Perfume dupe finder for AI agents

People searching for a perfume dupe finder want a cheaper bottle that smells close to one they already know. A chat model will invent a clone. scentrev-mcp will not: find_dupes returns community similars filtered by relative value.

Three tools, three jobs

• find_dupes — cheaper adjacent alternatives. This is the fragrance dupe API.
• get_reminds_of — closest smell, no price filter. Use it when “similar” matters more than “cheaper.”
• compare_fragrances — longevity, sillage, and value for two or more slugs, so the agent can say what you give up.

A prompt that stays honest

Resolve the reference with resolve_slug, call find_dupes, then compare_fragrances on the original plus the top two. Ask the model to quote n_records and to skip any row with a thin vote count. That is a dupe finder. A single name with no score is a guess.

The same call works as a cologne dupe finder. Men's, women's, and unisex fragrances share one catalog of 126,000+ scents. Point Cursor or Claude at the perfume MCP, or call the fragrance API with the same key.

For “summer office” or “winter signature” shortlists, use the perfume recommendation API instead. Dupes answer “what is cheaper and close.” Recommendations answer “what fits this brief.”

Questions

What is a perfume dupe finder?

A perfume dupe finder returns cheaper fragrances that the community treats as similar to a named scent. On scentrev-mcp the tool is find_dupes. It is a fragrance dupe API as well as an MCP tool.

How is find_dupes different from get_reminds_of?

find_dupes filters community similars by relative value, so the list leans cheaper. get_reminds_of returns closest-smelling alternatives with no price filter. Use compare_fragrances when you need longevity and sillage side by side.

Can I search cologne dupes with the same API?

Yes. A cologne dupe search uses the same catalog and the same find_dupes tool. Cologne, perfume, and fragrance are one database here, not three products.