pinterest-search

v2026.09.24

Find public Pinterest images from a keyword for moodboards and visual inspiration. Return usable image links, source pins, and titles.

GitHub
安装命令
npx skhub add postplusai/pinterest-search
Markdown
SKILL.md

Pinterest Search

Use this skill when the user wants public Pinterest image evidence — image addresses, moodboard source lists, or visual inspiration links — starting from a keyword.

Read references/search.md only when the keyword workflow needs clarification.

Job

Turn one keyword into a bounded set of Pinterest images. Return each result normalized to { image_url, pin_url, title }, deduplicated by image address. Run the smallest first pass that can answer the request, then stop and report scope, count, strongest results, and next action.

Reference Index

User asks forApply
Any Pinterest image search from a keywordreferences/search.md only if needed
Broader pass: more results, filter variation, or a second keywordreferences/search.md
Board scrape, profile scrape, pin-URL scrape, shopping/product fields, engagement metricsNot supported on the current public surface. Say so and stop
Non-Pinterest image sourcesHand off; run only the Pinterest lane here

First Question

Ask one question only when the answer changes the route, first-pass scope, or output shape.

MissingAsk
KeywordWhat keyword should I search Pinterest for?
Too broadWhich one keyword matters most for the first pass?
Filter intentShould I include all pins, or only video pins?

Do not ask the user for credentials, implementation choice, schema fields, hidden filters, or retry strategy.

Run Discipline

  1. Use the command below; read references/search.md only if needed.
  2. Run the narrowest collection that can answer the first pass; start at the minimum limit of 20.
  3. Normalize output to { image_url, pin_url, title } and deduplicate by image address.
  4. Stop after the first pass and report scope, count, strongest results, limits, and next action.

Do not present a bounded first pass as the full Pinterest catalog. Use only the public filters shown by the route.

Public Command Boundary

  • Run postplus research run pinterest-search --query <keyword> --kind <all|videos> --limit <n> --wait --output <result.json>; the minimum limit is 20.

  • Readiness diagnostics: postplus doctor --skill pinterest-search.

  • Inspect flags with postplus research run pinterest-search --help only when needed.

  • Keep the first pass bounded; expand only after inspecting the first result.

  • If the CLI returns a quote-confirmation challenge, obtain user approval for its scope and cost before running postplus quote confirm --json --challenge-file <challenge.json> and retry with the returned token.

Only if platform scope or evidence interpretation remains unclear, consult platform contract; it is not a preflight.

Evidence Quality

  1. Keep usable image records and source pin URLs; discard records without an image and deduplicate by image URL.
  2. If a completed pass is sparse or off-topic, try one focused keyword or supported kind change within scope.
  3. Allow at most two changed follow-up passes after successful but insufficient results, within approved scope and budget; do not repeat an identical request or hide a failed/pending operation.
  4. Stop when sufficient, at the bound, or when another pass would not help. Report useful evidence and uncertainty; preserve raw results and source links.

image_url is the asset, pin_url is its source page; missing titles stay missing and visual inspiration is not engagement evidence. Full machine fields belong to postplus research schema --route <route> --json; consult it only when required for processing, not before every request.

<!-- BEGIN GENERATED EXECUTION EXAMPLE -->
postplus research run pinterest-search \
  --query "example topic" \
  --wait \
  --output ./result.json

Follow the CLI's structured result and reported next action; do not infer recovery from free-text messages. Wait for explicit user approval when requested; an action does not authorize spending, publishing, or overwriting. Resume the same operation through its returned checkpoint or action; never resubmit uncertain work, repeat exhausted recovery, or switch providers to bypass failure.

<!-- END GENERATED EXECUTION EXAMPLE -->
发现
标签

此技能尚未发布标签。

版本
最新版本元数据

版本

v2026.09.24

发布时间

Sep 24, 2026

分类

未分类

许可证

NOASSERTION

源路径

skills/20-research/pinterest-search

默认分支

main

最新提交

51a1c76

Tree SHA

ee441c2