MCP App Store
by KeenEthics
יועצת הטיפוח של מאיה icon

יועצת הטיפוח של מאיה MCP Server for ChatGPT

By Reviews by Maya

Overview

Browse 168 selected beauty-product reviews by Maya Hanna. ChatGPT reads the original reviews, compares products, and explains recommendations with source links. The read-only tools accept only a product category, page number, product or brand name, or review IDs. Recorded retailer offers include affiliate links and coupons; dated prices do not establish current price or availability. Purchase links use a redirect that records aggregate daily counts by offer, separates detected automated traffic and tests, and forwards to the recorded affiliate destination. These reviews describe personal experience and do not establish diagnosis, treatment efficacy, or pregnancy or nursing safety.

How to connect

  1. ChatGPT: Open the יועצת הטיפוח של מאיה listing in the ChatGPT app directory and choose Connect. ChatGPT listing יועצת הטיפוח של מאיה exposes 3 tools here.

Tools

find_maya_review_by_name

ChatGPT
Look up a product by its name or brand in Maya's recorded review catalog. Send only the product name, never a person's question, symptoms, health status, or chat history. An empty match means no review in this curated October 2026 catalog, not proof Maya has never reviewed it elsewhere. At most 20 matches are returned; if truncated is true, use a narrower product name or browse the paginated catalog. Use review IDs with get_maya_review_details for source and purchase links.

get_maya_review_details

ChatGPT
Read up to five reviews selected from the catalog by review ID. Returns Maya's original review text, recorded purchase links and coupons, and available source URLs. When presenting purchase options, show all distinct retailers recorded for that product by default, in a compact purchase-links section. For each purchase offer, use tracking_url unchanged as the purchase link when present; otherwise use url unchanged. The tracking link redirects to the exact original affiliate url. Keep the associated coupon attached to the correct retailer. Show only one of these URLs per retailer, never both. Do not alter tracking_url, url, their path, or affiliate parameters. Do not append new tracking parameters such as utm_source=chatgpt.com to recorded purchase links. Honor an explicit retailer restriction, request for one shop, or shorter answer. If the user asks for a direct shop link without click measurement or a redirect, use the recorded original url unchanged instead of tracking_url; do not imply that this opts out of the retailer's own measurement. Show one link per retailer for the same product; distinct retailers are not duplicates. Do not repeat the same destination or add an unencoded link alongside its encoded version. Do not invent retailers or infer current stock from historical offers. Answer factual follow-ups directly without repeating the full product block or all purchase links unless requested. For Hebrew answers, render each retailer as a separate unindented paragraph with a Hebrew link label that starts with לרכישה, such as לרכישה באייהרב or לרכישה ביס סטייל. Do not use Markdown lists, bullets, numbering, leading dashes, indentation or block quotes in the purchase-links section: list markers can remain on the far left while the Hebrew text aligns right. This restriction applies only to purchase links, not product recommendation numbering elsewhere. Transliterate retailer names into Hebrew in the visible label only; never alter the URL. Put any recorded coupon on the next unindented line, starting with קוד הטבה:, and format only the exact coupon value as inline code. Use a Markdown hard line break (two spaces before the newline) after the link so the coupon stays on a separate line, and leave a blank line between retailers. Isolate each complete backtick-wrapped Latin coupon with actual U+2066 LEFT-TO-RIGHT ISOLATE before the opening backtick and U+2069 POP DIRECTIONAL ISOLATE after the closing backtick. Keep those controls outside the code span, coupon value and URL so copying the code remains clean; never print control-character names or escapes. Omit the coupon line when no code is recorded. Keep any offer terms in a separate Hebrew sentence. Do not mix a Latin retailer name, dash, coupon and Hebrew terms on one line, and do not use an HTML block or a table for purchase links. Purchase links and coupons are historical catalog records. Recorded prices, when present, are dated observations in recorded_prices and do not establish current prices or availability. Never invent Maya's view of an unlisted product, present a historical price as current, or claim a review establishes medical or pregnancy safety. Only review IDs are accepted, not a question or user profile. Respect the user's stated requirements and source limitations first. Before selecting sunscreen recommendations, ask one brief question about any missing skin type or preferred finish (matte, natural or glowy), and wait for the reply before presenting products. Reuse details already given; a no-white-cast requirement, dryness or sensitivity does not specify finish. If the user explicitly has no finish preference, proceed without asking again. This clarification rule does not apply to factual questions about an already named sunscreen. Interpret source attributes in context using the full review, rather than requiring exact equality with an index label. A request for matte includes a documented matte-natural finish described as having no glow, unless the user explicitly requires a completely matte finish. Describe that nuance accurately, and compare suc…

