Bigapi MCP server
{
"mcpServers": {
"bigapi": {
"command": "npx",
"args": [
"-y",
"@bigapi/mcp@latest"
],
"env": {
"BIGAPI_KEY": "<bigapi_key>",
"BIGAPI_URL": "https://..."
}
}
}
}{
"mcpServers": {
"bigapi": {
"command": "npx",
"args": [
"-y",
"@bigapi/mcp@latest"
],
"env": {
"BIGAPI_KEY": "<bigapi_key>",
"BIGAPI_URL": "https://..."
}
}
}
}{
"mcpServers": {
"bigapi": {
"command": "npx",
"args": [
"-y",
"@bigapi/mcp@latest"
],
"env": {
"BIGAPI_KEY": "<bigapi_key>",
"BIGAPI_URL": "https://..."
}
}
}
}{
"servers": {
"bigapi": {
"command": "npx",
"args": [
"-y",
"@bigapi/mcp@latest"
],
"env": {
"BIGAPI_KEY": "${input:bigapi_key}",
"BIGAPI_URL": "${input:bigapi_url}"
}
}
},
"inputs": [
{
"id": "bigapi_key",
"type": "promptString",
"description": "Bigapi key",
"password": true
},
{
"id": "bigapi_url",
"type": "promptString",
"description": "Bigapi url",
"password": true
}
]
}{
"mcpServers": {
"bigapi": {
"command": "npx",
"args": [
"-y",
"@bigapi/mcp@latest"
],
"env": {
"BIGAPI_KEY": "<bigapi_key>",
"BIGAPI_URL": "https://..."
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y @bigapi/mcpversion 0.8.2
What it does
MCP server for bigapi.dev that provides file operations AI agents cannot perform locally: rendering HTML/Markdown to PDF, taking screenshots, performing OCR, converting office documents, manipulating PDFs, and processing images. Designed for Claude Desktop, Cursor, Cline, Windsurf and other MCP-capable tools.
- Convert Markdown reports or HTML templates to PDF files for document generation and delivery.
- Extract text from scanned PDFs or photos using OCR with language support and searchability.
- Manipulate PDF files (merge, split, rotate, compress, convert to PDF/A) and convert office documents (DOCX, XLSX, PPTX) to PDF.
Tools this server exposes
| Tool | Does |
|---|---|
| render | Convert HTML, Markdown, or URL to PDF or PNG via server-side Chromium. |
| screenshot | Capture screenshots of public URLs with device presets and full-page capture. |
| ocr | Extract text from scanned PDFs or photos, outputting searchable PDF, plain text, or per-page JSON with language support. |
| office_to_pdf | Convert DOCX, XLSX, PPTX, ODT, RTF, CSV, TXT to PDF via server-side LibreOffice. |
| pdf_to_pdfa | Convert PDF to archival PDF/A-2b format with embedded fonts for long-term storage and compliance. |
| pdf_merge | Merge multiple PDF files into a single document. |
| pdf_split | Split a PDF into separate pages or page ranges. |
| pdf_rotate | Rotate pages within a PDF document. |
| pdf_compress | Reduce PDF file size. |
| pdf_to_images | Convert PDF pages to JPEG or PNG images. |
| image_process | Resize, crop, rotate, convert formats (webp, avif), compress, strip EXIF, or add watermarks to images. |
| image_info | Retrieve image format, dimensions, color space, and EXIF/ICC profile presence. |
| get_access | Obtain a free API key instantly with 100 free operations that never expire. |
| get_balance | Check current account balance. |
| get_usage | View usage statistics for operations performed. |
| set_monthly_cap | Set a monthly spending limit for API operations. |
| get_pricing | Retrieve pricing information for operations. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.