APICK OCR
Korean OCR and ID-document extraction exposed as an MCP tool for AI agents
- Pricing
- Paid
- Platforms
- API, CLI
- Runs
- Remotely
- Version
- 1.3.0
- Listing updated
- today
Similar MCP servers
About APICK OCR
APICK OCR is a paid MCP server for files and documents. It runs as a hosted remote server, so there is nothing to install locally.
The OCR server is one of several APICK endpoints and handles text recognition on images, with Korean identity documents as a focus. It is reached over streamable HTTP at a fixed URL and authenticated with an APICK license key sent as a bearer header, so an MCP client can call it without installing anything locally.
The README lists seven tools on this server: general image text extraction, identity-card field extraction, and reading of business registration certificates, registry documents, and seal certificates. Files are passed as public https URLs rather than uploaded inline, and results come back as structured JSON.
It suits developers and back-office teams who need to pull fields from Korean ID or business documents into an AI workflow. Billing is per successful call, and tool discovery does not require a key.
Install
Run in a terminal
claude mcp add --transport http ocr https://apick.app/mcp/ocrAdd to Settings → Connectors → Add custom connector
https://apick.app/mcp/ocrAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"ocr": {
"url": "https://apick.app/mcp/ocr"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"ocr": {
"type": "http",
"url": "https://apick.app/mcp/ocr"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"ocr": {
"serverUrl": "https://apick.app/mcp/ocr"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"ocr": {
"type": "http",
"url": "https://apick.app/mcp/ocr"
}
}
}Using several servers? Build one config file with the MCP config generator.
Tools this server provides (7)
Taken from the project documentation; the running server may differ.
- ocr
- Image text extraction.
- business_doc_ocr
- Business document reading.
- ocr_identi1
- Korean ID field extraction (resident registration card).
- ocr_identi2
- Korean ID field extraction (driver licence).
- ocr_identi3
- Korean ID field extraction (passport).
- ocr_identi4
- Korean ID field extraction (resident registration abstract).
- ocr_identi5
- Korean ID field extraction (alien registration card).
What it does
- Image text extraction
- Reads text from images supplied as public https URLs.
- Korean ID field extraction
- Extracts fields from resident registration, driver licence, passport and related documents.
- Business document reading
- Reads business registration certificates and other business paperwork.
- Streamable HTTP endpoint
- Connects to a hosted server without a local install or build step.
- Key-based access
- Uses a bearer license key; tool listing works without one.
Good for
- Extract fields from a scanned Korean ID for onboarding forms
- Read business registration certificates during vendor checks
- Turn image-based documents into text for an AI assistant to summarise
Pricing
Paid
Prepaid points charged per successful call; 1,000 free points on signup.
Prices change. Check apick.app before you buy.
APICK OCR FAQ
How do I authenticate to the OCR server?
The README says requests use an APICK license key sent in the Authorization header as a Bearer token.
How are files sent to the OCR tools?
The README says file-taking tools accept a public https URL pointing to the file, not an inline upload.
Are failed calls charged?
The README says failed calls are not charged and only successful calls are billed at REST API rates.
Is APICK OCR free?
No. APICK OCR is a paid product.
How do I add APICK OCR to Claude Code?
Run this command in your terminal: claude mcp add --transport http ocr https://apick.app/mcp/ocr
What are the best alternatives to APICK OCR?
The closest alternatives to APICK OCR are APICK, APICK Convert, Image & Scanned PDF to Text (OCR), Image to Text OCR and Inkscry: Image to Text OCR.
Where this comes from
- Description: written from the project's README, the project's website and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.