@vespermcp/mcp-server 1.2.9 → 1.2.11

package/build/cleaning/cleaner.js

@@ -63,7 +63,13 @@ export class DataCleaner {
                         reject(new Error(result.error));
                     }
                     else {
-                        resolve(result);
+                        resolve({
+                            success: true,
+                            rows_affected: Number(result.rows_affected ?? 0),
+                            columns_affected: Number(result.columns_affected ?? 0),
+                            output_path: result.output_path,
+                            logs: Array.isArray(result.logs) ? result.logs : [],
+                        });
                     }
                 }
                 catch (e) {

package/build/cleaning/executor.js

@@ -14,7 +14,7 @@ export class PipelineExecutor {
     /**
      * Run the full Auto-Cleaning Pipeline on a dataset file
      */
-    async runPipeline(datasetId, filePath, outputFormat = "csv", onProgress) {
+    async runPipeline(datasetId, filePath, outputFormat = "parquet", onProgress) {
         // ... (logging setup)
         const log = (m) => {
             console.error(`[Pipeline] ${m}`);
@@ -26,9 +26,10 @@ export class PipelineExecutor {
         // 2. Generate Plan
         log(`Generating cleaning plan...`);
         const plan = await this.planner.generatePlan(datasetId, qualityReport);
+        const rules = plan.operations;
         // If no cleaning needed, we still might need format conversion
         const needsConversion = !filePath.toLowerCase().endsWith(`.${outputFormat}`);
-        if (plan.operations.length === 0 && !needsConversion) {
+        if (rules.length === 0 && !needsConversion) {
             log(`No cleaning or conversion needed.`);
             return {
                 initial_quality: qualityReport,
@@ -42,9 +43,9 @@ export class PipelineExecutor {
             };
         }
         // 3. Execute Plan (includes conversion if requested)
-        log(`Executing ${plan.operations.length} operations (Format: ${outputFormat})...`);
-        plan.operations.forEach(op => console.error(`  - ${op.type}: ${op.reason}`));
-        const cleaningResult = await this.cleaner.clean(filePath, plan.operations, outputFormat);
+        log(`Executing ${rules.length} operations (Format: ${outputFormat})...`);
+        rules.forEach(op => console.error(`  - ${op.type}: ${op.reason}`));
+        const cleaningResult = await this.cleaner.clean(filePath, rules, outputFormat);
         if (cleaningResult.success) {
             log(`Cleaning complete. Output: ${cleaningResult.output_path}`);
         }

package/build/cleaning/planner.js

@@ -110,6 +110,10 @@ export class CleaningPlanner {
         }
         return plan;
     }
+    async generateRules(datasetId, report, ruleSet, targetInfo) {
+        const plan = await this.generatePlan(datasetId, report, ruleSet, targetInfo);
+        return plan.operations;
+    }
     shouldFixType(col) {
         if (col.inferred_type && col.inferred_type.includes("Numeric") && (col.type.includes("String") || col.type.includes("Utf8"))) {
             return true;

