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/gg-mo/stagecraft-agent-skill/stagecraftgit clone --depth 1 https://github.com/gg-mo/stagecraft-agent-skillWhat 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 | $0.00057 | $0.11551 |
| Opus 5 | $0.00028 | $0.05776 |
| Sonnet 5 | $0.00011 | $0.02310 |
| Haiku 4.5 | $0.00006 | $0.01155 |
Grade A, and why
stagecraft 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 2d ago.
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 — 1,162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stagecraft
Turn an interface into a premium, keynote-style experience — minimalist, restrained, high-tech — without sacrificing clarity, functionality, accessibility, or responsiveness.
Do not just make the UI "fancier." Make it feel more intentional, more coherent, and more expensive.
When to use
Invoke Stagecraft when the user wants UI that feels: premium, polished, keynote-style, Apple-like, cinematic, modern, minimal, elegant, launch-ready, high-end SaaS, or dramatic in a controlled way. Typical surfaces: landing pages, hero sections, onboarding flows, result pages, premium dashboards, product-reveal screens, AI answer interfaces, modals, loading states, pricing sections, showcase pages.
Also invoke when the user says "audit," "elevate," "refine," or "make this feel more premium" about an existing screen — run the audit flow in references/audit-checklist.md.
When not to use
Skip Stagecraft when the requested interface should feel playful, cute, gamified, youthful, expressive, maximalist, cartoonish, neon-heavy, or intentionally quirky. Apply cautiously to dense productivity tables, enterprise data-heavy surfaces, or workflows where minimal motion is important and content density outweighs stage presence.
Workflow
- Identify UI intent. Name what the screen is for, what must remain visible and usable, and what the primary focal region should be. Write this down in one or two sentences before touching styling.
- Apply the visual system. Load
references/visual-system.mdand use its palette, type scale, spacing grid, radius, and shadow tokens. Do not invent values. - Apply motion tokens. Load
references/motion-tokens.mdand use the named variants (sceneEnter,fadeUp,fadeIn,modalIn,scaleSettle,sceneExit,staggerContainer,staggerBlock). Do not invent durations or easing curves. - Apply reveal sequencing. Load
references/reveal-patterns.mdand pick one of the five patterns (scene-entry, focal-first, line-by-line, block-by-block, progressive-disclosure). Do not combine more than one primary pattern within a single element tree. - Run the audit. Load
references/audit-checklist.mdand score the output against eight dimensions before declaring done. Fix any dimension that is not at target.
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.
- 2d ago First seen · 1,162 lines · 57 tokens per session scan A 02e7d1a16edc
stagecraft is a cursor rule published in the GitHub repository gg-mo/stagecraft-agent-skill (5 stars, last pushed 4mo ago), licensed MIT. It adds 57 tokens to every session and 11,551 once invoked, about $0.0003 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 2026-08-31.
Other cursor rules, from other repositories
slushpile
Adversarial job search: pool-anchored fit scoring, adversarial review, channel-conditional verdicts, findings written back to the profile.
authenticity-check
Score how authentically text reads as a real human author's work, flag AI-generated, AI-templated, or derivative spans, and report suspicious Unicode provenance carriers. Use when the user asks if text is authentic, reads like AI, sounds like them, or contains a hidden AI watermark or invisible Unicode. Apply even…
cursor-rule
Use the Market Daily Review skill when the request matches its SKILL.md description.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.