System Prompts Leaks is a collection of captured system instructions used to guide AI chatbots and coding agents before they receive user messages. It serves researchers and developers studying how different AI assistants are directed.
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 skills add asgeirtj/system_prompts_leaks --skill tastegit clone --depth 1 https://github.com/asgeirtj/system_prompts_leaksWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/taste)<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/taste"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/taste/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/asgeirtj/system_prompts_leaks/taste"><img src="https://agentmods.dev/badge/skills/asgeirtj/system_prompts_leaks/taste.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00087 | $0.00485 |
| Opus 5.5 | $0.00035 | $0.00194 |
| Sonnet 5.5 | $0.00017 | $0.00097 |
| Haiku 4.5 | $0.00009 | $0.00049 |
Grade A, and why
taste 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 11d 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.
What it actually says
Anti-Slop: Frontend DON'Ts
What NOT to do when designing a UI or page. This is a filter, not a style; it never prescribes a look. Stacking several of the below is what reads as AI-made on sight. Adaptive, so if the user or another skill asks for something specific, do that instead; these are defaults, not overrides.
- purple / indigo / violet accent, or a purple-to-blue hero gradient (the top AI tell)
bg-clip-textgradient headline text- cream (
#faf8f4) as the whole-page background - aurora blobs / glowing colored shadows, or an identical glow on every card
- Inter everywhere, or Geist / Space Grotesk / Instrument Serif / Roboto / Arial as your only identity
- gray-400 / gray-500 body text
- one font weight and size throughout (no hierarchy)
- tracked all-caps micro-kickers above every section
- centered hero of eyebrow pill + headline + two CTAs
- the identical 3-card feature grid with a lucide icon in a rounded square
- an 8+ word, 48px+ headline that says nothing
- bento grid for everything
- the same eyebrow-headline-3cards section repeated back to back
rounded-2xl, or any single radius, on every element- glassmorphism blur + border + shadow on every surface
hover:scale-105lift on cards- emoji as feature icons (🚀 ⚡ ✨ 🎯 💡)
- left-border accent cards, a colored border on a rounded element, or cards nested inside cards
- fade-up / slide-up on everything as it scrolls in
- motion that ignores
prefers-reduced-motion - scribbly hand-drawn SVG mascots with no purpose
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.
- 11d ago First seen · 32 lines · 87 tokens per session scan A 380b59b28fa3
taste is a skill published in the GitHub repository asgeirtj/system_prompts_leaks (69,707 stars, last pushed 2d ago), licensed CC0-1.0. It adds 87 tokens to every session and 485 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-05.
Other skills, from other repositories
claude-design
Design one-off HTML artifacts (landing, deck, prototype).
sketch
Throwaway HTML mockups: 2-3 design variants to compare.
scrollcraft
Premium scroll-driven landing pages; scroll = timeline.
impeccable
Frontend design guidance, upstream-maintained (impeccable).
auteur
Design and build cinematic, award-level web pages.
motion-system
Define the app's motion language before animating anything -- a small set of durations, easings, and named moves that every animation draws from, so the app feels designed instead of decorated. Use before writing the first animation in a project, or when existing motion feels scattered.