vCard Contacts Parser Extended MCP. Extract reliable contact data from messy vCF files.

When you pull contacts off your phone, what do you get? You get a vCard (.vcf) file. These things are garbage—they’re bloated messes stuffed with binary noise, base64-encoded images, and decades of legacy data structures like BEGIN:VCARD. If you just throw that raw export at an LLM or your agent, it wastes tokens trying to parse irrelevant junk or, worse, it starts hallucinating fields because the data is too messy.

You need something dedicated.

The vCard Contacts Parser Extended handles this garbage and gives you usable intel. This tool's sole job is to take a file path pointing to a vCard export and convert that raw chaos into clean, structured JSON. Your AI client doesn't see the mess; it only sees what matters: actionable data points.

The core functionality lives in parse_vcard_contacts. When you run this tool, it reads the vCard file path and immediately strips away all the binary noise and structural overhead. It processes those messy contacts and outputs a perfectly structured JSON array ready for querying. You're getting raw contact data organized into predictable fields, which is exactly what your agent needs to work with.

Structuring Raw Contacts: The tool takes that volatile vCard file path and turns it into an easily queryable JSON format. This isn't just dumping the text; it’s parsing a massive array of contacts, ensuring every entry follows the same clean data schema. You get back structured contact details you can immediately work with.

Extracting Key Data Points: The parser doesn't just dump everything it finds. It isolates and organizes the specific fields that actually matter for business use. Think about it: phone numbers, email addresses, full names, company roles—it pulls them out individually. This granularity means your agent can reliably check if a contact has an office number versus a mobile number, or separate their first name from their last name without fail.

Handling Large Datasets: This is where the tool really shines. If you have thousands of contacts—like exporting an entire corporate directory that's loaded with data—the parser handles it. It processes these huge vCard files containing potentially thousands of entries without hitting context window limits or slowing down. You can pass in a gigantic file, and it reliably delivers structured JSON for every single contact.

The output is designed for machine readability. Instead of dealing with cryptic headers and semi-colon separated garbage, your agent gets pure data. It's perfect for tasks like generating mailing lists, cross-referencing client records, or populating databases in batches.

When you use parse_vcard_contacts, you’re essentially giving your AI client a clean API endpoint for contact intelligence. You provide the file path; it returns an array of objects, where each object is one perfectly parsed contact record. These records reliably contain separated fields for: names (first and last), phone numbers, email addresses, and detailed organization information.

The JSON format makes querying straightforward—your agent doesn't have to guess which field holds the primary phone number; it just looks at contact[].phone_number.

The efficiency of this process saves time. You don't waste compute cycles trying to decipher legacy data formats. Instead, your agent uses those resources doing actual work, like drafting follow-up emails or checking for duplicate entries across multiple client lists.