package/build/index.js

@@ -469,11 +469,20 @@ async function handlePrepareJob(jobId, query, requirements) {
  */
 async function handleCleanJob(jobId, datasetId, ops) {
     const update = (updates) => jobManager.updateJob(jobId, updates);
-    let filePath = path.join(dataRoot, "data", "raw", `${datasetId.replace(/\//g, "_")}.csv`);
+    const safeId = datasetId.replace(/\//g, "_");
+    const parquetPath = path.join(dataRoot, "data", "raw", `${safeId}.parquet`);
+    const csvPath = path.join(dataRoot, "data", "raw", `${safeId}.csv`);
+    let filePath = parquetPath;
+    if (!fs.existsSync(filePath)) {
+        filePath = csvPath;
+    }
     if (datasetId === "demo" || !fs.existsSync(filePath)) {
-        const demoPath = path.join(dataRoot, "e2e_demo_output", "raw_data.csv");
-        if (fs.existsSync(demoPath))
-            filePath = demoPath;
+        const demoParquetPath = path.join(dataRoot, "e2e_demo_output", "raw_data.parquet");
+        const demoCsvPath = path.join(dataRoot, "e2e_demo_output", "raw_data.csv");
+        if (fs.existsSync(demoParquetPath))
+            filePath = demoParquetPath;
+        else if (fs.existsSync(demoCsvPath))
+            filePath = demoCsvPath;
         else
             throw new Error(`Data file not found for ${datasetId}`);
     }
@@ -714,7 +723,7 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
             },
             {
                 name: "export_dataset",
-                description: "Export a dataset to a local directory. Use format='feather' (default) for 5-10× faster writes than CSV. Add fast=true to skip quality/cleaning steps.",
+                description: "Export a dataset to a local directory. Use format='parquet' (default) for efficient analytics and broad interoperability. Add fast=true to skip quality/cleaning steps.",
                 inputSchema: {
                     type: "object",
                     properties: {
@@ -729,7 +738,7 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
                         format: {
                             type: "string",
                             enum: ["feather", "parquet", "csv", "jsonl", "arrow"],
-                            description: "Output format. feather (fastest), parquet (best compression), csv (human-readable). Default: feather.",
+                            description: "Output format. parquet (default, analytics-friendly), feather (fast local IO), csv (human-readable).",
                         },
                         compression: {
                             type: "string",
@@ -800,7 +809,7 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
                         output_format: {
                             type: "string",
                             enum: ["feather", "parquet", "csv", "jsonl", "arrow"],
-                            description: "Output format (default: feather).",
+                            description: "Output format (default: parquet).",
                         },
                         compression: {
                             type: "string",
@@ -1144,12 +1153,19 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
         }
         case "analyze_quality": {
             const datasetId = String(request.params.arguments?.dataset_id);
-            let filePath = path.join(dataRoot, "data", "raw", `${datasetId.replace(/\//g, "_")}.csv`);
+            const safeId = datasetId.replace(/\//g, "_");
+            const parquetPath = path.join(dataRoot, "data", "raw", `${safeId}.parquet`);
+            const csvPath = path.join(dataRoot, "data", "raw", `${safeId}.csv`);
+            let filePath = fs.existsSync(parquetPath) ? parquetPath : csvPath;
             // Demo Fallback for easy testing
             if (datasetId === "demo" || !fs.existsSync(filePath)) {
-                const demoPath = path.join(dataRoot, "e2e_demo_output", "raw_data.csv");
-                if (fs.existsSync(demoPath)) {
-                    filePath = demoPath;
+                const demoParquetPath = path.join(dataRoot, "e2e_demo_output", "raw_data.parquet");
+                const demoCsvPath = path.join(dataRoot, "e2e_demo_output", "raw_data.csv");
+                if (fs.existsSync(demoParquetPath)) {
+                    filePath = demoParquetPath;
+                }
+                else if (fs.existsSync(demoCsvPath)) {
+                    filePath = demoCsvPath;
                 }
                 else if (datasetId !== "demo") {
                     return {
@@ -1165,11 +1181,18 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
         }
         case "preview_cleaning": {
             const datasetId = String(request.params.arguments?.dataset_id);
-            let filePath = path.join(dataRoot, "data", "raw", `${datasetId.replace(/\//g, "_")}.csv`);
+            const safeId = datasetId.replace(/\//g, "_");
+            const parquetPath = path.join(dataRoot, "data", "raw", `${safeId}.parquet`);
+            const csvPath = path.join(dataRoot, "data", "raw", `${safeId}.csv`);
+            let filePath = fs.existsSync(parquetPath) ? parquetPath : csvPath;
             if (datasetId === "demo" || !fs.existsSync(filePath)) {
-                const demoPath = path.join(dataRoot, "e2e_demo_output", "raw_data.csv");
-                if (fs.existsSync(demoPath)) {
-                    filePath = demoPath;
+                const demoParquetPath = path.join(dataRoot, "e2e_demo_output", "raw_data.parquet");
+                const demoCsvPath = path.join(dataRoot, "e2e_demo_output", "raw_data.csv");
+                if (fs.existsSync(demoParquetPath)) {
+                    filePath = demoParquetPath;
+                }
+                else if (fs.existsSync(demoCsvPath)) {
+                    filePath = demoCsvPath;
                 }
                 else {
                     throw new McpError(ErrorCode.InvalidParams, `Local data file not found for ${datasetId}. Please run prepare_dataset first.`);
@@ -1291,7 +1314,7 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
             // If NOT fast mode, run quality/cleaning pipeline first (only for csv/parquet compat)
             if (!fastMode) {
                 const currentExt = path.extname(sourcePath).substring(1).toLowerCase();
-                const pipelineFmt = (requestedFormat === "csv" || requestedFormat === "parquet") ? requestedFormat : "csv";
+                const pipelineFmt = (requestedFormat === "csv" || requestedFormat === "parquet") ? requestedFormat : "parquet";
                 if (currentExt !== pipelineFmt) {
                     console.error(`[Export] Running quality/cleaning pipeline (use fast=true to skip)...`);
                     try {

