Landing Optimizer logo

Landing Optimizer

CommunityPopular
aaron-he-zhu
landing-optimizer

Use when the user asks to "optimize our landing page for influencer traffic", "fix our promo-code landing page", or "improve conversion from a creator campaign"; produces a message-match audit, page-structure and social-proof recommendations, a promo-code/CTA conversion plan, and an A/B test roadmap. Not for measuring campaign results after launch — use performance-analyzer. 落地页优化/达人流量转化提升

Overview

Publisheraaron-he-zhu
Repositoryaaron-marketing-skills
Skill namelanding-optimizer
Stars
2.8K
Forks
361
Bundled files
1
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by aaron-he-zhu on GitHub. Read the source before you install it.

Installation

Install the Landing Optimizer AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git /tmp/aaron-marketing-skills
mkdir -p .claude/skills
cp -r /tmp/aaron-marketing-skills/influencer/report/landing-optimizer .claude/skills/landing-optimizer
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Landing Optimizer in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Landing Optimizer on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Landing Optimizer is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Landing Optimizer

This skill helps you create and optimize landing pages specifically for influencer marketing traffic. When users click from an influencer's post, the landing experience should feel connected and optimized for conversion.

Cross-discipline (paid ads): this is also the paid-ads post-click skill — the page half of the ROAS Offer message-match (it pairs with ad-creative-builder, which owns the ad half). The same diagnose-and-fix flow applies to paid landing pages; save paid runs under memory/ad/landing-optimizer/. On paid runs, message-match the page against the offer-claims-registry ledger when present: offer terms, promo codes, and dates against memory/claims/offers.md, and claim wording against the approved variants in memory/claims/claims-ledger.md.

Quick Start

Shortest invocation:

Optimize our landing page for traffic from [influencer campaign]

Common scenario — diagnose and fix a low-converting creator page:

Our influencer landing page has [X%] conversion rate. How can we improve it?

Skill Contract

  • Reads: a transient landing-page locator plus opaque page_ref/snapshot ref and current state, conversion rate and goal, traffic source, stable opaque creator_ref, platforms/content type, and any proposed creator display name, message, quote, asset, embed, or screenshot. Every creator reuse also reads the exact frozen approved_asset_ref plus creator-content-auditor approval_ref, and a rights record that is active, dated/evidenced, unexpired, and explicitly scoped to channel, territory, format, duration, and paid-vs-organic use. Inputs come from the user when no tool is connected.
  • Writes: return the optimization plan inline by default; save it to memory/influencer/landing-optimizer/YYYY-MM-DD-<topic>.md (or the declared paid path) only with exact WARM-save authorization. Saved artifacts and handoffs keep creator_ref, page_ref, snapshot_ref, frozen asset/approval refs, and opaque rights/evidence refs only—never a raw creator handle/name, profile/content/page URL, email, provider ID, or embedded creator media.
  • Promotes: only with separate exact authorization, promote durable facts — active campaign ref, opaque page ref, baseline conversion rate, promo code ref, primary creator_ref — to memory/hot-cache.md.
  • Done when:
    • Message-match score and named fixes are produced for the page.
    • A prioritized conversion plan (CTA, promo-code experience, friction, mobile) exists with evidence-labeled impact or Unknown/NEEDS_INPUT.
    • An A/B test roadmap with at least one hypothesis and success metric is written.
    • Every proposed creator name/quote/asset/embed/screenshot reuse has the exact frozen auditor approval and an active dated scoped-rights row covering the whole implementation/test duration; blocked reuse remains NEEDS_INPUT and is neither copied nor tested.
  • Primary next skill: performance-analyzer — measure whether the optimizations moved conversion.

Handoff Summary

Emit the standard shape from skill-contract.md §Handoff Summary Format.

Data Sources

This family needs no live integrations (Tier 1). The skill works from a transient page locator, current conversion data, opaque creator/page/evidence refs, approved messaging, and the rights inputs supplied by the user. A brief, draft, public post, or contract label does not substitute for the exact frozen creator-content-auditor approval plus current scoped-rights evidence.

Optional connectors that can deepen the analysis when available:

  • ~~analytics — pull live conversion rate, bounce rate, scroll depth, and add-to-cart events instead of asking.
  • ~~A/B testing platform — read past test results and feed sample-size/duration estimates.
  • ~~CMS / landing page builder — inspect current page structure and copy directly.
  • ~~social platform analytics — confirm the creator's actual messaging and audience.

See CONNECTORS.md for the verified free/keyless recipe per category. Every step degrades gracefully to user-supplied inputs.

Instructions

When a user requests landing page help, work through these steps. Each step's fill-in template, ASCII layout, and HTML snippet live in references/templates.md — keyed by the same step numbers.

Creator-reuse gate: before copying, proposing, publishing, or testing any creator display name, quote, claim excerpt, video, image, thumbnail, embed, screenshot, badge, testimonial, creator-specific path, or creator-linked tracking token, require all of the following for that exact reuse: stable opaque creator_ref; exact frozen approved_asset_ref; matching creator-content-auditor approval_ref with approved status for that version; rights status active; status_observed_at; opaque status_evidence_ref; unexpired start/end or perpetual duration; and explicit channel, territory, format, duration, and paid | organic | both scope matching the page and its entire proposed experiment/flight. Resolve any permitted display name or media locator only transiently at implementation. If any field is missing, stale, non-active, expired during the proposed test, disputed, revoked, unknown, or out of scope, return NEEDS_INPUT for that reuse and do not copy the name/quote, embed or screenshot the asset, publish a variant, or start a test. You may still audit non-creator page elements with opaque snapshot refs and propose generic placeholders.

