Enhance AI validation with progress tracking and prompt debugging

inventory-server/src/prompts/product-validation.txt

@@ -1,6 +1,6 @@
-I will provide a JSON array with product data. Process the array by combining all products from validData and invalidData arrays into a single array, excluding any fields starting with “__”, such as “__index” or “__errors”. Process each product according to the reference guidelines below. If a field is not included in the data, do not include it in your response unless the specific field guidelines below say otherwise. Please respond with:
+I will provide a JSON array with product data. Process the array by combining all products from validData and invalidData arrays into a single array, excluding any fields starting with “__”, such as “__index” or “__errors”. Process each product according to the reference guidelines below. If a field is not included in the data, do not include it in your response unless the specific field guidelines below say otherwise. 
 
-Respond in the following JSON format:
+Respond in the following JSON structure in minified format (single line, no whitespace):
 {
   "correctedData": [], // Array of corrected products
   "changes": [], // Array of strings describing each change made
@@ -11,7 +11,203 @@ Using the provided guidelines, focus on:
 1. Correcting typos and any incorrect spelling or grammar
 2. Standardizing product names
 3. Correcting and enhancing descriptions by adding details, keywords, and SEO-friendly language
-4. Fixing any obvious errors in measurements, prices, or quantities
+4. Fixing any obvious errors or inconsistencies between similar products in measurements, prices, or quantities
 5. Adding correct categories, themes, and colors
 
