amazon-research
Amazon Research
Follow shared release-shell rules in:
postplus-sharedrelease-shell rules
Use this skill for Amazon platform-data work.
Default use cases:
- 看 Amazon 竞品
- 查价格带
- 抓评论和差评
- 看 best sellers / new releases
- 判断某个产品在 Amazon 上怎么打
Do not default to web search for these when this skill can cover them. Use web only for external facts or supplemental context.
Task Shapes
Classify the request first:
- Product discovery
- keywords
- category pages
- search URLs
- collection key:
amazon-products - alternate collection key:
amazon-free-products
- ASIN enrichment
- known ASIN list
- need titles, prices, ratings, seller details
- collection key:
amazon-asins
- Review mining
- need reviews, complaints, low-star patterns
- collection key:
amazon-reviews - alternate collection key:
amazon-reviews-v2
- Bestseller mapping
- need trend products or leaderboard products
- collection key:
amazon-bestsellers
Default Workflow
Use the lightest valid chain:
- collect
- normalize
- analyze
- synthesize
Prefer:
- small sample first
- URL input when URLs are known
- ASIN input for review work
Tell the user:
- "我会先用少量关键词或 ASIN 做第一轮 Amazon 采集,输出排名表、详情表、评论摘要或榜单;如果信号够强,再交给 sourcing-selection 做跨证据判断。"
Release-Shell Execution Contract
- keep request files, raw datasets, normalized datasets, and analysis caches
under
<work-folder>/.postplus/amazon/ - keep only final user-facing summaries or shortlisted exports outside
.postplus/ - compile the request into a small input JSON before the expensive collection step when URLs, ASINs, or keyword seeds need shaping
- start with a bounded first pass:
- one keyword set
- one ASIN batch
- one bestseller page
- if hosted capability is unavailable, unauthorized, or returns a stable network error, stop immediately instead of switching to ad hoc shell glue
Normalized schema:
schemas/amazon-dataset.schema.json
Main scripts:
${CLAUDE_SKILL_DIR}/_postplus_shared/00-core/shared-collection/scripts/collection_actor_run.mjsscripts/normalize_amazon_dataset.mjsscripts/analyze_amazon_dataset.mjs
Routing Reminder
If the user asks strategic Amazon questions like:
- 最像的竞品有哪些
- 价格锚点是什么
- 评论里最常见的问题是什么
- 我们要怎么打
Treat them as Amazon platform-data questions first, not generic strategy questions.
Downstream Handoff
After delivering results, offer to hand off to sourcing-selection for supply-side
validation or to drill deeper into a specific ASIN's reviews.
More from postplusai/postplus-skills
audio-transcription
Transcribe local or remote audio into durable text and timestamp artifacts using hosted Whisper models. Use this when the job is speech-to-text from audio files and you need request/response persistence, optional timestamps, and subtitle-ready outputs.
84google-trends-research
Research Google Trends search-intent signals for topic discovery, keyword momentum, regional interest, and rising queries without treating search trends as the same thing as platform content heat or marketplace demand.
78seedance-submitter
Use when preparing, submitting, polling, or debugging Seedance 2.0 video generation jobs from product images, storyboard images, UGC scripts, voiceover copy, or promptPlan request JSON. Use for splitting scripts into render segments, uploading references, creating request JSON, submitting jobs through the hosted capability, polling predictions, and handing off local render paths.
76social-media-publisher
Prepare and, after explicit approval, publish social posts through the PostPlus platform-owned Postiz workspace.
76facebook-research
Research Facebook pages, public follower or following surfaces, and public posts using hosted collection capability. Use this when the user wants Facebook account research, follower-surface sampling, or public post metrics.
76x-tools
Local execution tools for X/Twitter hosted collection workflows, including actor runs, dataset normalization, tweet ranking, account ranking, audience graph construction, and language clustering.
75