agentsclimarketplace

Pulse find exact same product

Skill joomcode/joompulse-skills/skills/pulse-find-exact-same-product

Finds product listings that appear to represent the same real-world product as a reference item. Use this skill when a user wants to find duplicate listings, match a product across listings, identify identical products by title or URL, compare product photos against marketplace listings, or find the same product on Mercado Livre.From its SKILL.md

Install
npx -y skills add joomcode/joompulse-skills --skill pulse-find-exact-same-product

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

3.5 KB, 656 tokens by cl100k_base, as published. Nobody here has run it

Find Exact Same Product

This skill helps find listings that appear to represent the same real-world product as a reference product.

"The same product" means the candidate listing matches the reference item's identity: brand, model, variant, color, size, capacity, pack count, bundle contents, and other defining attributes visible from the provided product data. Formatting, casing, punctuation, and translation differences are acceptable when the underlying product is clearly the same.

Prerequisites

  • JoomPulse MCP access is configured for the current agent environment.
  • The user provides a product title, product URL, product identifier, or product image.
  • The available JoomPulse tools can search marketplace listings and return enough product data to compare candidates.

If JoomPulse MCP access is unavailable, stop and explain that the skill requires JoomPulse MCP setup before it can search or compare products.

Workflow

  1. Resolve the reference product.

    • If the user provides a title, use it directly as the reference.
    • If the user provides a product URL or identifier, fetch the available product details first.
    • If the user provides an image, use the available image-based product search workflow to generate candidate listings.
  2. Search for candidate listings.

    • Use JoomPulse product search tools to find plausible candidates.
    • Treat search results as candidate generation, not final truth.
    • Prefer a compact candidate set first, then broaden only when recall is clearly too low.
  3. Enrich candidates.

    • Fetch product attributes, title, seller/listing metadata, images, and links when available.
    • Keep enough source data to explain why a candidate was accepted or rejected.
  4. Decide exact matches.

    • Accept a candidate only when the defining attributes match the reference.
    • Reject candidates when brand, model, size, color, capacity, pack count, bundle contents, or other defining attributes differ.
    • If data is insufficient for a confident match, do not mark it as exact.
    • Do not rely on title similarity alone when important attributes are missing or ambiguous.
  5. Return results.

    • List confirmed matches with product links when available.
    • Include a short rationale for each accepted match.
    • If there are no confident matches, say so clearly and summarize what was checked.
    • When useful, include rejected near-matches separately with the key reason they were rejected.

Output Format

Use a concise table when there are multiple candidates:

ResultProductLinkWhy it matches
MatchProduct titleProduct URLSame brand, model, size, and variant

For a single match, a short paragraph with the product link and rationale is enough.

Notes

  • Prefer precision over recall. A smaller list of confident matches is better than a broad list of uncertain candidates.
  • Do not expose raw private data, internal identifiers, or implementation details in the final answer.
  • If the user asks for bulk matching, process items one by one and make uncertainty explicit for each item.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most product growth skills give in 656 tokens

Counted across 694 of the 879 authors here whose files we hold, read 2026-09-06

  • Check for product marketing context firstin 49 of 694, across 20 files
  • Validate the why before building featuresin 18 of 694, across 4 files
  • Respond to every comment in real-timein 17 of 694, across 6 files
  • Structure launch marketing across three channel typesin 16 of 694, across 4 files
  • Recruit early users one-on-onein 13 of 694, across 2 files
  • Ask one question at a timein 13 of 694
  • Rank features using ICE scoringin 12 of 694, across 3 files
  • Identify primary conversion goalin 11 of 694, across 3 files
  • Identify traffic contextin 11 of 694, across 3 files
  • Evaluate headline effectivenessin 11 of 694, across 3 files
  • Check visual hierarchy and scannabilityin 11 of 694, across 3 files
  • Run product diagnosticsin 11 of 694, across 3 files

Said here and by no other author read

  • Ask which marketplace to use when unclear
  • Never mix data from the two marketplaces
  • Search short terms in both languages on Shopee.
  • Echo back marketplace and reference product
  • Confirm Shopee matches on brand category and price
  • List confirmed matches with their links

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.