-Use only the provided data and your own knowledge to make changes. Do not make assumptions or make up information that you're not sure about. If you're unable to make a change you're confident about, leave the field as is.
+Use only the provided data and your own knowledge to make changes. Do not make assumptions or make up information that you're not sure about. If you're unable to make a change you're confident about, leave the field as is. All data passed in should be validated, corrected, and returned. All values should be strings. Do not leave out any fields that were present in the original data.
+
+----------PRODUCT FIELD GUIDELINES----------
+
+Fields: supplier, private_notes, company, line, subline, artist
+Changes: Not allowed
+Required: Return if present in the original data
+Instructions: If present, return these fields exactly as provided with no changes
+
+Fields: upc, supplier_no, notions_no, item_number
+Changes: Formatting only
+Required: Return if present in the original data
+Instructions: If present, trim outside white space and return these fields exactly as provided with no other changes
+
+Fields: hts_code
+Changes: Minimal, you can correct formatting, obvious errors or inconsistencies
+Required: Return if present in the original data
+Instructions: If present, trim white space and any characters that are not a number or decimal point and return as a string
+
+Fields: image_url
+Changes: Formatting only
+Required: Return if present in the original data
+Instructions: If present, convert all comma-separated values to valid https:// URLs and return
+
+Fields: msrp, cost_each
+Changes: Minimal, you can correct formatting, obvious errors or inconsistencies
+Required: Return if present in the original data
+Instructions: If present, strip any currency symbols and return as a string with exactly two decimal places, even if the last place is a 0.
+
+Fields: qty_per_unit, case_qty
+Changes: Minimal, you can correct formatting, obvious errors or inconsistencies
+Required: Return if present in the original data
+Instructions: If present, strip non-numeric characters and return
+
+Fields: ship_restrictions
+Changes: Only add a value if it's not already present
+Required: You must always return a value for this field, even if it's not provided in the original data
+Instructions: Always return a value exactly as provided, or return 0 if no value is provided. Do not leave this field out even if it's not provided.
+
+Fields: eta
+Changes: Minimal, you can correct formatting, obvious errors or inconsistencies
+Required: Return if present in the original data
+Instructions: If present, return a full month name, day is optional, no year ever (e.g. “January” or “March 3”). This value is not required if not provided.
+
+Fields: name
+Changes: Allowed to conform to guidelines, to fix typos or formatting
+Required: You must always return a value for this field, even if it's not provided in the original data
+Instructions: Always return a value that is corrected and enhanced per additional guidelines below
+
+Fields: description
+Changes: Full creative control allowed within guidelines
+Required: You must always return a value for this field, even if it's not provided in the original data
+Instructions: Always return a value that is corrected and enhanced per additional guidelines below
+
+Fields: weight, length, width, height
+Changes: Allowed to correct obvious errors or inconsistencies or to add missing values
+Required: You must always return a value for this field, even if it's not provided in the original data
+Instructions: Always return a reasonable value (weights in ounces and dimensions in inches) that is validated against similar provided products and your knowledge of general object measurements (e.g. a sheet of paper is not going to be 3 inches thick, a pack of stickers is not going to be 250 ounces, this sheet of paper is very likely going to be the same size as that other sheet of paper from the same line). If a value is unusual or unreasonable, change it to match similar products or to be more reasonable. Do not return 0 or null for any of these fields.
+
+Fields: coo
+Changes: Formatting only
+Required: Return if present in the original data
+Instructions: If present, return a valid two character country code, using capital letters
+
+Fields: tax_cat
+Changes: Allowed to correct obvious errors or inconsistencies or to add missing values
+Required: You must always return a value for this field, even if it's not provided in the original data
+Instructions: Always return a valid numerical tax code ID from the Available Tax Codes array below. Give preference to the value provided, but correct it if another value is more accurate. You must return a value for this field. 0 should be the default value in most cases.
+
+Fields: size_cat
+Changes: Allowed to correct obvious errors or inconsistencies or to add missing values
+Required: Return if present in the original data or if not present and applicable
+Instructions: If present or if applicable, return a valid numerical size category ID from the Available Size Categories array below. Give preference to the value provided, but correct it if another value is more accurate. A value is not required if none of the size categories apply.
+
+Fields: themes
+Changes: Allowed to correct obvious errors or inconsistencies or to add missing values
+Required: Return if present in the original data or if not present and applicable
+Instructions: If present, confirm that each provided theme matches what you understand to be a theme of the product. Remove any themes that do not match and add any themes that are missing. Most products will have zero or one theme. Return a comma-separated list of numerical theme IDs from the Available Themes array below. If you choose a sub-theme, you do not need to include its parent theme in the list.
+
+Fields: colors
+Changes: Allowed to correct obvious errors or inconsistencies or to add missing values
+Required: Return if present in the original data or if not present and applicable
+Instructions: If present or if applicable, return a comma-separated list of numerical color IDs from the Available Colors array below, using the product name as the primary guide. A value is not required if none of the colors apply. Most products will have zero colors.
+
+Fields: categories
+Changes: Allowed to correct obvious errors or inconsistencies or to add missing values
+Required: You must always return at least one value for this field, even if it's not provided in the original data
+Instructions: Always return a comma-separated list of one or more valid numerical category IDs from the Available Categories array below. Give preference to the values provided, particularly if the other information isn't enough to determine a category, but correct them or add new categories if another value is more accurate. Do not return categories in the Deals or Black Friday categories, and strip these from the list if present. If you choose a subcategory at any level, you do not need to include its parent categories in the list. You must return at least one category.