  1. Assess current state — capture campaign ref, transient page locator plus opaque page/snapshot refs, traffic source, current conversion rate, goal, and the traffic context (creator_ref, platforms, content type, approved message ref, promo code ref, audience). Keep raw locators transient.
  2. Evaluate message match — compare only a creator message cleared by the reuse gate against the supplied page snapshot across message, value prop, offer, product, and tone; produce a Message Match Score (X/10) and named fixes. If the frozen approval or rights row is missing, keep the creator side Unknown, return NEEDS_INPUT, and do not quote or paraphrase it into page copy. For paid runs, also verify the page's offer/promo terms against memory/claims/offers.md when the ledger exists — an ad's "50% off" promise is only true while the offer row is live.
  3. Page structure — recommend the influencer-traffic layout (hero → social proof → product → more proof → FAQ → final CTA) and give section-by-section fixes. Any creator-specific slot stays an opaque placeholder until the reuse gate passes.
  4. Social proof integration — use a creator name, quote, asset, embed, screenshot, badge, or testimonial only when its exact frozen approval and rights scope pass; otherwise omit it and return NEEDS_INPUT. Apply the same gate separately to every additional creator.
  5. Conversion optimization — tune CTA copy/placement, design the promo-code experience (auto-apply via URL param, prominent display, confirmation), cut friction, and check mobile (load speed, thumb-friendly CTA, scroll depth).
  6. A/B testing plan — rank supported tests by impact/effort, then write at least one hypothesis with variants, sample size, duration, and success metric. Do not include or start a creator-name/asset/quote variant unless the approved rights duration covers the full planned test and resulting publication period.
  7. Influencer-specific pages — decide whether a dedicated creator page is warranted. A creator name in the path or page, creator-linked tracking token, and every personalized asset each require the reuse gate; otherwise use a generic campaign page and opaque tracking ref.
  8. Performance tracking — set targets for load time, bounce, CR, add-to-cart, AOV; define UTM params and events for attribution.

Return the finished plan inline. Offer memory/influencer/landing-optimizer/YYYY-MM-DD-<topic>.md (or memory/ad/landing-optimizer/ for paid runs) for exact WARM-save authorization, and ask separately before any HOT promotion. Before save/handoff, replace raw creator identities, media/page/profile URLs, and copied creator text with opaque refs; the persisted plan resolves nothing directly.

Example

User: "Our dated analytics export shows 1.2% CR versus our source-dated 2–3% target. Use creator_ref: creator-042, approved_asset_ref: asset-v7, and its frozen creator-content-auditor approval_ref. The supplied rights row is active, observed today with an opaque evidence ref, and covers US web landing-page display of the approved name, exact quote, video embed, and screenshot formats for both paid and organic traffic through the full six-week test/flight. The approved asset says 'smooth texture'; the page snapshot leads with 'high protein', omits the video, does not auto-apply the promo, and places the mobile CTA below the fold. Build a plan."

Output (abridged — full version in references/templates.md):

  • Diagnosis: 1.2% CR, below the supplied 2–3% target for influencer traffic.
  • Issues: message mismatch (the frozen approved asset says "smooth texture", while the page snapshot leads with "high protein"); the approved creator asset is absent; the promo is not auto-applied; the mobile CTA is below the fold.
  • Priority fixes: test the exact frozen approved video in the hero within its active scoped rights, auto-apply the promo, match the headline to approved wording, and move the mobile CTA above the fold. Any lift is Unknown until the predeclared A/B test reaches its decision rule; do not add isolated lift estimates into a promised CR.
  • Test plan: wk1 hero changes, wk2 headline A/B, wk3 CTA copy.

Reference Materials

  • templates.md — all step fill-in templates, ASCII layouts, HTML snippets, the full worked example, and tips.

  • skill-contract.md — shared contract and Handoff Summary format.

  • state-model.md — memory tiers and save-path conventions.

  • CONNECTORS.md — free/keyless data recipes per connector category.

  • conversion-quality.md — advisory conversion rubric (non-veto) to sanity-check the optimization plan.

  • Sibling skills in the influencer-marketing family:

Next Best Skill

Primary: performance-analyzer — measure whether the optimizations actually moved conversion, AOV, and attribution.

Alternates (same Report family):

  • content-amplifier — when the audit shows the page needs more creator content to feature.
  • roi-calculator — when the page's conversion is validated and you want to translate it into ROI and payback math.

Termination note: Maintain a visited-set this session. If a recommended skill has already been invoked, stop and report the chain as complete rather than re-running it. Hard stop at chain depth 3 to avoid loops.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Landing Optimizer AI skill do?

Use when the user asks to "optimize our landing page for influencer traffic", "fix our promo-code landing page", or "improve conversion from a creator campaign"; produces a message-match audit, page-structure and social-proof recommendations, a promo-code/CTA conversion plan, and an A/B test roadmap. Not for measuring campaign results after launch — use performance-analyzer. 落地页优化/达人流量转化提升

Why use Landing Optimizer on TypingMind?

Because you install it once and use it with any model. Landing Optimizer is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Landing Optimizer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/aaron-he-zhu/aaron-marketing-skills/tree/main/influencer/report/landing-optimizer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Landing Optimizer?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Landing Optimizer?

As many as you like. As long as a model supports skills, you can use Landing Optimizer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Landing Optimizer AI skill free?

Yes. It is published on GitHub by aaron-he-zhu under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