package/build/ingestion/ingestor.js

@@ -152,7 +152,7 @@ export class DataIngestor {
     /**
      * Generates a safe local filename for a dataset ID
      */
-    getTargetPath(datasetId, extension = "csv") {
+    getTargetPath(datasetId, extension = "parquet") {
         const safeId = datasetId.replace(/\//g, "_").replace(/:/g, "_");
         return path.join(this.rawDataDir, `${safeId}.${extension}`);
     }

package/build/python/config.py

@@ -155,12 +155,7 @@ def _get_from_env(name: str) -> Optional[str]:
 
 
 def get_key(name: str) -> Optional[str]:
-    # 1) env vars (highest priority)
-    env_val = _get_from_env(name)
-    if env_val:
-        return env_val
-
-    # 2) keyring (secure)
+    # 1) keyring (secure)
     if HAS_KEYRING:
         try:
             val = keyring.get_password(SERVICE_NAME, name)
@@ -169,14 +164,21 @@ def get_key(name: str) -> Optional[str]:
         except Exception:
             pass
 
-    # 3) encrypted fallback config.toml
+    # 2) encrypted fallback config.toml
     fallback = _read_fallback_toml()
     enc = fallback.get(name)
-    if not enc:
-        return None
-    secret = _get_or_create_local_secret()
-    method = fallback.get("__method__", "fernet" if HAS_FERNET else "xor")
-    return _decrypt_value(enc, method, secret)
+    if enc:
+        secret = _get_or_create_local_secret()
+        method = fallback.get("__method__", "fernet" if HAS_FERNET else "xor")
+        dec = _decrypt_value(enc, method, secret)
+        if dec:
+            return dec
+
+    # 3) env vars (fallback only)
+    env_val = _get_from_env(name)
+    if env_val:
+        return env_val
+    return None
 
 
 def set_key(name: str, value: str) -> Dict[str, str]:

package/build/python/kaggle_engine.py

@@ -20,14 +20,13 @@ def _ensure_auth() -> Dict[str, Any]:
         }
 
     # Priority:
-    # 1) Existing env vars
-    # 2) secure local store (keyring or ~/.vesper/config.toml)
+    # 1) secure local store (keyring or ~/.vesper/config.toml)
+    # 2) existing env vars
     # 3) ~/.kaggle/kaggle.json handled by KaggleApi.authenticate()
-    if not os.getenv("KAGGLE_USERNAME") or not os.getenv("KAGGLE_KEY"):
-        keys = get_all()
-        if keys.get("kaggle_username") and keys.get("kaggle_key"):
-            os.environ["KAGGLE_USERNAME"] = keys["kaggle_username"]
-            os.environ["KAGGLE_KEY"] = keys["kaggle_key"]
+    keys = get_all()
+    if keys.get("kaggle_username") and keys.get("kaggle_key"):
+        os.environ["KAGGLE_USERNAME"] = keys["kaggle_username"]
+        os.environ["KAGGLE_KEY"] = keys["kaggle_key"]
 
     api = KaggleApi()
     try:

package/build/python/vesper/core/asset_downloader.py

@@ -62,23 +62,19 @@ class AssetDownloader:
 
     @staticmethod
     def _hydrate_kaggle_credentials() -> None:
-        username = os.getenv("KAGGLE_USERNAME")
-        key = os.getenv("KAGGLE_KEY")
-
-        if not username or not key:
-            try:
-                from config import get_all  # type: ignore
-                keys = get_all() or {}
-            except Exception:
-                keys = {}
+        try:
+            from config import get_all  # type: ignore
+            keys = get_all() or {}
+        except Exception:
+            keys = {}
 
-            username = username or keys.get("kaggle_username")
-            key = key or keys.get("kaggle_key")
+        username = keys.get("kaggle_username") or os.getenv("KAGGLE_USERNAME")
+        key = keys.get("kaggle_key") or os.getenv("KAGGLE_KEY")
 
-            if username:
-                os.environ["KAGGLE_USERNAME"] = str(username)
-            if key:
-                os.environ["KAGGLE_KEY"] = str(key)
+        if username:
+            os.environ["KAGGLE_USERNAME"] = str(username)
+        if key:
+            os.environ["KAGGLE_KEY"] = str(key)
 
         username = os.getenv("KAGGLE_USERNAME")
         key = os.getenv("KAGGLE_KEY")