+
+----------PRODUCT NAMING GUIDELINES----------
+If there's only one of this type of product in a line: [Line Name] [Product Name] - [Company]
+Example: "Cosmos Infinity Chipboard - Stamperia"
+Example: "Serene Petals 6x6 Paper Pad - Prima"
+
+Multiple similar products in a line: [Differentiator] [Product Type] - [Line Name] - [Company]
+Example: "Ice & Shells Stencil - Arctic Antarctic - Stamperia"
+Example: "Astronomy Paper - Cosmos Infinity - Stamperia"
+
+Standalone products: [Product Name] - [Company]
+Example: "Hedwig Puffy Stickers - Paper House Productions"
+Example: "Heart Tree Dies - Lawn Fawn"
+
+Color-based products: [Color] [Product Name] - [Company]
+Example: "Green Valley Enamel Dots - Altenew"
+Example: "Magenta Aqua Pigment - Brutus Monroe"
+
+Complex products: [Differentiator] [Line] [Product Type] - [Company]
+Example: "Size 6 Round Black Velvet Watercolor Brush - Silver Brush Limited" (Size 6 Round is the differentiator, Black Velvet is the line, Watercolor Brush is the product type)
+
+These should not be included in the name, unless there are multiple products that are otherwise identical:
+- Product size
+- Product weight
+- Number of pages
+- How many are in the package
+
+Naming Conventions:
+- Paper sizes: Use "12x12", "8x8", "6x6" (no spaces or units of measure)
+- Company names must match backend exactly
+- Always capitalize every word in the name, including short articles like "The" and "An"
+- Use "Idea-ology" (not "idea-ology" or "Ideaology")
+- All stamps are "Stamp Set" (not "Clear Stamps" or "Rubber Stamps")
+- All dies are "Dies" or "Die" (not "Die Set")
+- Brands with their own naming conventions should be respected, such as "Doodle Cuts" for dies from Doodlebug
+
+Special Brand Rules - Ranger: 
+Format: [Product Name] - [Designer Line] - Ranger
+Possible Designers: Dylusions, Dina Wakley MEdia, Simon Hurley create., Wendy Vecchi
+Example: "Stacked Stencil - Dina Wakley MEdia - Ranger"
+
+Special Brand Rules - Tim Holtz products from Ranger:
+Format: [Color] [Product Name/Type] - Tim Holtz Distress - Ranger
+Example: "Mermaid Lagoon Tim Holtz Distress Oxide Ink Pad - Ranger"
+
+Special Brand Rules - Tim Holtz products from Sizzix or Stampers Anonymous:
+Format: [Product Name] [Product Type] by Tim Holtz - [Company]
+Example: "Leaf Fragments Thinlits Dies by Tim Holtz - Sizzix"
+
+Special Brand Rules - Tim Holtz products from Advantus/Idea-ology:
+Format: [Product Name] - Tim Holtz Idea-ology
+Example: "Tiny Vials - Tim Holtz Idea-ology"
+
+Special Brand Rules - Dies from Sizzix:
+Include die type plus "Dies" or "Die"
+Examples:
+"Art Nouveau 3-D Textured Impressions Embossing Folder - Sizzix"
+"Pocket Pals Thinlits Dies - Sizzix"
+"Butterfly Wishes Framelits Dies & Stamps - Sizzix"
+
+Important Notes
+- Ensure that product names are consistent across all products of the same type
+- Use the minimum amount of information needed to uniquely identify the product
+- Put detailed specifications in the product description, not its name
+
+Incorrect example: MVP Rugby - Collection Pack - Photoplay
+Notes: there should be no dash between the line and the product
+
+Incorrect Example: A2 Easel Cards - Black - Photoplay
+Notes: the differentiating factor should come first: “Black A2 Easel Cards - Photoplay”. Size is ok to include here because this is the name printed on the package.
+
+Incorrect Example: 6” - Scriber Needle Modeling Tool
+Notes: this product only comes in one size, so 6” isn’t needed. The company name should also be included.
+
+Incorrect Example: Slick - White - Tulip Dimensional Fabric Paint 4oz
+Notes: color should be first, then type, then product, then company, so “White Slick Dimensional Fabric Paint - Tulip”. It appears there’s only one size available so no need to differentiate in the name.
+
+Incorrect Example: Silhouette Adhesive Cork Sheets 5”X7” 8/Pkg
+Notes: should be “Adhesive Cork Sheets - Silhouette”
+
+Incorrect Example: Galaxy - Opaque - American Crafts Color Pour Resin Dyes
+Notes: “Galaxy Opaque Dye Set - Color Pour Resin - American Crafts”
+
+Incorrect Example: Slate - Lion Brand Truboo Yarn
+Notes: [Differentiator] [Line] [Product Type] - [Company] : “Slate Truboo Yarn - Lion Brand”
+
+Incorrect Example: Rose Quartz Dylusions Shimmer Paint
+Notes: “Rose Quartz Shimmer Paint - Dylusions - Ranger”
+
+
+----------PRODUCT DESCRIPTION GUIDELINES----------
+Product descriptions are an extremely important part of the listing and are the most important part of your response. Care should be taken to ensure they are correct, helpful, and SEO-friendly.
+
+If a description is provided in the data, use it as a starting point. Correct any spelling errors, typos, poor grammar, or awkward phrasing. If necessary and you have the information, add more details, describe how the customer could use it, etc. Use complete sentences and keep SEO in mind.
+
+If no description is provided, make one up using the product name, the information you have, and the other provided guidelines. At minimum, a description should be one complete sentence that starts with a capital letter and ends with a period. Unless the product is extremely complex, 2-4 sentences is usually sufficient if you have enough information.
+
+Important Notes:
+- Every description should state exactly what's included in the product (e.g. "Includes one 12x12 sheet of patterned cardstock." or "Includes one 6x12 sheet with 27 unique stickers." or "Includes 55 pieces." or "Package includes machine, power cord, 12 sheets of cardstock, 3 dies, and project instructions.")
+- Do not use the word "our" in the description (this usually shows up when we copy a description from the manufacturer). Instead use "these" or "[Company name] [product]" or similar. (e.g. don't use "Our journals are hand-made in the USA", instead use "These journals are hand made..." or "Archer & Olive journals are handmade...")
+- Don't include fluff like “this is perfect for all your paper crafts” most of the time. If the product helps to solve a unique problem or has a unique feature, by all means describe it, but if it’s just a normal sheet of paper or pack of stickers, you don’t have to pretend like it’s the best thing ever.
+- State as many facts as you can about the product, considering the viewpoint of the customer and what they would want to know when looking at it. They probably want to know dimensions, what products it’s compatible with, how thick the paper is, how many sheets are included, whether the sheets are double-sided or not, which items are in the kit, etc. Say as much as you possibly can with the information that you have.
+- !!DO NOT make up information if you aren't sure about it. A minimal correct description is better than a long incorrect one!!
+
+Avoid/remove:
+- The word "Imported"
+- Any warnings about Prop 65, choking hazards, etc
+- The manufacturer's name if it's included as the very first thing in the description
+- Any statement similar to "comes in a variety of colors, each sold separately"

