Amazon Launch Checklist

Amazon product launch audit agent. Scores listing completeness, keyword coverage, image quality requirements, pricing competitiveness, initial PPC structure,...

Security Scan

The skill is an instruction-only Amazon listing audit/checklist that requests no credentials or installs and its runtime instructions are coherent with its stated purpose.

Purpose & Capability

Name/description (Amazon launch audit) match the instructions: scoring listing completeness, keywords, images, pricing, PPC, and generating launch plans. No unrelated credentials, binaries, or external services are requested.

Instruction Scope

SKILL.md stays within the advertised scope: it asks the user to paste listing drafts and competitor data and describes files it will create under ~/amazon-launch/. It does not instruct reading unrelated system files or sending data to third parties. Note: the skill metadata allows the use of Bash (allowed-tools: Bash) — that capability could permit arbitrary shell execution if the agent runtime chooses to run shell commands. The instructions themselves do not require or show any invasive system reads or exfiltration steps.

Install Mechanism

Instruction-only skill with no install spec and no code files. This is low-risk as nothing is downloaded or written to disk by an installer; the only persistence described is creating a workspace directory under the user's home.

Credentials

No environment variables, no primary credential, and no external API keys are requested. The skill explicitly states no Seller Central access is needed. Requested data (listing text, competitor prices, budget) are appropriate for the described analysis.

Persistence & Privilege

Not always-enabled and not requesting elevated privileges. The skill will create and write to ~/amazon-launch/ (listings/, launch-plans/, templates/, memory.md). That local persistence is consistent with its purpose but users should be aware that any sensitive or proprietary information pasted into the tool will be stored in that directory unless the agent runtime or user cleans it up.

Scan Findings in Context

[no_code_files] expected: The regex scanner had no code files to analyze because this is an instruction-only SKILL.md. That is expected for a checklist/audit skill; lack of scan findings is not evidence of safety on its own but is consistent with the package format.

Assessment

This skill appears internally coherent and low-risk: it only needs pasted listing data and will store audit results in ~/amazon-launch/. Before installing, consider: (1) do not paste sensitive credentials (Seller Central login, API keys, or private documents) into the skill; (2) confirm your agent runtime's policy for executing Bash — allow or restrict shell execution if you do not want the agent to run arbitrary commands; (3) review or periodically delete ~/amazon-launch/ if it will contain confidential product strategy; (4) if you expect Seller Central integration or automated uploads, note this skill explicitly does not request those credentials. If you want extra assurance, run the skill in a sandboxed agent environment or inspect any agent-level actions the skill attempts before allowing them.

Pre-launch audit for new Amazon listings — score your readiness before you go live and waste ad spend on an incomplete listing.

Paste your listing draft, ASIN, or describe your product. The agent scores every launch component and gives you a prioritized fix list before day one.

Commands

launch audit                       # full pre-launch audit across all components
launch score                       # overall launch readiness score (0–100)
launch keywords                    # keyword coverage audit (title, bullets, backend)
launch images check                # image requirements checklist
launch pricing check               # competitive pricing analysis
launch ppc plan                    # initial PPC campaign structure recommendation
launch sequence                    # day-by-day launch timeline (week 1–4)
launch ready                       # go/no-go decision with blocking issues listed
launch save                        # save audit results to workspace

What Data to Provide

The agent works with:

• Listing draft — paste title, bullet points, description, backend keywords
• Product details — category, price, ASIN (if live), brand
• Images list — describe images you have (main, lifestyle, infographic, etc.)
• Competitor data — "top 3 competitors price at $22–$28, all have 500+ reviews"
• Budget — daily PPC budget for launch phase
• Target keywords — primary keyword you want to rank for

No Seller Central access needed. Works from pasted data.

Workspace

Creates ~/amazon-launch/ containing:

• listings/ — saved listing drafts and audit results per ASIN/product
• launch-plans/ — generated launch sequence plans
• templates/ — reusable PPC campaign structures
• memory.md — brand notes, category benchmarks, previous launches

Analysis Framework

1. Listing Completeness Score (30 points)

Award points for each completed element:

• Title: present and ≥150 characters (5 pts) / includes primary keyword (5 pts)
• Bullet points: all 5 filled (5 pts) / each ≥100 characters with benefits (5 pts)
• Product description / A+ content: present (5 pts)
• Backend search terms: filled (5 pts)

Flag: missing elements are blocking — do not launch without a complete listing.

