@privacyscrubber/mcp-server 1.0.1 → 1.0.2

package/CONTRIBUTING.md

@@ -0,0 +1,52 @@
+# Contributing to PrivacyScrubber MCP Server
+
+We welcome contributions to the PrivacyScrubber Model Context Protocol (MCP) server integration, client configurations, and testing suites.
+
+---
+
+## 🛠️ Development Setup
+
+To set up a local development environment:
+
+1.  **Clone the Repository:**
+    ```bash
+    git clone https://github.com/moxno/privacyscrubber-mcp.git
+    cd privacyscrubber-mcp
+    ```
+
+2.  **Install Dependencies:**
+    ```bash
+    npm install
+    ```
+
+3.  **Run the Server Locally:**
+    You can spin up the server transport using:
+    ```bash
+    node index.js
+    ```
+    *Note: The server uses StdIO transport, so it will sit waiting for JSON-RPC input on standard streams.*
+
+---
+
+## 🧪 Testing Your Changes
+
+Before submitting a Pull Request, ensure that all integration and cryptographic verification tests pass cleanly.
+
+### 1. Cryptographic Signature Assertions
+Verify that local license key validations and RSA signatures parse correctly:
+```bash
+node test-mcp.js
+```
+
+### 2. End-to-End JSON-RPC Subprocess Suite
+Run the full simulation test which spawns the server, registers protocol versions, lists capabilities, processes files (TXT and DOCX), and asserts binary file rejections:
+```bash
+node test-mcp-deep.js
+```
+
+---
+
+## 🔒 Regex Rules & PII Detection Core
+The core detection engine and regex rules are bundled inside `scrubber-core.cjs` which maintains parity with the PrivacyScrubber Chrome Extension and web application. 
+
+If you notice a false positive or an uncaught entity type (e.g. name matching edge cases), please report it directly through the main project issues or email us at `support@privacyscrubber.com` so we can sync the ruleset across all platforms.

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Ilya Sibiryakov (BrandMeWeb)
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -1,24 +1,42 @@
 # @privacyscrubber/mcp-server
 
+[![NPM Version](https://img.shields.io/npm/v/@privacyscrubber/mcp-server?color=blue)](https://www.npmjs.com/package/@privacyscrubber/mcp-server)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![Smithery Compatible](https://smithery.ai/badge/@privacyscrubber/mcp-server)](https://smithery.ai/server/@privacyscrubber/mcp-server)
+[![Security: 100% Local](https://img.shields.io/badge/Security-100%25%20Local-emerald)](https://privacyscrubber.com)
+[![Parity: 100% Core Match](https://img.shields.io/badge/Parity-100%25%20Core%20Match-blueviolet)](https://privacyscrubber.com)
+
 **Zero-Trust Data Sanitization (ZTDS) Model Context Protocol (MCP) Server.**
-Locally scrubs PII, secrets, credentials, and custom patterns from files and text contexts before sending them to LLMs.
+Locally scrubs PII, secrets, credentials, and custom regex rules from files and text contexts before they reach remote LLM providers.
 
 ---
 
-## 🔒 Security & Privacy Guarantees
+## 🔒 Zero-Trust Data Flow
+
+All sensitive parameters, identifiers, and variables are intercepted locally inside your machine's RAM. They are replaced by tokens (e.g. `[EMAIL_1]`) before being sent to the AI. Once the AI responds, the tokens are safely swapped back to original values in your local context.
 
-*   **100% Offline / Local Execution:** All PII detection, regex matching, and tokenization happen entirely inside your local machine's RAM. No prompt context or file data is ever sent to any remote servers.
-*   **Volatile In-Memory Mapping:** The token replacement map (`sessionMap`) is kept in volatile RAM only. It is never persisted to disk, cookies, or database storage.
-*   **Support for Reverse-Scrubbing (Reveal):** Replaces tokens in LLM responses back with original values securely inside your local context.