inventory-server/src/routes/ai-validation.js

@@ -38,51 +38,119 @@ router.get('/debug', async (req, res) => {
     console.log('Debug endpoint called');
     const pool = req.app.locals.pool;
     
-    // Load taxonomy data first
+    // Get a real supplier, company, and artist ID from the database
-    console.log('Loading taxonomy data...');
+    const [suppliers] = await pool.query('SELECT supplierid FROM suppliers LIMIT 1');
-    const taxonomy = await getTaxonomyData(pool);
+    const [companies] = await pool.query('SELECT cat_id FROM product_categories WHERE type = 1 LIMIT 1');
-    console.log('Taxonomy data loaded:', {
+    const [artists] = await pool.query('SELECT cat_id FROM product_categories WHERE type = 40 LIMIT 1');
-      categoriesCount: taxonomy.categories.length,
+    
-      themesCount: taxonomy.themes.length,
+    // Create a sample product with real IDs
-      colorsCount: taxonomy.colors.length,
+    const productsToUse = [{
-      taxCodesCount: taxonomy.taxCodes.length,
+      supplierid: suppliers[0]?.supplierid || 1234,
-      sizeCategoriesCount: taxonomy.sizeCategories.length
+      company: companies[0]?.cat_id || 567,
-    });
+      artist: artists[0]?.cat_id || 890
-
+    }];
-    // Then load the prompt
+    
-    console.log('Loading prompt...');
+    return await generateDebugResponse(pool, productsToUse, res);
-    const currentPrompt = await loadPrompt(pool);
-    const sampleData = [{ name: "Sample Product" }];
-    const fullPrompt = currentPrompt + '\n' + JSON.stringify(sampleData);
-
-    const response = {
-      cacheStatus: {
-        isCacheValid: isCacheValid(),
-        lastUpdated: cache.lastUpdated ? new Date(cache.lastUpdated).toISOString() : null,
-        timeUntilExpiry: cache.lastUpdated ? 
-          Math.max(0, CACHE_TTL - (Date.now() - cache.lastUpdated)) / 1000 + ' seconds' : 
-          'expired',
-      },
-      taxonomyStats: taxonomy ? {
-        categories: countItems(taxonomy.categories),
-        themes: taxonomy.themes.length,
-        colors: taxonomy.colors.length,
-        taxCodes: taxonomy.taxCodes.length,
-        sizeCategories: taxonomy.sizeCategories.length
-      } : null,
-      basePrompt: currentPrompt,
-      sampleFullPrompt: fullPrompt,
-      promptLength: fullPrompt.length,
-    };
-
-    console.log('Sending response with stats:', response.taxonomyStats);
-    res.json(response);
   } catch (error) {
     console.error('Debug endpoint error:', error);
     res.status(500).json({ error: error.message });
   }
 });
 
+// New POST endpoint for debug with products
+router.post('/debug', async (req, res) => {
+  try {
+    console.log('Debug POST endpoint called');
+    const pool = req.app.locals.pool;
+    const { products } = req.body;
+    
+    console.log('Received products:', {
+      isArray: Array.isArray(products),
+      length: products?.length,
+      firstProduct: products?.[0],
+      lastProduct: products?.[products?.length - 1]
+    });
+    
+    if (!Array.isArray(products)) {
+      console.error('Invalid input: products is not an array');
+      return res.status(400).json({ error: 'Products must be an array' });
+    }
+
+    if (products.length === 0) {
+      console.error('Invalid input: products array is empty');
+      return res.status(400).json({ error: 'Products array cannot be empty' });
+    }
+    
+    // Clean the products array to remove any internal fields
+    const cleanedProducts = products.map(product => {
+      const { __errors, __index, ...cleanProduct } = product;
+      return cleanProduct;
+    });
+    
+    return await generateDebugResponse(pool, cleanedProducts, res);
+  } catch (error) {
+    console.error('Debug POST endpoint error:', error);
+    res.status(500).json({ error: error.message });
+  }
+});
+
+// Helper function to generate debug response
+async function generateDebugResponse(pool, productsToUse, res) {
+  // Load taxonomy data first
+  console.log('Loading taxonomy data...');
+  const taxonomy = await getTaxonomyData(pool);
+  console.log('Taxonomy data loaded:', {
+    categoriesCount: taxonomy.categories.length,
+    themesCount: taxonomy.themes.length,
+    colorsCount: taxonomy.colors.length,
+    taxCodesCount: taxonomy.taxCodes.length,
+    sizeCategoriesCount: taxonomy.sizeCategories.length,
+    suppliersCount: taxonomy.suppliers.length,
+    companiesCount: taxonomy.companies.length,
+    artistsCount: taxonomy.artists.length
+  });
+
+  // Load the prompt using the same function used by validation
+  console.log('Loading prompt...');
+  const prompt = await loadPrompt(pool, productsToUse);
+  const fullPrompt = prompt + '\n' + JSON.stringify(productsToUse);
+
+  const response = {
+    cacheStatus: {
+      isCacheValid: isCacheValid(),
+      lastUpdated: cache.lastUpdated ? new Date(cache.lastUpdated).toISOString() : null,
+      timeUntilExpiry: cache.lastUpdated ? 
+        Math.max(0, CACHE_TTL - (Date.now() - cache.lastUpdated)) / 1000 + ' seconds' : 
+        'expired',
+    },
+    taxonomyStats: taxonomy ? {
+      categories: countItems(taxonomy.categories),
+      themes: taxonomy.themes.length,
+      colors: taxonomy.colors.length,
+      taxCodes: taxonomy.taxCodes.length,
+      sizeCategories: taxonomy.sizeCategories.length,
+      suppliers: taxonomy.suppliers.length,
+      companies: taxonomy.companies.length,
+      artists: taxonomy.artists.length,
+      // Add filtered counts when products are provided
+      filtered: productsToUse ? {
+        suppliers: taxonomy.suppliers.filter(([id]) => 
+          productsToUse.some(p => Number(p.supplierid) === Number(id))).length,
+        companies: taxonomy.companies.filter(([id]) => 
+          productsToUse.some(p => Number(p.company) === Number(id))).length,
+        artists: taxonomy.artists.filter(([id]) => 
+          productsToUse.some(p => Number(p.artist) === Number(id))).length
+      } : null
+    } : null,
+    basePrompt: prompt,
+    sampleFullPrompt: fullPrompt,
+    promptLength: fullPrompt.length,
+  };
+
+  console.log('Sending response with stats:', response.taxonomyStats);
+  return res.json(response);
+}
+
 // Helper function to count total items in hierarchical structure
 function countItems(items) {
   return items.reduce((count, item) => {
@@ -167,6 +235,46 @@ async function getTaxonomyData(pool) {
   // Fetch size categories
   const [sizeCategories] = await pool.query('SELECT cat_id, name FROM product_categories WHERE type=50 ORDER BY name');
 
