a page, panel, overlay, or window already runs a mandate behind a button, assist, automatic action, or mode; when the agent-disclosure guard names a file; or during a surface check.

The English summary is being prepared. Showing the original description.

armanisadeghi/ai-matrx4Updated Oct 23, 2026

The method for getting a cheaper model to match an agent's big-model answers by fixing instructions, inputs and setup. Use when asked to move an agent to a smaller or cheaper model, prove one model beats another, or compare models on an agent's test cases.

The English summary is being prepared. Showing the original description.

armanisadeghi/ai-matrx4Updated Oct 23, 2026

The watch-fix-rerun method for making an existing platform agent or automated process efficient and correct: baseline its ledger, run one unit yourself, fix the class behind every wasted call, rerun, record. Use when asked to improve, optimize, watch, or 'make efficient' an agent, a sandbox session, a sync job, or any recurring automated process.

The English summary is being prepared. Showing the original description.

armanisadeghi/ai-matrx4Updated Oct 23, 2026

Redux state for agent execution and firing agent shortcuts. Use when editing features/agents/redux/, building agent UI, creating a conversation, touching assembleRequest or NDJSON stream state, wiring agent-state selectors, adding a per-conversation capability, or triggering a shortcut from a button, menu, or code (useShortcutTrigger, launchAgentExecution).

The English summary is being prepared. Showing the original description.

armanisadeghi/ai-matrx4Updated Oct 23, 2026

See all skills by armanisadeghi

Report a problem with this skill