Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/monthu56/landforgenpx agentmods add skills/monthu56/landforge/landing-experimentWrote 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/monthu56/landforge/landing-experiment)<a href="https://agentmods.dev/skills/monthu56/landforge/landing-experiment"><img src="https://agentmods.dev/badge/skills/monthu56/landforge/landing-experiment/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/monthu56/landforge/landing-experiment"><img src="https://agentmods.dev/badge/skills/monthu56/landforge/landing-experiment.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 | $0.00113 | $0.01286 |
| Opus 5 | $0.00056 | $0.00643 |
| Sonnet 5 | $0.00023 | $0.00257 |
| Haiku 4.5 | $0.00011 | $0.00129 |
Grade A, and why
landing-experiment 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 8d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
landing-experiment — A/B-эксперименты на лендинге
Готовит A/B-тест по правилам проекта (правила A/B встроены в этот скилл):
варианты живут на одном URL (inline-сплит, без клоакинга), вариант прокидывается в
аналитику, всё фиксируется в аудит-журнале. Опирается на готовую A/B-инфраструктуру в
index.html (AB_ACTIVE, cookie-сплит, ym(...,'params',{ab:…})).
Жёсткие правила
- Один URL, без клоакинга (SEO-безопасность). Роботу и людям — один контент.
Control (вариант A) — в HTML; вариант B применяется JS поверх. Никаких скрытых от роботов
текстов/ключей. Отдельные URL под варианты — только с
canonical+noindex(не наш случай). - Одна переменная за тест (иначе результат не интерпретировать).
- Анти-FOOC и анти-CLS: вариант фиксируется до первого paint, подмена не двигает вёрстку.
- Трафик-гейт: всерьёз тест запускают после аналитики и ≥100 визитов/день. Меньше — фоновый сплит без выводов или честный before/after (пометка «B/A»).
- Каждый шаг — в аудит-журнал (skill
landing-journal).
Workflow
1. Определить эксперимент
Зафиксируй (шаблон — reference/experiment-design.md):
- Гипотеза в форме «Если [изменение], то [метрика] вырастет, потому что […]».
- Переменная — ровно одна. Бери из бэклога гипотез (приоритизация по ICE) или сформулируй.
- Variant A (control) и Variant B — точное различие.
- Primary-метрика + MDE; secondary/прокси; guardrail (что не ухудшить).
- Сплит 50/50, cookie 30 дней; длительность ≥ целого числа недель; критерий решения.
Если гипотеза/метрика не заданы — добери через AskUserQuestion (предложи топ по ICE из §3).
2. Подготовить варианты
- Control — существующий контент лендинга (в HTML).
- Variant B — реализуй inline-сплитом по
reference/ab-implementation.md: настройTEST-имя, cookie-сплит, применение варианта до paint, передачу варианта в Метрику (params) и GA4 (event/dimension), цели с меткой варианта (data-goal). - Несколько вариантов (B, C…) — допустимо, но это A/B/n: нужно больше трафика; помни про MDE.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 71 lines · 113 tokens per session scan A db83819128f1
landing-experiment is a skill published in the GitHub repository monthu56/landforge (10 stars, last pushed 3mo ago), licensed MIT. It adds 113 tokens to every session and 1,286 once invoked, about $0.0006 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 skills, from other repositories
Landing Page Copy Improver
Rewrites landing page copy to be more compelling, benefit-focused, and conversion-optimized.
landing-page-scaffold
Generate a self-contained landing page HTML prototype with five standard sections (Hero → Social Proof → Features → Pricing → CTA), outputting a single inlined HTML file ready to preview in any browser. Triggered when the user asks to build, generate, or scaffold a landing page, wireframe, or marketing page prototype…
ads-performance-analytics
How to read paid media dashboards without fooling yourself. Attribution models, platform reporting quirks, multi-platform reconciliation, ROAS vs LTV horizon traps, statistical noise in performance metrics, incrementality testing, and the failure modes that produce expensive lessons. Triggers on read paid media…
ads-creative-development
How to produce ad creative that converts at performance scale. Hook patterns, format selection, video pacing, variation systems, sequential testing methodology, fatigue detection, brand-voice alignment without conversion dilution, and platform-specific creative norms. Triggers on ad creative, ad design, hook patterns…
paid-media-strategy
A discipline for running paid media that does not light money on fire. Hypothesis writing for paid spend, channel selection, budget allocation, audience targeting, bid strategy, campaign types, what NOT to spend on, attribution reality, and the failure modes that produce expensive lessons. Triggers on paid media…
ecommerce-full-pipeline
An e-commerce workflow assistant covering product discovery, 1688 product collection, listing on multiple platforms, promotional copy, and short-video creation. 1688 is a Chinese online wholesale marketplace.