+  // Fetch suppliers
+  const [suppliers] = await pool.query(`
+    SELECT supplierid, companyname as name 
+    FROM suppliers 
+    WHERE companyname <> '' 
+    ORDER BY companyname
+  `);
+
+  // Fetch companies (type 1)
+  const [companies] = await pool.query(`
+    SELECT cat_id, name 
+    FROM product_categories 
+    WHERE type = 1 
+    ORDER BY name
+  `);
+
+  // Fetch artists (type 40)
+  const [artists] = await pool.query(`
+    SELECT cat_id, name 
+    FROM product_categories 
+    WHERE type = 40 
+    ORDER BY name
+  `);
+
+  // Fetch lines (type 2)
+  const [lines] = await pool.query(`
+    SELECT cat_id, name 
+    FROM product_categories 
+    WHERE type = 2 
+    ORDER BY name
+  `);
+
+  // Fetch sub-lines (type 3)
+  const [subLines] = await pool.query(`
+    SELECT cat_id, name 
+    FROM product_categories 
+    WHERE type = 3 
+    ORDER BY name
+  `);
+
   // Format categories into a hierarchical structure
   const formatHierarchy = (items, level = 1, parentId = null) => {
     return items
@@ -198,7 +306,12 @@ async function getTaxonomyData(pool) {
     themes: formatThemes(themes),
     colors: colors.map(c => [c.color, c.name]),
     taxCodes: (taxCodes || []).map(tc => [tc.tax_code_id, tc.name]),
-    sizeCategories: (sizeCategories || []).map(sc => [sc.cat_id, sc.name])
+    sizeCategories: (sizeCategories || []).map(sc => [sc.cat_id, sc.name]),
+    suppliers: suppliers.map(s => [s.supplierid, s.name]),
+    companies: companies.map(c => [c.cat_id, c.name]),
+    artists: artists.map(a => [a.cat_id, a.name]),
+    lines: lines.map(l => [l.cat_id, l.name]),
+    subLines: subLines.map(sl => [sl.cat_id, sl.name])
   };
   cache.lastUpdated = Date.now();
 
@@ -206,18 +319,113 @@ async function getTaxonomyData(pool) {
 }
 
 // Load the prompt from file and inject taxonomy data
-async function loadPrompt(pool) {
+async function loadPrompt(pool, productsToValidate = null) {
-  if (cache.validationPrompt && isCacheValid()) {
-    return cache.validationPrompt;
-  }
-
   const promptPath = path.join(__dirname, '..', 'prompts', 'product-validation.txt');
   const basePrompt = await fs.readFile(promptPath, 'utf8');
   
