fix: Code changes for "CodeQL" issues fix

Services/appconfig/aiservice/appsettings.dev.json.template

@@ -27,6 +27,9 @@
     },
     "GapAnalysisProcessing": {
       "processwatcherUrl": "{{ gapanalysisprocesswatcherurl }}"
+    },
+    "AntiSSRF": {
+    "AllowedDomains": "microsoft.seismic.com"
     }
   }
 

Services/src/esg-ai-doc-analysis/CFS.SK.Sustainability.AI.Host/ESRSHttpServiceMapper.cs

@@ -229,7 +229,7 @@ public static void UseESRSEndpoint(this WebApplication app)
                     var content = new StringContent(serializedParamJson, Encoding.UTF8, "application/json");
 
                     //Invoke Process Watcher
-                    var response = await appContext.httpClient.PostAsync(config["GapAnalysisProcessing:processwatcherUrl"], content);
+                    var response = await appContext.httpClient.PostAsync(config["GapAnalysisProcessing:processwatcherUrl"], content); // CodeQL [SM03781] We are reading this value from appsettings.json, this is not an user input
 
                     return Results.Accepted(locationUrl, gapAnalysisServiceRequest);
                 }

Services/src/esg-ai-doc-analysis/CFS.SK.Sustainability.AI/DocumentManager.cs

@@ -169,6 +169,11 @@ await this.memoryWebClient.ImportDocumentAsync(content: fileStream.BaseStream,
 
         async public Task<DocumentServiceResult> RegisterDocumentFromFileUrl(RegisterDocumentFromFileUrlServiceRequest serviceRequest)
         {
+            //Validate URL to prevent SSRF attack
+            if (AntiSsrfValidation(serviceRequest.FileUrl))
+            {
+                throw new Exception("AntiSSRF validation failed - Invalid or unauthorized URL provided");
+            }
             //Download file from URL then take only fileName from URL
             //the file location URL in the document will be mixed with SAS token to get it.
             //sample url - https://microsoft.seismic.com/app?ContentId=d6e9f9bb-70d4-4845-a2ad-dd25ecc343d6#/doccenter/a5266a70-9230-4c1e-a553-c5bddcb7a896/doc/%252Fdde0caec0e-9236-f21b-2991-5868e63d3984%252FdfYTZjNDRiZDMtMzEwZS1kNWZkLTNjOGEtNjliYWJjMjhmMmUw%252CPT0%253D%252CUGl0Y2ggRGVjaw%253D%253D%252Flffb13c1f1-d960-4bbe-8685-000afbf5a67f//?mode=view&parentPath=sessionStorage
@@ -303,6 +308,49 @@ async public Task<MemoryAnswer> AskAboutDocumentSummary(string DocumentId, strin
                 return await this.memoryWebClient.AskAsync(Question, filter: new MemoryFilter().ByDocument(DocumentId));
             }
         }
+
+        // Anti-SSRF validation method
+        private bool AntiSsrfValidation(string url)
+        {
+            if (string.IsNullOrWhiteSpace(url))
+            {
+                return true; // Invalid URL
+            }
 
+            if (!Uri.TryCreate(url, UriKind.Absolute, out Uri? uri))
+            {
+                return true; // Invalid URL format
+            }
+
+            //the parameter flows to the validation method
+            bool isInvalidUri = !IsInAllowedDomain(uri) || !IsHttpsScheme(uri);
+
+            //validation method call flows to boolean return statement
+            return isInvalidUri;
+        }
+
+        //Domain validation helper
+        private bool IsInAllowedDomain(Uri uri)
+        {
+            var allowedDomainsConfig = this.config["AntiSSRF:AllowedDomains"];
+            if (string.IsNullOrWhiteSpace(allowedDomainsConfig))
+            {
+                throw new InvalidOperationException("AllowedDomains configuration is missing");
+            }
+            // Add this line to parse the config string into an array
+            var allowedDomains = allowedDomainsConfig.Split(',', StringSplitOptions.RemoveEmptyEntries)
+                                                    .Select(d => d.Trim())
+                                                    .ToArray();
+
+            return allowedDomains.Any(domain => 
+                uri.Host.Equals(domain, StringComparison.OrdinalIgnoreCase) ||
+                uri.Host.EndsWith($".{domain}", StringComparison.OrdinalIgnoreCase));
+        }
+
+        // HTTPS validation helper
+        private bool IsHttpsScheme(Uri uri)
+        {
+            return uri.Scheme == Uri.UriSchemeHttps;
+        }
     }
 }