+```text
+[Raw Input / Files] ──> [MCP sanitize_text] ──> [Masked Tokens] ──> [LLM API]
+                               │                                       │
+                        (In-Memory Map)                             (Result)
+                               │                                       │
+[Original Output] <─── [MCP reveal_text] <─────────────────────────────┘
+```
 
 ---
 
-## 🚀 Installation & Usage
+## 🚀 Installation
+
+### 1. Install via Smithery
+To automatically configure and run with your preferred client, install using Smithery:
+```bash
+npx -y @smithery/cli install @privacyscrubber/mcp-server --write-to-clients
+```
 
-### 1. Instant Run with NPX
-You can run the MCP server directly using `npx`:
+### 2. Instant Run with NPX
+Run the server directly without local installation:
 ```bash
-npx @privacyscrubber/mcp-server
+npx -y @privacyscrubber/mcp-server
 ```
 
 ---
@@ -26,7 +44,7 @@ npx @privacyscrubber/mcp-server
 ## ⚙️ Client Integrations
 
 ### Claude Desktop
-Add the server configuration to your Claude Desktop config file:
+Add this to your Claude Desktop config file:
 *   **macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
 *   **Windows:** `%APPDATA%\Claude\claude_desktop_config.json`
 
@@ -37,7 +55,7 @@ Add the server configuration to your Claude Desktop config file:
       "command": "npx",
       "args": ["-y", "@privacyscrubber/mcp-server"],
       "env": {
-        "PRIVACYSCRUBBER_KEY": "YOUR_OPTIONAL_PRO_OR_TEAMS_KEY"
+        "PRIVACYSCRUBBER_KEY": "YOUR_OPTIONAL_PRO_LICENSE_KEY"
       }
     }
   }
@@ -45,22 +63,93 @@ Add the server configuration to your Claude Desktop config file:
 ```
 
 ### Cursor / Windsurf
-1. Open settings (Settings -> Features -> MCP).
+1. Navigate to Settings -> Features -> MCP.
 2. Add new MCP server:
     *   **Name:** `privacyscrubber`
     *   **Type:** `command`
     *   **Command:** `npx -y @privacyscrubber/mcp-server`
-3. If you have a PRO key, add `PRIVACYSCRUBBER_KEY` as an environment variable in your system shell or configure it locally.
+3. Optional: Set `PRIVACYSCRUBBER_KEY` as an environment variable in your system shell.
+
+---
+
+## 🛠️ Provided Tools & JSON-RPC Specifications
+
+### 1. `sanitize_text`
+Redacts PII, secrets, API keys, and credentials from a text block and populates the volatile local replacement mapping.
+
+*   **Arguments:**
+    *   `text` (string, required): The raw content or logs to sanitize.
+    *   `profile` (string, optional): Gated industry detection profile (e.g., 'General', 'Dev', 'Medical', 'Legal', 'Compliance'). Defaults to 'General'.
+*   **JSON-RPC Call Example:**
+    ```json
+    {
+      "method": "tools/call",
+      "params": {
+        "name": "sanitize_text",
+        "arguments": {
+          "text": "Contact me at dev-key-1234 or jane.doe@company.com",
+          "profile": "General"
+        }
+      }
+    }
+    ```
+*   **Response Example:**
+    ```json
+    {
+      "content": [
+        {
+          "type": "text",
+          "text": "Contact me at [SECRET_1] or [EMAIL_1]"
+        }
+      ]
+    }
+    ```
+
+### 2. `reveal_text`
+Detokenizes the AI response back to the original values locally.
+
+*   **Arguments:**
+    *   `text` (string, required): The response from the LLM containing tokenized placeholders.
+*   **JSON-RPC Call Example:**
+    ```json
+    {
+      "method": "tools/call",
+      "params": {
+        "name": "reveal_text",
+        "arguments": {
+          "text": "Please reach out to [EMAIL_1] regarding the update."
+        }
+      }
+    }
+    ```
+*   **Response Example:**
+    ```json
+    {
+      "content": [
+        {
+          "type": "text",
+          "text": "Please reach out to jane.doe@company.com regarding the update."
+        }
+      ]
+    }
+    ```
+
+### 3. `sanitize_file`
+Reads a local file, extracts text, sanitizes it, and returns the redacted template for LLM analysis.
+*   **Supported Formats:** Plain text (source code, logs, CSV, JSON, markdown) and Microsoft Word (`.docx`) documents.
+*   **Arguments:**
+    *   `filePath` (string, required): Absolute file path to read and sanitize.
+    *   `profile` (string, optional): The industry detection profile.
 
 ---
 
-## 🛠️ Provided Tools
+## 🌐 Browser Extension & Web Client
 
-1.  `sanitize_text` — Sanitizes a raw string using the selected detection profile.
-2.  `reveal_text` — Detokenizes a response containing PII tokens back to the original text.
-3.  `sanitize_file` — Reads a local file, sanitizes its contents, and outputs the safe version (supports up to 10MB).