   // Get taxonomy data
   const taxonomy = await getTaxonomyData(pool);
-  
+
-  // Format taxonomy data for the prompt
+  // Add system instructions to the prompt
+  const systemInstructions = `You are a specialized e-commerce product data processor for a crafting supplies website tasked with providing complete, correct, appealing, and SEO-friendly product listings. You should write professionally, but in a friendly and engaging tone.
+
+`;
+
+  // If we have products to validate, create a filtered prompt
+  if (productsToValidate) {
+    console.log('Creating filtered prompt for products:', productsToValidate);
+    
+    // Extract unique values from products for non-core attributes
+    const uniqueValues = {
+      supplierIds: new Set(),
+      companyIds: new Set(),
+      artistIds: new Set(),
+      lineIds: new Set(),
+      subLineIds: new Set()
+    };
+
+    // Collect any values that exist in the products
+    productsToValidate.forEach(product => {
+      Object.entries(product).forEach(([key, value]) => {
+        if (value === undefined || value === null) return;
+        
+        // Map field names to their respective sets
+        const fieldMap = {
+          supplierid: 'supplierIds',
+          supplier: 'supplierIds',
+          company: 'companyIds',
+          artist: 'artistIds',
+          line: 'lineIds',
+          subline: 'subLineIds'
+        };
+        
+        if (fieldMap[key]) {
+          uniqueValues[fieldMap[key]].add(Number(value));
+        }
+      });
+    });
+
+    console.log('Unique values collected:', {
+      suppliers: Array.from(uniqueValues.supplierIds),
+      companies: Array.from(uniqueValues.companyIds),
+      artists: Array.from(uniqueValues.artistIds),
+      lines: Array.from(uniqueValues.lineIds),
+      subLines: Array.from(uniqueValues.subLineIds)
+    });
+
+    // Create mixed taxonomy with filtered non-core data and full core data
+    const mixedTaxonomy = {
+      // Keep full data for core attributes
+      categories: taxonomy.categories,
+      themes: taxonomy.themes,
+      colors: taxonomy.colors,
+      taxCodes: taxonomy.taxCodes,
+      sizeCategories: taxonomy.sizeCategories,
+      // For non-core data, only include items that are actually used
+      suppliers: taxonomy.suppliers.filter(([id]) => uniqueValues.supplierIds.has(Number(id))),
+      companies: taxonomy.companies.filter(([id]) => uniqueValues.companyIds.has(Number(id))),
+      artists: taxonomy.artists.filter(([id]) => uniqueValues.artistIds.has(Number(id))),
+      lines: taxonomy.lines.filter(([id]) => uniqueValues.lineIds.has(Number(id))),
+      subLines: taxonomy.subLines.filter(([id]) => uniqueValues.subLineIds.has(Number(id)))
+    };
+
+    console.log('Filtered taxonomy counts:', {
+      suppliers: mixedTaxonomy.suppliers.length,
+      companies: mixedTaxonomy.companies.length,
+      artists: mixedTaxonomy.artists.length,
+      lines: mixedTaxonomy.lines.length,
+      subLines: mixedTaxonomy.subLines.length
+    });
+
+    // Format taxonomy data for the prompt, only including sections with values
+    const taxonomySection = `
+All Available Categories:
+${JSON.stringify(mixedTaxonomy.categories)}
+
+All Available Themes:
+${JSON.stringify(mixedTaxonomy.themes)}
+
+All Available Colors:
+${JSON.stringify(mixedTaxonomy.colors)}
+
+All Available Tax Codes:
+${JSON.stringify(mixedTaxonomy.taxCodes)}
+
+All Available Size Categories:
+${JSON.stringify(mixedTaxonomy.sizeCategories)}${mixedTaxonomy.suppliers.length ? `\n\nSuppliers Used In This Data:\n${JSON.stringify(mixedTaxonomy.suppliers)}` : ''}${mixedTaxonomy.companies.length ? `\n\nCompanies Used In This Data:\n${JSON.stringify(mixedTaxonomy.companies)}` : ''}${mixedTaxonomy.artists.length ? `\n\nArtists Used In This Data:\n${JSON.stringify(mixedTaxonomy.artists)}` : ''}${mixedTaxonomy.lines.length ? `\n\nLines Used In This Data:\n${JSON.stringify(mixedTaxonomy.lines)}` : ''}${mixedTaxonomy.subLines.length ? `\n\nSub-Lines Used In This Data:\n${JSON.stringify(mixedTaxonomy.subLines)}` : ''}
+
+----------Here is the product data to validate----------`;
+
+    // Return the filtered prompt without caching
+    return systemInstructions + basePrompt + '\n' + taxonomySection;
+  }
+
+  // For debug/display purposes, if no products provided and cache is valid, return cached prompt
+  if (!productsToValidate && cache.validationPrompt && isCacheValid()) {
+    return cache.validationPrompt;
+  }
+
+  // Generate and cache the full unfiltered prompt
   const taxonomySection = `
 Available Categories:
 ${JSON.stringify(taxonomy.categories)}
@@ -234,10 +442,22 @@ ${JSON.stringify(taxonomy.taxCodes)}
 Available Size Categories:
 ${JSON.stringify(taxonomy.sizeCategories)}
 
+Available Suppliers:
+${JSON.stringify(taxonomy.suppliers)}
+
+Available Companies:
+${JSON.stringify(taxonomy.companies)}
+
+Available Artists:
+${JSON.stringify(taxonomy.artists)}
+
+Available Shipping Restrictions:
+${JSON.stringify(taxonomy.shippingRestrictions)}
+
 Here is the product data to validate:`;
 
-  // Combine the prompt sections
+  // Cache the full prompt only when no specific products are provided
-  cache.validationPrompt = basePrompt + '\n' + taxonomySection;
+  cache.validationPrompt = systemInstructions + basePrompt + '\n' + taxonomySection;
   cache.lastUpdated = Date.now();
 
   return cache.validationPrompt;
@@ -256,19 +476,15 @@ router.post('/validate', async (req, res) => {
       return res.status(400).json({ error: 'Products must be an array' });
     }
 
-    // Load the prompt and append the products data
+    // Load the prompt with the products data to filter taxonomy
-    const basePrompt = await loadPrompt(req.app.locals.pool);
+    const prompt = await loadPrompt(req.app.locals.pool, products);
-    const fullPrompt = basePrompt + '\n' + JSON.stringify(products);
+    const fullPrompt = prompt + '\n' + JSON.stringify(products);
     console.log('📝 Generated prompt:', fullPrompt);
 
     console.log('🤖 Sending request to OpenAI...');
     const completion = await openai.chat.completions.create({
-      model: "gpt-4-turbo-preview",
+      model: "gpt-4o",
       messages: [
-        {
-          role: "system",
-          content: "You are a specialized e-commerce product data processor for a crafting supplies website tasked with providing complete, correct, appealing, and SEO-friendly product listings. You should write professionally, but in a friendly and engaging tone."
-        },
         {
           role: "user",
           content: fullPrompt

inventory/src/lib/react-spreadsheet-import/src/steps/ValidationStep/ValidationStep.tsx

@@ -65,6 +65,7 @@ import {
   DialogTitle,
 } from "@/components/ui/dialog"
 import { ScrollArea } from "@/components/ui/scroll-area"
+import { Code } from "@/components/ui/code"
 
 type Props<T extends string> = {
   initialData: (Data<T> & Meta)[]
@@ -724,6 +725,26 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
     isOpen: false,
   });
 