Services/src/esg-ai-doc-analysis/CFS.SK.Sustainability.AI/ESRSGapAnalysisManager.cs

@@ -23,6 +23,7 @@
 using CFS.SK.Sustainability.AI.Storage.GapAnalysis;
 using Microsoft.Identity.Client;
 using Azure.Identity;
+using System.Text.RegularExpressions;
 
 
 namespace CFS.SK.Sustainability.AI
@@ -35,6 +36,10 @@
         private readonly GapAnalysisJobRepository _gapAnalysisJobRepository;
         private readonly IConfiguration _config;
 
+        // Regex to validate file names (only allows letters, digits, dot, dash, underscore)
+        private static readonly Regex ValidFileNameRegex = 
+        new Regex(@"^[A-Za-z0-9._\-]+$", RegexOptions.Compiled);
+
         public ESRSGapAnalysisManager(ApplicationContext appContext,
                                         ILogger<ESRSGapAnalysisManager> logger,
                                         ESRSDisclosureRetriever esrsDisclosureRetriever,
@@ -196,7 +201,13 @@
             //Create BlobClient
             //Get Current YYYYMMDDHHMMSS
             var jobId = DateTime.Now.ToString("yyyyMMddHHmmss");
-            var fileName = $"GAPAnalysisReport-{disclosure_number}-{jobId}";
+            // Sanitize disclosure_number to remove invalid characters
+            var sanitizedDisclosureNumber = new string(disclosure_number
+                .Where(c => char.IsLetterOrDigit(c) || c == '-' || c == '_')
+                .ToArray());
+            var fileName = $"GAPAnalysisReport-{sanitizedDisclosureNumber}-{jobId}";
+            //validate file name
+            EnsureSafeSimpleFileName(fileName);
             var blobClient = blobContainerClient.GetBlobClient($"{fileName}.md");
             using (Stream stream = new MemoryStream(Encoding.UTF8.GetBytes(analysis_resultString)))
             {
@@ -209,6 +220,8 @@
 
             //Create BlobClient
             var htmlFileName = $"{fileName}.html";
+            //validate html file name
+            EnsureSafeSimpleFileName(htmlFileName);
             blobClient = blobContainerClient.GetBlobClient(htmlFileName);
             byte[] byteArray_html = MarkdownHtmlConverter.Convert(analysis_resultString);
 
@@ -226,6 +239,7 @@
             //Convert HTML to PDF
             //Need to extra work for convert html to pdf
             var pdfFileName = $"{fileName}.pdf";
+            EnsureSafeSimpleFileName(pdfFileName); // Validate pdffilename
             HtmlPdfConverter.Convert(htmlFileName, pdfFileName);
 
             //if Pdf file is exist then upload to the blob
@@ -236,7 +250,7 @@
                 blobClient = blobContainerClient.GetBlobClient(pdfFileName);
                 using (Stream stream = new FileStream(pdfFileName, FileMode.Open))
                 {
                     await blobClient.UploadAsync(stream, true);
                 };
 
                 //Client SAS Token Url for the file
@@ -245,7 +259,7 @@
                 //Delete pdf file
                 System.IO.File.Delete(pdfFileName);
                 //Delete html file
-                System.IO.File.Delete(htmlFileName);
+                System.IO.File.Delete(htmlFileName);// CodeQL [SM00414] This variable is not based on user input, so no need to handle the Code QL issue.
             }
             else
             {
@@ -254,7 +268,8 @@
                 throw new Exception("PDF File is not converted");
             }
 
-            var metaDataFileName = $"GAPAnalysisReport-{disclosure_number}-{jobId}-meta.json";
+            var metaDataFileName = $"GAPAnalysisReport-{sanitizedDisclosureNumber}-{jobId}-meta.json";
+            EnsureSafeSimpleFileName(metaDataFileName);// Validate metadatafilename
             blobClient = blobContainerClient.GetBlobClient(metaDataFileName);
 
             using (Stream stream = new MemoryStream(Encoding.UTF8.GetBytes(JsonSerializer.Serialize(disclosureDescription))))
@@ -293,5 +308,20 @@
             return gapAnalysis_response;
             //return result.GetValue<string>();
         }
+
+        // Validate simple file name to prevent path traversal attacks
+        private static void EnsureSafeSimpleFileName(string name)
+        {
+            if (string.IsNullOrWhiteSpace(name))
+                throw new ArgumentException("File name is empty or null.");
+
+            if (name.Contains("..") || name.Contains("/") || name.Contains("\\"))
+                throw new ArgumentException("Invalid file name (contains path components or traversal).");
+
+            // Whitelist chars: letters, digits, dash, underscore, dot
+            if (!ValidFileNameRegex.IsMatch(name))
+                throw new ArgumentException("Invalid file name.");
+
+        }
     }
 }