+Looking for real-time protection directly inside your web browser?
+*   **Chrome Extension:** Get the [PrivacyScrubber Chrome Extension](https://chromewebstore.google.com/detail/privacyscrubber-%E2%80%94-pii-red/pimoejgefeilajmmbpghifdmhdlkgjol) to sanitize prompts directly inside ChatGPT, Claude, and Gemini in real-time.
+*   **Web Sandbox:** Use the zero-server browser sanitization tools at [PrivacyScrubber Homepage](https://privacyscrubber.com/).
 
 ---
 
-## 📄 License & Commercial Key
-Standard use includes the **Free Tier** (limits to the `General` PII profile). To unlock 22+ specialized industry profiles (DevOps, Medical, Legal, Finance) and custom regex rules, acquire a license at [privacyscrubber.com/pricing](https://privacyscrubber.com/pricing).
+## 📄 License & Commercial Upgrade
+Standard use is free under the **Free Tier** (limits to the `General` PII profile). To unlock 22+ specialized industry profiles (DevOps, Medical, Legal, Finance) and custom regex rules, acquire a commercial license at [privacyscrubber.com/pricing](https://privacyscrubber.com/pricing).

package/SECURITY.md

@@ -0,0 +1,30 @@
+# Security Policy (SECURITY.md)
+
+PrivacyScrubber is built with a zero-server, client-side first architecture. We take data privacy and vulnerability management extremely seriously.
+
+---
+
+## 🔒 Security Posture & Guarantees
+
+The PrivacyScrubber MCP server operates under a strict **Zero-Trust Data Sanitization (ZTDS)** design:
+1.  **100% Offline / Local Redaction:** All PII extraction, pattern matching, and tokenization logic run locally inside your system's Node.js runtime process memory (volatile RAM).
+2.  **No Telemetry & No Remote Logs:** The server contains zero network request code for telemetry, prompts transmission, or log collection.
+3.  **Strict Gating Audits:** Cryptographic key checks (for PRO/TEAMS licenses) are validated locally on-device using a signed RSA public key signature verification. No license verification calls are made to any remote databases.
+
+### Verification (Airplane Mode Test)
+You can verify the zero-network posture at any time:
+1.  Disconnect your machine's internet access (Wi-Fi/Ethernet).
+2.  Launch the MCP server locally using Cursor, Windsurf, or Claude Desktop.
+3.  Execute file sanitization and detokenization. All tools remain 100% operational offline.
+
+---
+
+## 🐛 Reporting a Vulnerability
+
+If you discover a potential security vulnerability (e.g. regex engine backtracking denial of service, memory leaks, or local file access overflows), please do not report it in the public GitHub Issues tracker. 
+
+Instead, report it privately through our coordinated disclosure channel:
+- **Email:** `security@privacyscrubber.com`
+- **Response SLA:** Our team reviews all reports within **24 hours** and will provide a status update and remediation timeline.
+
+Thank you for helping keep the developer ecosystem secure!

package/index.js

@@ -113,6 +113,20 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
             }
           },
           required: ["text"]
+        },
+        outputSchema: {
+          type: "object",
+          properties: {
+            text: {
+              type: "string",
+              description: "The sanitized text output."
+            }
+          },
+          required: ["text"]
+        },
+        annotations: {
+          priority: 1.0,
+          audience: ["developer", "user"]
         }
       },
       {
@@ -127,6 +141,20 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
             }
           },
           required: ["text"]
+        },
+        outputSchema: {
+          type: "object",
+          properties: {
+            text: {
+              type: "string",
+              description: "The detokenized text with original values restored."
+            }
+          },
+          required: ["text"]
+        },
+        annotations: {
+          priority: 1.0,
+          audience: ["developer", "user"]
         }
       },
       {
@@ -135,7 +163,7 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
         inputSchema: {
           type: "object",
           properties: {
-            filePath: {
+            file_path: {
               type: "string",
               description: "Absolute path to the file to sanitize."
             },
@@ -144,7 +172,21 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
               description: "The detection profile to use. Available: 'General' (Free), or PRO profiles: 'Dev' (Engineering/Code), 'Medical', 'Pharma', 'Legal', 'Compliance', 'CCPA', 'Finance', 'Bizops', 'Sales', 'WealthMgmt', 'Insurance', 'Accounting', 'HR', 'Security', 'Marketing', 'Support', 'RealEstate', 'Agents', 'Academic', 'Creative', 'Tech', 'Personal'. Defaults to 'General'."