+  const [aiValidationProgress, setAiValidationProgress] = useState<{
+    isOpen: boolean;
+    status: string;
+    step: number;
+  }>({
+    isOpen: false,
+    status: "",
+    step: 0,
+  });
+
+  const [currentPrompt, setCurrentPrompt] = useState<{
+    isOpen: boolean;
+    prompt: string | null;
+    isLoading: boolean;
+  }>({
+    isOpen: false,
+    prompt: null,
+    isLoading: false,
+  });
+
   // Memoize filtered data to prevent recalculation on every render
   const filteredData = useMemo(() => {
     if (!filterByErrors) return data
@@ -967,7 +988,18 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
   const handleAiValidation = async () => {
     try {
       setIsAiValidating(true);
-      console.log('Sending data for AI validation:', data);
+      setAiValidationProgress({
+        isOpen: true,
+        status: "Preparing data for validation...",
+        step: 1
+      });
+      console.log('Sending data for validation:', data);
+
+      setAiValidationProgress(prev => ({
+        ...prev,
+        status: "Sending data to AI service and awaiting response...",
+        step: 2
+      }));
 
       const response = await fetch(`${config.apiUrl}/ai-validation/validate`, {
         method: 'POST',
@@ -981,6 +1013,12 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
         throw new Error('AI validation failed');
       }
 
+      setAiValidationProgress(prev => ({
+        ...prev,
+        status: "Processing AI response...",
+        step: 3
+      }));
+
       const result = await response.json();
       console.log('AI validation response:', result);
       
@@ -988,6 +1026,12 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
         throw new Error(result.error || 'AI validation failed');
       }
 
+      setAiValidationProgress(prev => ({
+        ...prev,
+        status: "Applying corrections...",
+        step: 4
+      }));
+
       // Update the data with AI suggestions
       if (result.correctedData && Array.isArray(result.correctedData)) {
         // Log the differences
@@ -1027,6 +1071,16 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
         isOpen: true,
       });
 
+      setAiValidationProgress(prev => ({
+        ...prev,
+        status: "Validation complete!",
+        step: 5
+      }));
+
+      setTimeout(() => {
+        setAiValidationProgress(prev => ({ ...prev, isOpen: false }));
+      }, 1000);
+
     } catch (error) {
       console.error('AI Validation Error:', error);
       toast({
@@ -1034,11 +1088,82 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
         description: error instanceof Error ? error.message : "An error occurred during AI validation",
         variant: "destructive",
       });
+      setAiValidationProgress(prev => ({
+        ...prev,
+        status: "Validation failed",
+        step: -1
+      }));
     } finally {
       setIsAiValidating(false);
     }
   };
 