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@vespermcp/mcp-server",
-  "version": "1.2.9",
+  "version": "1.2.11",
   "description": "AI-powered dataset discovery, quality analysis, and preparation MCP server with multimodal support (text, image, audio, video)",
   "type": "module",
   "main": "build/index.js",
@@ -91,5 +91,6 @@
     "tsx": "^4.21.0",
     "typescript": "^5.9.3",
     "vitest": "^4.0.17"
-  }
+  },
+  "packageManager": "pnpm@10.18.1+sha512.77a884a165cbba2d8d1c19e3b4880eee6d2fcabd0d879121e282196b80042351d5eb3ca0935fa599da1dc51265cc68816ad2bddd2a2de5ea9fdf92adbec7cd34"
 }

package/src/python/config.py

@@ -155,12 +155,7 @@ def _get_from_env(name: str) -> Optional[str]:
 
 
 def get_key(name: str) -> Optional[str]:
-    # 1) env vars (highest priority)
-    env_val = _get_from_env(name)
-    if env_val:
-        return env_val
-
-    # 2) keyring (secure)
+    # 1) keyring (secure)
     if HAS_KEYRING:
         try:
             val = keyring.get_password(SERVICE_NAME, name)
@@ -169,14 +164,21 @@ def get_key(name: str) -> Optional[str]:
         except Exception:
             pass
 
-    # 3) encrypted fallback config.toml
+    # 2) encrypted fallback config.toml
     fallback = _read_fallback_toml()
     enc = fallback.get(name)
-    if not enc:
-        return None
-    secret = _get_or_create_local_secret()
-    method = fallback.get("__method__", "fernet" if HAS_FERNET else "xor")
-    return _decrypt_value(enc, method, secret)
+    if enc:
+        secret = _get_or_create_local_secret()
+        method = fallback.get("__method__", "fernet" if HAS_FERNET else "xor")
+        dec = _decrypt_value(enc, method, secret)
+        if dec:
+            return dec
+
+    # 3) env vars (fallback only)
+    env_val = _get_from_env(name)
+    if env_val:
+        return env_val
+    return None
 
 
 def set_key(name: str, value: str) -> Dict[str, str]:

package/src/python/kaggle_engine.py

@@ -20,14 +20,13 @@ def _ensure_auth() -> Dict[str, Any]:
         }
 
     # Priority:
-    # 1) Existing env vars
-    # 2) secure local store (keyring or ~/.vesper/config.toml)
+    # 1) secure local store (keyring or ~/.vesper/config.toml)
+    # 2) existing env vars
     # 3) ~/.kaggle/kaggle.json handled by KaggleApi.authenticate()
-    if not os.getenv("KAGGLE_USERNAME") or not os.getenv("KAGGLE_KEY"):
-        keys = get_all()
-        if keys.get("kaggle_username") and keys.get("kaggle_key"):
-            os.environ["KAGGLE_USERNAME"] = keys["kaggle_username"]
-            os.environ["KAGGLE_KEY"] = keys["kaggle_key"]
+    keys = get_all()
+    if keys.get("kaggle_username") and keys.get("kaggle_key"):
+        os.environ["KAGGLE_USERNAME"] = keys["kaggle_username"]
+        os.environ["KAGGLE_KEY"] = keys["kaggle_key"]
 
     api = KaggleApi()
     try:

package/src/python/vesper/core/asset_downloader.py

@@ -62,23 +62,19 @@ class AssetDownloader:
 
     @staticmethod
     def _hydrate_kaggle_credentials() -> None:
-        username = os.getenv("KAGGLE_USERNAME")
-        key = os.getenv("KAGGLE_KEY")
-
-        if not username or not key:
-            try:
-                from config import get_all  # type: ignore
-                keys = get_all() or {}
-            except Exception:
-                keys = {}
+        try:
+            from config import get_all  # type: ignore
+            keys = get_all() or {}
+        except Exception:
+            keys = {}
 
-            username = username or keys.get("kaggle_username")
-            key = key or keys.get("kaggle_key")
+        username = keys.get("kaggle_username") or os.getenv("KAGGLE_USERNAME")
+        key = keys.get("kaggle_key") or os.getenv("KAGGLE_KEY")
 
-            if username:
-                os.environ["KAGGLE_USERNAME"] = str(username)
-            if key:
-                os.environ["KAGGLE_KEY"] = str(key)
+        if username:
+            os.environ["KAGGLE_USERNAME"] = str(username)
+        if key:
+            os.environ["KAGGLE_KEY"] = str(key)
 
         username = os.getenv("KAGGLE_USERNAME")
         key = os.getenv("KAGGLE_KEY")