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 rules/vercel/ai-elements/ultracitegit clone --depth 1 https://github.com/vercel/ai-elementsWhat 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 | $0.03481 | $0.03481 |
| Opus 5 | $0.01741 | $0.01741 |
| Sonnet 5 | $0.00696 | $0.00696 |
| Haiku 4.5 | $0.00348 | $0.00348 |
Grade A, and why
ultracite 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.
The source is not reproduced here
No licence file
A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.
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 · 333 lines · 3,481 tokens per session scan A a7fac75b8f46
ultracite is a cursor rule published in the GitHub repository vercel/ai-elements (2,380 stars, last pushed 10d ago), with no licence file. It adds 3,481 tokens to every session, about $0.0174 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 2026-08-30.
Other cursor rules, from other repositories
cursorrules
You are using StyleSeed, an AI design-method engine for product UI and other visual artifacts.
components
React component architecture for creating composable, accessible components with data attributes. Use when creating/updating composable components, not for higher-level feature/page components.
slate-plan
Review Slate v2 architecture/API plans against React 19.2 runtime performance, Slate-close unopinionated DX, legacy-regression proof, research evidence, and shadcn-style composability; write a scored plan and keep completion pending until every required pass and closure gate is complete.
sync-shadcn
Autogoal-backed planning, status, review, dashboard, apply, and tracking for upstream shadcn docs syncs into Plate docs. Use when the user asks for sync-shadcn, sync-shadcn status, sync-shadcn review, sync-shadcn dashboard, sync-shadcn apply, a scoped sync-shadcn lane, to sync shadcn docs, audit newer shadcn docs…
dev-browser
Fallback browser automation with persistent Chrome state. Use only when Browser Use is unavailable or blocked.
plate-plan
Define or update Plate editor-behavior specs with a pass-gated, evidence-scored plan. Use for behavior law, authority maps, protocol rows, parity gates, research-backed winner shifts, editor chrome contracts, and implementation-roadmap changes.