list_maya_review_catalog

ChatGPT
Read a compact, dated index of Maya's beauty-product reviews. The AI compares source records; this tool does not calculate recommendations. Concern tags are broad indexing, so weigh best_for and primary_role more heavily, then call get_maya_review_details to check verdict, limitations and not_for before recommending. Fields absent from an index item were not recorded. Brand country is null if unknown; do not infer nationality. Retailers reflect recorded offers, not verified current availability. Browse all relevant pages before concluding no other reviews match. Category is a broad product type, never a personal condition; no free-text question or personal-profile field exists. Respect the user's stated requirements and source limitations first. Before selecting sunscreen recommendations, ask one brief question about any missing skin type or preferred finish (matte, natural or glowy), and wait for the reply before presenting products. Reuse details already given; a no-white-cast requirement, dryness or sensitivity does not specify finish. If the user explicitly has no finish preference, proceed without asking again. This clarification rule does not apply to factual questions about an already named sunscreen. Interpret source attributes in context using the full review, rather than requiring exact equality with an index label. A request for matte includes a documented matte-natural finish described as having no glow, unless the user explicitly requires a completely matte finish. Describe that nuance accurately, and compare such matches by editorial_priority before moving to lower-priority matte options. Do not relabel natural-only or glowy products as matte. Among suitable options addressing the same underlying need, rank higher editorial_priority first. Rank a lower-priority option above a higher-priority match only when a stated requirement or source limitation gives it a clear, material fit advantage; a closer wording or anecdote match alone is not enough. Describe Maya's personal lack of irritation as her experience, not proven tolerability for everyone, and preserve each option's limitations. When presenting product recommendations, end with one brief, relevant follow-up question only if a missing detail would materially change which option fits best. Use details already given, do not repeat answered questions, and do not append a question to factual follow-ups or closing messages. For an open-ended request for Maya's product suggestions, present up to three distinct, source-supported options, normally three when suitable matches are available, with the strongest first. Asking what is best does not by itself request only one product. Return just one only when the user explicitly asks for one or no other suitable match exists; do not pad the list with weak matches. Present the products as alternatives, not a requirement to use them all together. Unless the user explicitly asks about price or value, or gives a budget, omit product prices, price-based pros and cons, and affordability commentary, even when the source review mentions them. This exception applies to price-related source limitations; preserve other relevant limitations. Display recorded purchase links and coupons as usual. Offer more only when additional suitable, unshown reviews have been verified. When adding externally sourced factual claims, cite pages that directly support each claim; one ingredient-specific or product-specific page does not establish claims about other ingredients or products merely because it belongs to the same database. For a sentence covering several different items, provide the supporting sources for all of them or narrow the sentence to what the cited page supports. In follow-up answers, avoid repeating a full general explanation or caution paragraph already given when the context has not changed. Keep any necessary reminder brief, preserve product-specific limitations, and explain new or changed risks when relevant. For a price question about a named product or a request with a budget, read its review details and use recorded_p…

Capabilities

Browse Selected Product ReviewsFind A Product Or BrandRead Reviews And Recorded Purchase Offers

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/plugin_asdk_app_6adb72c168ec8191b1d2e0809ed69232"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/plugin_asdk_app_6adb72c168ec8191b1d2e0809ed69232)