+  // Add function to fetch current prompt
+  const showCurrentPrompt = async () => {
+    try {
+      setCurrentPrompt(prev => ({ ...prev, isLoading: true, isOpen: true }));
+      
+      // Debug log the data being sent
+      console.log('Sending products data:', {
+        dataLength: data.length,
+        firstProduct: data[0],
+        lastProduct: data[data.length - 1]
+      });
+
+      // Clean the data to ensure we only send what's needed
+      const cleanedData = data.map(item => {
+        const { __errors, __index, ...rest } = item;
+        return rest;
+      });
+
+      console.log('Cleaned data sample:', {
+        length: cleanedData.length,
+        firstProduct: cleanedData[0],
+        lastProduct: cleanedData[cleanedData.length - 1]
+      });
+      
+      // Use POST to send products in request body
+      const response = await fetch(`${config.apiUrl}/ai-validation/debug`, {
+        method: 'POST',
+        headers: {
+          'Content-Type': 'application/json',
+        },
+        body: JSON.stringify({ products: cleanedData })
+      });
+
+      if (!response.ok) {
+        const errorText = await response.text();
+        console.error('Debug endpoint error:', {
+          status: response.status,
+          statusText: response.statusText,
+          body: errorText
+        });
+        throw new Error(`Failed to fetch prompt: ${response.status} ${response.statusText}`);
+      }
+      const debugData = await response.json();
+      
+      // Log the response stats
+      console.log('Debug response stats:', {
+        promptLength: debugData.promptLength,
+        taxonomyStats: debugData.taxonomyStats
+      });
+      
+      setCurrentPrompt(prev => ({ 
+        ...prev, 
+        prompt: debugData.sampleFullPrompt,
+        isLoading: false 
+      }));
+    } catch (error) {
+      console.error('Error fetching prompt:', error);
+      toast({
+        title: "Error",
+        description: error instanceof Error ? error.message : "Failed to fetch current prompt",
+        variant: "destructive",
+      });
+      setCurrentPrompt(prev => ({ ...prev, isLoading: false }));
+    }
+  };
+
   return (
     <div className="flex h-[calc(100vh-9.5rem)] flex-col">
       <CopyDownDialog 
@@ -1047,6 +1172,25 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
         onConfirm={executeCopyDown}
         fieldLabel={copyDownField?.label || ""}
       />
+      <Dialog open={currentPrompt.isOpen} onOpenChange={(open) => setCurrentPrompt(prev => ({ ...prev, isOpen: open }))}>
+        <DialogContent className="max-w-4xl h-[80vh]">
+          <DialogHeader>
+            <DialogTitle>Current AI Prompt</DialogTitle>
+            <DialogDescription>
+              This is the exact prompt that would be sent to the AI for validation
+            </DialogDescription>
+          </DialogHeader>
+          <ScrollArea className="flex-1">
+            {currentPrompt.isLoading ? (
+              <div className="flex items-center justify-center h-full">
+                <Loader2 className="h-8 w-8 animate-spin" />
+              </div>
+            ) : (
+              <Code className="whitespace-pre-wrap p-4">{currentPrompt.prompt}</Code>
+            )}
+          </ScrollArea>
+        </DialogContent>
+      </Dialog>
       <AlertDialog open={showSubmitAlert} onOpenChange={setShowSubmitAlert}>
         <AlertDialogPortal>
           <AlertDialogOverlay className="z-[1400]" />
@@ -1074,6 +1218,34 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
           </AlertDialogContent>
         </AlertDialogPortal>
       </AlertDialog>
+      <Dialog open={aiValidationProgress.isOpen} onOpenChange={() => {}}>
+        <DialogContent className="sm:max-w-md">
+          <DialogHeader>
+            <DialogTitle>AI Validation Progress</DialogTitle>
+          </DialogHeader>
+          <div className="space-y-4 py-4">
+            <div className="flex items-center gap-4">
+              <div className="flex-1">
+                <div className="h-2 w-full bg-secondary rounded-full overflow-hidden">
+                  <div 
+                    className="h-full bg-primary transition-all duration-500" 
+                    style={{ 
+                      width: `${(aiValidationProgress.step / 5) * 100}%`,
+                      backgroundColor: aiValidationProgress.step === -1 ? 'var(--destructive)' : undefined
+                    }}
+                  />
+                </div>
+              </div>
+              <div className="text-sm text-muted-foreground w-12 text-right">
+                {aiValidationProgress.step === -1 ? '❌' : `${Math.round((aiValidationProgress.step / 5) * 100)}%`}
+              </div>
+            </div>
+            <p className="text-center text-sm text-muted-foreground">
+              {aiValidationProgress.status}
+            </p>
+          </div>
+        </DialogContent>
+      </Dialog>
       <Dialog 
         open={aiValidationDetails.isOpen} 
         onOpenChange={(open) => setAiValidationDetails(prev => ({ ...prev, isOpen: open }))}
@@ -1141,6 +1313,14 @@ export const ValidationStep = <T extends string>({ initialData, file, onBack }:
                   )}
                   AI Validate
                 </Button>
+                <Button
+                  variant="outline"
+                  size="sm"
+                  onClick={showCurrentPrompt}
+                  disabled={data.length === 0}
+                >
+                  Show Prompt
+                </Button>
                 <div className="flex items-center gap-2">
                   <Switch
                     checked={filterByErrors}

inventory/src/pages/Import.tsx

@@ -40,7 +40,7 @@ const BASE_IMPORT_FIELDS = [
     label: "Supplier #",
     key: "supplier_no",
     description: "Supplier's product identifier",
-    alternateMatches: ["sku", "item#", "mfg item #", "item"],
+    alternateMatches: ["sku", "item#", "mfg item #", "item", "supplier #"],
     fieldType: { type: "input" },
     width: 180,
     validations: [
@@ -52,6 +52,7 @@ const BASE_IMPORT_FIELDS = [
     label: "Notions #",
     key: "notions_no",
     description: "Internal notions number",
+    alternateMatches: ["notions #"],
     fieldType: { type: "input" },
     width: 110,
     validations: [
@@ -109,7 +110,7 @@ const BASE_IMPORT_FIELDS = [
     label: "Qty Per Unit",
     key: "qty_per_unit",
     description: "Quantity of items per individual unit",
-    alternateMatches: ["inner pack", "inner", "min qty", "unit qty", "min. order qty"],
+    alternateMatches: ["inner pack", "inner", "min qty", "unit qty", "min. order qty", "supplier qty/unit"],
     fieldType: { type: "input" },
     width: 90,
     validations: [
@@ -121,7 +122,7 @@ const BASE_IMPORT_FIELDS = [
     label: "Cost Each",
     key: "cost_each",
     description: "Wholesale cost per unit",
-    alternateMatches: ["wholesale", "wholesale price"],
+    alternateMatches: ["wholesale", "wholesale price", "supplier cost each"],
     fieldType: { 
       type: "input",
       price: true 
@@ -264,7 +265,7 @@ const BASE_IMPORT_FIELDS = [
     label: "Country Of Origin",
     key: "coo",
     description: "2-letter country code (ISO)",
-    alternateMatches: ["coo"],
+    alternateMatches: ["coo", "country of origin"],
     fieldType: { type: "input" },
     width: 100,
     validations: [
@@ -275,7 +276,7 @@ const BASE_IMPORT_FIELDS = [
     label: "HTS Code",
     key: "hts_code",
     description: "Harmonized Tariff Schedule code",
-    alternateMatches: ["taric"],
+    alternateMatches: ["taric","hts"],
     fieldType: { type: "input" },
     width: 130,
     validations: [