             }
           },
-          required: ["filePath"]
+          required: ["file_path"]
+        },
+        outputSchema: {
+          type: "object",
+          properties: {
+            text: {
+              type: "string",
+              description: "The sanitized content of the file."
+            }
+          },
+          required: ["text"]
+        },
+        annotations: {
+          priority: 1.0,
+          audience: ["developer", "user"]
         }
       }
     ]
@@ -169,7 +211,7 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
           content: [
             {
               type: "text",
-              text: `⚠️ [PrivacyScrubber] Advanced profile '${targetProfile}' is locked in the FREE tier. Falling back to 'General' profile.${reason}\nTo unlock 22+ advanced industry profiles, custom regex, and unlimited logs protection, set PRIVACYSCRUBBER_KEY to your PRO license key.\nGet a key at: https://privacyscrubber.com/pricing\n\n--- Sanitized Output (General Profile) ---\n`
+              text: `⚠️ [PrivacyScrubber] Advanced profile '${targetProfile}' is locked in the FREE tier. Falling back to 'General' profile.${reason}\nTo unlock 22+ advanced industry profiles, custom regex, and unlimited logs protection, set PRIVACYSCRUBBER_KEY to your PRO license key.\nGet a key at: https://privacyscrubber.com/pricing\nBrowser Extension: https://chromewebstore.google.com/detail/privacyscrubber-%E2%80%94-pii-red/pimoejgefeilajmmbpghifdmhdlkgjol\n\n--- Sanitized Output (General Profile) ---\n`
             },
             {
               type: "text",
@@ -204,7 +246,20 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
     }
 
     if (name === "sanitize_file") {
-      const { filePath, profile = "General" } = args;
+      const rawPath = args.file_path || args.filePath;
+      if (!rawPath) {
+        return {
+          isError: true,
+          content: [
+            {
+              type: "text",
+              text: "Error: Missing required parameter 'file_path'."
+            }
+          ]
+        };
+      }
+      const { profile = "General" } = args;
+      const filePath = rawPath;
       
       const resolvedPath = path.resolve(filePath);
       if (!fs.existsSync(resolvedPath)) {
@@ -246,7 +301,71 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
         };
       }
 
-      const content = fs.readFileSync(resolvedPath, "utf8");
+      if (resolvedPath.toLowerCase().endsWith(".docx")) {
+        try {
+          const mammoth = require('mammoth');
+          const result = await mammoth.extractRawText({ path: resolvedPath });
+          const content = result.value;
+          const targetProfile = profile.trim();
+          const isAdvanced = targetProfile.toLowerCase() !== "general";
+          const license = checkLicenseStatus();
+
+          let prefix = "";
+          let finalProfile = targetProfile;
+
+          if (isAdvanced && !license.isPro) {
+            const reason = license.error ? ` (${license.error})` : "";
+            prefix = `⚠️ [PrivacyScrubber] Advanced profile '${targetProfile}' is locked in the FREE tier. Falling back to 'General' profile.${reason}\nGet a PRO key at: https://privacyscrubber.com/pricing\nBrowser Extension: https://chromewebstore.google.com/detail/privacyscrubber-%E2%80%94-pii-red/pimoejgefeilajmmbpghifdmhdlkgjol\n\n`;
+            finalProfile = "General";
+          }
+
+          const sanitized = performSanitization(content, finalProfile);
+          return {
+            content: [
+              {
+                type: "text",
+                text: `${prefix}${sanitized}`
+              }
+            ]
+          };
+        } catch (docxError) {
+          return {
+            isError: true,
+            content: [
+              {
+                type: "text",
+                text: `Error: Failed to parse DOCX file: ${docxError.message}`
+              }
+            ]
+          };
+        }
+      }
+
+      const buffer = fs.readFileSync(resolvedPath);
+      
+      // Check for null bytes in the first 8KB to detect binary files
+      let isBinary = false;
+      const checkLen = Math.min(buffer.length, 8000);
+      for (let i = 0; i < checkLen; i++) {
+        if (buffer[i] === 0) {
+          isBinary = true;
+          break;
+        }
+      }