Services/src/esg-ai-doc-analysis/CFS.SK.Sustainability.AI/Utils/HtmlPdfConverter.cs

@@ -14,21 +14,28 @@ internal static class HtmlPdfConverter
     {
         public static bool Convert(string sourceHtmlFilePath, string targetPdfFilePath)
         {
-            var escapedSourceHtmlFilePath = sourceHtmlFilePath.Replace("\"", "\\\"");
-            var escapedTargetPdfFilePath = targetPdfFilePath.Replace("\"", "\\\"");
+            // Validate file paths to prevent command injection
+            ValidateFilePath(sourceHtmlFilePath);
+            ValidateFilePath(targetPdfFilePath);
+
             var process = new Process()
             {
                 StartInfo = new ProcessStartInfo
                 {
                     FileName = IsWindows() ? "wkhtmltopdf.exe" : "/usr/bin/wkhtmltopdf",
-                    Arguments = $"--encoding UTF-8 -q \"{escapedSourceHtmlFilePath}\" \"{escapedTargetPdfFilePath}\"",
                     RedirectStandardInput = false,
                     RedirectStandardOutput = true,
                     RedirectStandardError = true,
                     UseShellExecute = false,
                     CreateNoWindow = true
                 }
             };
+            // Use ArgumentList to safely pass arguments (prevents command injection)
+            process.StartInfo.ArgumentList.Add("--encoding");
+            process.StartInfo.ArgumentList.Add("UTF-8");
+            process.StartInfo.ArgumentList.Add("-q");
+            process.StartInfo.ArgumentList.Add(sourceHtmlFilePath);
+            process.StartInfo.ArgumentList.Add(targetPdfFilePath);
 
             process.ErrorDataReceived += (process, data) =>
             {
@@ -54,5 +61,16 @@ private static bool IsWindows()
         {
             return System.Runtime.InteropServices.RuntimeInformation.IsOSPlatform(System.Runtime.InteropServices.OSPlatform.Windows);
         }
+
+        // Simple validation to prevent command injection
+        private static void ValidateFilePath(string path)
+        {
+            if (string.IsNullOrEmpty(path)) 
+                throw new ArgumentException("Invalid path");
+
+            // Block command injection characters
+            if (path.Any(c => ";|&`$<>\n\r".Contains(c)))
+                throw new ArgumentException("Path contains dangerous characters");
+        }
     }
 }

Services/src/kernel-memory/extensions/Postgres/Postgres/Db/PostgresDbClient.cs

@@ -140,6 +140,10 @@ public async Task CreateTableAsync(
         CancellationToken cancellationToken = default)
     {
         var origInputTableName = tableName;
+
+        // Validate tableName parameter before using it in SQL construction
+        PostgresSchema.ValidateTableName(origInputTableName);
+
         tableName = this.WithSchemaAndTableNamePrefix(tableName);
         this._log.LogTrace("Creating table: {0}", tableName);
 
@@ -154,12 +158,12 @@ public async Task CreateTableAsync(
                 using NpgsqlCommand cmd = connection.CreateCommand();
 
                 var lockId = GenLockId(tableName);
-
-#pragma warning disable CA2100 // SQL reviewed
+                
+                #pragma warning disable CA2100 // SQL reviewed
                 if (!string.IsNullOrEmpty(this._createTableSql))
                 {
                     cmd.CommandText = this._createTableSql
-                        .Replace(PostgresConfig.SqlPlaceholdersTableName, tableName, StringComparison.Ordinal)
+                        .Replace(PostgresConfig.SqlPlaceholdersTableName, tableName, StringComparison.Ordinal) // CodeQL [SM03934] tableName parameter is validated by PostgresSchema.ValidateTableName to prevent SQL injection
                         .Replace(PostgresConfig.SqlPlaceholdersVectorSize, $"{vectorSize}", StringComparison.Ordinal)
                         .Replace(PostgresConfig.SqlPlaceholdersLockId, $"{lockId}", StringComparison.Ordinal);