# vCard Contacts Parser MCP for AI Agents AI Agent Connect > vCard Contacts Parser converts massive .vcf contact exports from iPhone and Android devices into structured JSON. It allows your AI client to interact with your local address book as a clean, searchable database. Instead of forcing your agent to wade through bloated files full of base64 images and legacy formatting, this Connector provides only the relevant names, phone numbers, and emails. It's the fastest way to give your agent memory of your personal and professional network while keeping everything on your machine. ## Overview - **Category:** productivity - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_nMwIDpe8t4IiDDLs4hL1zHyXkLRrKAbAmIxeZbyH/ai-agent-connect - **Tags:** vcard, contact-management, data-parsing, json-conversion, address-book, file-processing ## Description When you export your phone's address book, you usually end up with a bloated .vcf file. These files are messy because they contain legacy formatting and heavy base64-encoded profile pictures. If you try to feed that raw file into an AI, it will likely choke on the size or start making up phone numbers because it can't parse the noise. This Connector fixes that by acting as a local intelligence engine. It strips away the junk and turns the raw data into a clean JSON array that your AI client can actually understand. You get a reliable way to ask your agent for specific details, like finding a colleague's office number or a friend's email, without ever uploading your private contact list to a third-party cloud. It's a core part of the Vinkius ecosystem designed to make your local data actually useful for your daily tasks. ## Tools ### parse_vcard_contacts Provide a file path to convert a vCard export into a structured JSON array. This lets your AI client query names, phones, and emails without getting bogged down by image data. ## Prompt Examples **Prompt:** ``` Search my contacts.vcf and give me a list of everyone who works at 'Vinkius'. ``` **Response:** ``` I found 3 contacts working at Vinkius: * **John Silva** (DevOps) * **Sarah Connor** (Design) * **Mike Ross** (Legal) Would you like me to draft an email to any of them? ``` **Prompt:** ``` Extract all the email addresses from this vCard export and format them as a CSV. ``` **Response:** ``` I've extracted the email addresses from your export. Here they are in CSV format: Name,Email Alice Smith,alice@example.com Bob Jones,bob@example.org Let me know if you need any other fields included. ``` **Prompt:** ``` Look through my contacts and find the phone number for 'Plumber'. ``` **Response:** ``` I found the entry for **Plumber**. * **Phone Number:** +1 (555) 019-8372 Is there anything else you need to find in your contacts? ``` ## Capabilities ### Strip out base64 image data Removes heavy profile pictures from your contact files to save space. ### Extract phone numbers Pull out country codes and numbers into a clean format for your agent. ### Parse organization details Identifies company names and job titles from your vCard exports. ### Convert vCards to JSON Turns raw .vcf files into a structured format that AI clients can read. ### Process massive contact lists Handles files with over 5,000 contacts without hitting limits. ## Use Cases ### Finding a specific person An executive assistant asks the agent to find the phone number for the 'Plumber' in their contacts. ### Bulk email extraction A marketer wants to pull all email addresses from a 2,000-person vCard export to create a mailing list. ### Organization filtering A recruiter asks the agent to list everyone in their phonebook who works for 'Vinkius'. ### Contact verification A user wants to check if a specific contact already exists in their phonebook before adding a new one. ## Benefits - Keep your personal data private because this Connector runs entirely on your local machine. - Eliminate AI hallucinations by using parse_vcard_contacts to extract precise contact details. - Process huge contact lists since it handles files with 5,000+ entries without breaking. - Save your AI's context window by stripping out heavy base64-encoded profile pictures. - Get clean, queryable JSON that lets your agent quickly find specific people or organizations. ## How It Works The bottom line is your AI gets a clean, searchable version of your phonebook without the technical overhead. 1. Provide the Connector with the absolute file path to your .vcf export. 2. The Connector parses the file to strip out binary noise and extract contact details. 3. Your AI client receives a clean JSON array it can query for specific names or numbers. ## Frequently Asked Questions **Can the vCard Contacts Parser handle my whole phone book?** Yes, it can process files with over 5,000 contacts at once. This means you don't have to split your exports into smaller chunks to get your AI to see them. **Does vCard Contacts Parser upload my contacts to the cloud?** No, it runs 100% locally on your machine. Your personal data stays on your hardware, which is a huge win for privacy. **What kind of files does vCard Contacts Parser work with?** It specifically parses .vcf files, which are the standard exports for iPhone and Android contact lists. **Can I use vCard Contacts Parser to find a specific person's number?** Yes, once the file is parsed, you can just ask your agent to find a specific name or company. It will search the structured data and give you the result. **Will vCard Contacts Parser make my AI hallucinate less?** Yes, because it strips out the messy binary noise and base64 images that usually confuse AI models. It only gives the AI the clean text it needs. **Does vCard Contacts Parser work with Claude or Cursor?** It works with any MCP-compatible client, including Claude, Cursor, and Windsurf. You just connect it through the Vinkius catalog. **Is my address book uploaded to the cloud?** Never. The vCard parsing is executed completely local on your device. Only the extracted text representation is provided to the AI context. **Does it support multiple contacts in a single file?** Yes! It perfectly parses multi-vCard files exported from iOS, Google Contacts, or Android devices, handling thousands of entries seamlessly. **What happens to the contact profile pictures?** Profile pictures (PHOTO;ENCODING=b) are intentionally ignored and stripped during parsing to preserve AI context tokens and prevent crashes.