+
+      if (isBinary) {
+        return {
+          isError: true,
+          content: [
+            {
+              type: "text",
+              text: `Error: Binary file format detected. The local MCP server only supports plain text files (e.g., source code, logs, CSV, markdown, JSON). To sanitize complex formats like PDF or DOCX, please use the web interface at: https://privacyscrubber.com/`
+            }
+          ]
+        };
+      }
+
+      const content = buffer.toString("utf8");
       const targetProfile = profile.trim();
       const isAdvanced = targetProfile.toLowerCase() !== "general";
       const license = checkLicenseStatus();
@@ -256,7 +375,7 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
 
       if (isAdvanced && !license.isPro) {
         const reason = license.error ? ` (${license.error})` : "";
-        prefix = `⚠️ [PrivacyScrubber] Advanced profile '${targetProfile}' is locked in the FREE tier. Falling back to 'General' profile.${reason}\nGet a PRO key at: https://privacyscrubber.com/pricing\n\n`;
+        prefix = `⚠️ [PrivacyScrubber] Advanced profile '${targetProfile}' is locked in the FREE tier. Falling back to 'General' profile.${reason}\nGet a PRO key at: https://privacyscrubber.com/pricing\nBrowser Extension: https://chromewebstore.google.com/detail/privacyscrubber-%E2%80%94-pii-red/pimoejgefeilajmmbpghifdmhdlkgjol\n\n`;
         finalProfile = "General";
       }
 

package/llms.txt

@@ -0,0 +1,37 @@
+# PrivacyScrubber MCP Server
+
+> Zero-Trust Data Sanitization (ZTDS) Model Context Protocol (MCP) Server. Locally scrubs PII, secrets, credentials, and custom patterns from text and files before they reach remote LLM APIs.
+
+## Main URL
+- Repository: https://github.com/moxno/privacyscrubber-mcp
+- Web sandbox: https://privacyscrubber.com/
+- Chrome extension: https://chromewebstore.google.com/detail/privacyscrubber-%E2%80%94-pii-red/pimoejgefeilajmmbpghifdmhdlkgjol
+
+## Provided Tools
+
+The server runs locally in StdIO transport and exposes three tools:
+
+### sanitize_text
+Scrubs PII (names, emails, phones, locations), API keys, and passwords from input text.
+- Inputs:
+  - `text` (string, required): Raw text block or log segment to sanitize.
+  - `profile` (string, optional): Gated industry detection profile (e.g. 'General', 'Dev', 'Medical', 'Legal', 'Compliance'). Defaults to 'General'.
+- Output: Masked template string containing placeholders (e.g. `[EMAIL_1]`).
+
+### reveal_text
+Detokenizes AI responses by restoring original PII values locally in-memory.
+- Inputs:
+  - `text` (string, required): Response from the LLM containing tokens like `[EMAIL_1]`.
+- Output: Detokenized restored string.
+
+### sanitize_file
+Reads a local file, checks format compatibility, extracts raw text, and redacts PII.
+- Supported Formats: Source code files, CSV, JSON, markdown, text logs, and Microsoft Word (`.docx`) documents.
+- Binary formats (like PDF, ZIP) are automatically rejected with a redirection notice to prevent corrupting inputs.
+- Inputs:
+  - `filePath` (string, required): Absolute local file path.
+  - `profile` (string, optional): Industry detection profile.
+- Output: Sanitized file text template.
+
+## Environment Variables
+- `PRIVACYSCRUBBER_KEY` (string, optional): License key to unlock advanced industry detection profiles. If not set, the server runs on the Free tier and defaults all sanitizations to the 'General' profile.

package/manifest.json

@@ -0,0 +1,28 @@
+{
+  "manifest_version": "0.3",
+  "name": "privacyscrubber-mcp",
+  "version": "1.0.1",
+  "description": "Zero-Trust Data Sanitization (ZTDS) MCP Server for AI IDEs and Claude Desktop. Locally scrubs PII, secrets, and credentials before sending context to LLMs.",
+  "author": {
+    "name": "PrivacyScrubber",
+    "email": "legal@privacyscrubber.com"
+  },
+  "server": {
+    "type": "node",
+    "entry_point": "index.js",
+    "mcp_config": {
+      "command": "node",
+      "args": ["${__dirname}/index.js"],
+      "env": {
+        "PRIVACYSCRUBBER_KEY": "${user_config.PRIVACYSCRUBBER_KEY}"
+      }
+    }
+  },
+  "user_config": {
+    "PRIVACYSCRUBBER_KEY": {
+      "type": "string",
+      "title": "PrivacyScrubber License Key",
+      "description": "Optional PRO or TEAMS license key. Leave blank to use the Free tier."