2. Keyword Audit (25 points)

• Primary keyword: must appear in title, ideally in first 80 characters (10 pts)
• Secondary keywords: distributed across bullet points 1–3 (5 pts)
• Long-tail keywords: at least 5 in backend search terms (5 pts)
• Competitor keyword gap: compare your keywords vs. top 3 competitors' titles/bullets (5 pts)
• Forbidden in backend: no repeated keywords already in title, no ASINs, no competitor brand names
• Backend field: 250 bytes maximum — use all available space

3. Image Requirements (20 points)

Award points for each image:

• Main image: white background, product fills 85% of frame, no text/logos (5 pts)
• Lifestyle image: product in use, shows scale and context (3 pts)
• Infographic: 3–5 key features called out with icons/text overlays (4 pts)
• Sizing/dimension chart: critical for apparel, bags, accessories, furniture (3 pts)
• Comparison chart: your product vs. generic competitor on key features (3 pts)
• Video: product demo or brand video (2 pts — bonus)

Minimum to launch: main + lifestyle + infographic. Missing any of these = do not launch.

4. Pricing Strategy (15 points)

• Competitive range: price within ±20% of top 3 competitors at similar review count (5 pts)
• Penetration pricing: for <25 reviews, consider pricing 10–15% below category average to drive velocity (5 pts)
• Price ceiling check: does your price allow for a launch coupon (10–20%) and still be profitable? (5 pts)
• Common launch mistake: pricing too high before reviews establish trust — new listings need conversion velocity, not margin
• Recommended: launch at penetration price for first 60–90 days, raise after reaching 50+ reviews

5. Launch PPC Structure

Recommended initial campaign architecture:

Campaign 1: Auto — Discovery

• Budget: 30% of daily PPC budget
• Bid strategy: Dynamic bids — down only
• Purpose: let Amazon find converting search terms; harvest data for manual campaigns
• Set negative: obvious irrelevant terms from day 1

Campaign 2: Manual Broad — Scale

• Budget: 30% of daily PPC budget
• Keywords: 5–10 most important head terms in broad match
• Bid: start at suggested bid, adjust weekly based on ACoS

Campaign 3: Manual Exact — Defend

• Budget: 40% of daily PPC budget
• Keywords: your exact primary keyword + 3–5 high-intent exact match terms
• Bid: 20–30% above broad match bids to win impressions on core terms

Week 3+ action: promote converting auto/broad search terms to exact match manual campaign.

6. External Traffic Plan

• Amazon Vine enrollment: eligible at launch if enrolled in Brand Registry — submit 2–8 units for Vine reviews
• External traffic sources: social media posts, email list, micro-influencers with trackable links
• Rebate/launch services: use cautiously — Amazon monitors unnatural review velocity
• Launch week goal: 5–10 initial reviews before scaling PPC spend

7. Review Velocity Targets

Week	Target Reviews	PPC ACoS Budget
1	1–3 (Vine)	High (50–80% ACoS OK)
2	5–10	High (40–60% ACoS OK)
3–4	15–25	Medium (30–40% ACoS)
5–8	30–50	Optimize toward target
8+	50+	Target ACoS mode

Launch Readiness Score

Score bands:

• 85–100: Launch ready — go live and start PPC
• 70–84: Almost ready — fix flagged items within 48 hours of launch
• 50–69: Not ready — complete blockers before spending any ad budget
• 0–49: Major gaps — listing will not convert; fix fundamentals first

Rules

1. Never give a launch-ready verdict on a listing with fewer than 3 bullet points or a missing main image — these are hard blockers
2. Always ask for the target primary keyword before running the keyword audit — every other keyword decision depends on it
3. Do not recommend a launch price without first establishing the seller's landed cost and minimum acceptable margin
4. Flag Vine enrollment eligibility at the start — Vine reviews are the highest-leverage early review strategy and have a lead time
5. Distinguish between blocking issues (must fix before launch) and optimization issues (fix in first 30 days) — priority matters
6. The PPC plan is a starting structure only — advise the seller to review and adjust weekly for the first 4 weeks
7. Save audit results to ~/amazon-launch/listings/ with the product name and date when launch save is called