+    }
+  }
+}

package/package.json

@@ -1,9 +1,13 @@
 {
   "name": "@privacyscrubber/mcp-server",
-  "version": "1.0.1",
+  "version": "1.0.2",
+  "mcpName": "io.github.moxno/privacyscrubber-mcp",
   "description": "Zero-Trust Data Sanitization (ZTDS) MCP Server for AI IDEs and Claude Desktop. Locally scrubs PII, secrets, and credentials before sending context to LLMs.",
   "main": "index.js",
   "type": "module",
+  "publishConfig": {
+    "access": "public"
+  },
   "bin": {
     "mcp-server": "index.js",
     "privacyscrubber-mcp": "index.js"
@@ -12,7 +16,8 @@
     "start": "node index.js"
   },
   "dependencies": {
-    "@modelcontextprotocol/sdk": "^0.6.0"
+    "@modelcontextprotocol/sdk": "^0.6.0",
+    "mammoth": "^1.8.0"
   },
   "keywords": [
     "mcp",
@@ -26,6 +31,14 @@
     "claude-desktop",
     "cursor"
   ],
-  "author": "Ilya Sibiryakov (Brand Me Web)",
-  "license": "MIT"
+  "author": "Ilya Sibiryakov (BrandMeWeb)",
+  "license": "MIT",
+  "homepage": "https://privacyscrubber.com/pii-mcp/",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/moxno/privacyscrubber-mcp.git"
+  },
+  "bugs": {
+    "url": "https://github.com/moxno/privacyscrubber-mcp/issues"
+  }
 }

package/server.json

@@ -0,0 +1,29 @@
+{
+  "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
+  "name": "io.github.moxno/privacyscrubber-mcp",
+  "description": "Zero-Trust PII & secrets sanitizer. Locally scrubs data in-memory before sending context to LLMs.",
+  "repository": {
+    "url": "https://github.com/moxno/privacyscrubber-mcp",
+    "source": "github"
+  },
+  "version": "1.0.2",
+  "packages": [
+    {
+      "registryType": "npm",
+      "identifier": "@privacyscrubber/mcp-server",
+      "version": "1.0.2",
+      "transport": {
+        "type": "stdio"
+      },
+      "environmentVariables": [
+        {
+          "description": "Optional PRO or TEAMS license key to unlock specialized detection profiles.",
+          "isRequired": false,
+          "format": "string",
+          "isSecret": true,
+          "name": "PRIVACYSCRUBBER_KEY"
+        }
+      ]
+    }
+  ]
+}

package/smithery.yaml

@@ -0,0 +1,20 @@
+# Smithery.ai configuration for PrivacyScrubber MCP Server
+startCommand:
+  type: "stdio"
+  configSchema:
+    type: "object"
+    properties:
+      PRIVACYSCRUBBER_KEY:
+        type: "string"
+        title: "PrivacyScrubber License Key"
+        description: "Optional PRO or TEAMS license key. Leave blank to run on the Free tier."
+  commandFunction: |-
+    (config) => ({
+      command: 'node',
+      args: ['./index.js'],
+      env: {
+        PRIVACYSCRUBBER_KEY: config.PRIVACYSCRUBBER_KEY || ''
+      }
+    })
+exampleConfig:
+  PRIVACYSCRUBBER_KEY: ""

package/test-mcp-deep.js

@@ -0,0 +1,209 @@
+import { spawn } from 'child_process';
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = path.dirname(__filename);
+
+const indexScript = path.resolve(__dirname, 'index.js');
+
+// Helper to send JSON-RPC and wait for response
+async function runMcpSession() {
+  console.log("Starting MCP Server process...");
+  
+  // Set license key to valid PRO key for advanced test
+  const validKey = "eyJ0eXBlIjoicHJvIiwiZXhwaXJlcyI6MjA5OTA2MjE3NCwiaXNzdWVkQXQiOjE3ODM3MDIxNzQsIm5vbmNlIjoiMmI4MTllN2ZmM2ZmOTFmYiJ9.ZmhaKDGf9vG0nTH0nyOy3EInsuUmzjBOk9IOyiqzt6Y5s3UG+2yRExuKBoeXWymbpJt3NIJNM9sVxxl+lcop5wqbi2LNtPY4MuwBRA7pO4nn3Bes5R0lxLbEYVE8Iiw3zbfK4uVYQ53BJ7El6JCeFKPJ5WbKUsPLsjb1Lr2iQzW3ODOMaM5jKdgAGcNpuWQ373D7SW7I03Jec9kvP5hL7j3u4DV8ZzuQFzy2Nh+uMH54suydg2sNIpxrRQyxpGv7rZjTO1KT+xzzqnjqX4Pein0e6GrC5E4JUEggADtXPFMKW5SEiWzeeirB5RUPMO/F927S5fFuOYHAfuar2FqErA==";
+  
+  const mcpProcess = spawn('node', [indexScript], {
+    env: {
+      ...process.env,
+      PRIVACYSCRUBBER_KEY: validKey
+    }
+  });
+
+  let buffer = '';
+  const pendingRequests = new Map();
+  let nextId = 1;
+
+  mcpProcess.stdout.on('data', (data) => {
+    buffer += data.toString();
+    // Parse JSON-RPC delimited by newlines
+    let newlineIndex;
+    while ((newlineIndex = buffer.indexOf('\n')) !== -1) {
+      const line = buffer.substring(0, newlineIndex).trim();
+      buffer = buffer.substring(newlineIndex + 1);
+      if (line) {
+        try {
+          const response = JSON.parse(line);
+          if (response.id && pendingRequests.has(response.id)) {
+            const resolve = pendingRequests.get(response.id);
+            pendingRequests.delete(response.id);
+            resolve(response);
+          }
+        } catch (e) {
+          console.error("Failed to parse incoming line as JSON:", line, e);
+        }
+      }
+    }
+  });
+
+  mcpProcess.stderr.on('data', (data) => {
+    console.error(`[Server Stderr]: ${data.toString().trim()}`);
+  });
+
+  const sendRequest = (method, params) => {
+    return new Promise((resolve) => {
+      const id = nextId++;
+      pendingRequests.set(id, resolve);
+      const requestObj = {
+        jsonrpc: '2.0',
+        id,
+        method,
+        params
+      };
+      mcpProcess.stdin.write(JSON.stringify(requestObj) + '\n');
+    });
+  };
+
+  try {
+    // 1. Initialize
+    console.log("--> Sending 'initialize' request...");
+    const initResponse = await sendRequest('initialize', {
+      protocolVersion: '2024-11-05',
+      capabilities: {},
+      clientInfo: { name: 'test-client', version: '1.0.0' }
+    });
+    console.log("<-- Initialize response received:", JSON.stringify(initResponse).substring(0, 150) + "...");
+
+    // 2. List tools
+    console.log("--> Sending 'tools/list' request...");
+    const toolsResponse = await sendRequest('tools/list', {});
+    const tools = toolsResponse.result?.tools || [];
+    console.log(`<-- Tools list received. Total tools: ${tools.length}`);
+    tools.forEach(t => console.log(`  - Tool: ${t.name} (${t.description.substring(0, 60)}...)`));
+
+    if (tools.length !== 3) {
+      throw new Error(`Expected 3 tools, got ${tools.length}`);
+    }
+
+    // 3. Test sanitize_text (General profile)
+    console.log("--> Calling 'sanitize_text' with General profile...");
+    const rawText = "Contact John Doe at john.doe@example.com or phone 555-0199.";
+    const sanitizeResponse = await sendRequest('tools/call', {
+      name: 'sanitize_text',
+      arguments: {
+        text: rawText,
+        profile: 'General'
+      }
+    });
+    const sanitizedText = sanitizeResponse.result?.content?.[0]?.text;
+    console.log("<-- Sanitized Output:", sanitizedText);
+    
+    if (sanitizedText.includes("john.doe@example.com") || sanitizedText.includes("555-0199")) {
+      throw new Error("Sanitization failed to redact email or phone number.");
+    }
+    console.log("✅ Sanitize text success.");
+
+    // 4. Test reveal_text (detokenization)
+    console.log("--> Calling 'reveal_text' to restore tokens...");
+    const revealResponse = await sendRequest('tools/call', {
+      name: 'reveal_text',
+      arguments: {
+        text: `Please email ${sanitizedText.match(/\[EMAIL_\d+\]/)?.[0] || '[EMAIL_1]'} back.`
+      }
+    });
+    const revealedText = revealResponse.result?.content?.[0]?.text;
+    console.log("<-- Revealed Output:", revealedText);
+    if (!revealedText.includes("john.doe@example.com")) {
+      throw new Error("Reveal failed to restore the original email.");
+    }
+    console.log("✅ Reveal text success.");
+
+    // 5. Test sanitize_file (text file)
+    const tempFile = path.resolve(__dirname, 'temp-test-file.txt');
+    fs.writeFileSync(tempFile, "API Key: secret-key-value-12345\nUser: jane.smith@company.com", "utf8");
+    console.log(`--> Calling 'sanitize_file' on text file: ${tempFile}`);
+    
+    const fileResponse = await sendRequest('tools/call', {
+      name: 'sanitize_file',
+      arguments: {
+        filePath: tempFile,
+        profile: 'Dev'
+      }
+    });
+    fs.unlinkSync(tempFile);
+    
+    const fileSanitized = fileResponse.result?.content?.[0]?.text;
+    console.log("<-- Sanitized File Output:", fileSanitized);
+    if (fileSanitized.includes("secret-key-value-12345") || fileSanitized.includes("jane.smith@company.com")) {
+      throw new Error("File sanitization failed to redact credentials.");
+    }
+    console.log("✅ Sanitize file success.");
+
+    // 6. Test sanitize_file (binary file rejection)
+    const tempBinFile = path.resolve(__dirname, 'temp-binary-file.bin');
+    const binBuffer = Buffer.alloc(100);
+    binBuffer.write("GIF89a", 0);
+    binBuffer[10] = 0; // Null byte
+    fs.writeFileSync(tempBinFile, binBuffer);
+    console.log(`--> Calling 'sanitize_file' on binary file: ${tempBinFile}`);
+
+    const binFileResponse = await sendRequest('tools/call', {
+      name: 'sanitize_file',
+      arguments: {
+        filePath: tempBinFile,
+        profile: 'General'
+      }
+    });
+    fs.unlinkSync(tempBinFile);
+
+    const isBinError = binFileResponse.result?.isError || binFileResponse.error;
+    const binErrorText = binFileResponse.result?.content?.[0]?.text || binFileResponse.error?.message;
+    console.log("<-- Binary File Response Error status:", isBinError, "| message:", binErrorText);
+
+    if (!isBinError || !binErrorText?.includes("Binary file format detected")) {
+      throw new Error("Binary file check failed to reject binary file with clean message.");
+    }
+    console.log("✅ Sanitize file binary rejection success.");
+
+    // 7. Test sanitize_file (DOCX file parsing support)
+    const docxSource = path.resolve(__dirname, '../PrivacyScrubber/outreach-campaigns/guest-posts/Techbullion_GuestPost.docx');
+    const docxDest = path.resolve(__dirname, 'fixture-test.docx');
+    
+    if (fs.existsSync(docxSource)) {
+      fs.copyFileSync(docxSource, docxDest);
+      console.log(`--> Calling 'sanitize_file' on DOCX document: ${docxDest}`);
+
+      const docxResponse = await sendRequest('tools/call', {
+        name: 'sanitize_file',
+        arguments: {
+          filePath: docxDest,
+          profile: 'General'
+        }
+      });
+      fs.unlinkSync(docxDest);
+
+      const docxText = docxResponse.result?.content?.[0]?.text;
+      console.log("<-- DOCX Sanitized text preview (first 150 chars):", docxText?.substring(0, 150) + "...");
+      
+      if (!docxText || docxResponse.result?.isError || docxText.includes("Error") || docxText.length < 50) {
+        throw new Error("DOCX document parsing failed or returned invalid text content.");
+      }
+      console.log("✅ Sanitize DOCX document success.");
+    } else {
+      console.log("⚠️ Skipped DOCX test case: fixture file not found in main project path.");
+    }
+
+    console.log("\n🎉 All deep integration tests passed successfully!");
+    mcpProcess.kill();
+    process.exit(0);
+
+  } catch (error) {
+    console.error("❌ Test failed:", error.message);
+    mcpProcess.kill();
+    process.exit(1);
+  }
+}
+
+runMcpSession();

package/test-mcp.js

@@ -10,7 +10,7 @@ const require = createRequire(import.meta.url);
 const __filename = fileURLToPath(import.meta.url);
 const __dirname = path.dirname(__filename);
 
-const scrubberCorePath = path.resolve(__dirname, '../chrome-extension/scrubber-core.js');
+const scrubberCorePath = path.resolve(__dirname, './scrubber-core.cjs');
 const mcpServerIndex = path.resolve(__dirname, 'index.js');
 
 console.log("Loading modules for verification test...");