convert_file
Convert File
Convert a file at a public URL to another format and provide all resulting downloadable files.
Ask your agent in plain language. Each of these maps to a real tool on the server.
convert_file
Convert a file at a public URL to another format and provide all resulting downloadable files.
generate_thumbnail
Generate PNG, JPG, or WEBP thumbnails from a video, document, or image and provide the resulting downloadable files.
add_watermark
Add exactly one text or image watermark to a PDF, image, or video. Watermarking keeps the input format; use convert_file for a separate format conversion.
merge_files
Merge files from public URLs, in order, into one downloadable PDF. Non-PDF inputs are converted to PDF by CloudConvert.
process_pdf
OCR, encrypt, decrypt, split, extract, rotate, or convert a PDF to PDF/A using the corresponding PDF operation. Provide all resulting downloadable files.
list_jobs
List recent jobs with tasks, status, and an optional exact tag filter. Follow nextPage until absent; jobs normally expire 24 hours after ending, so this is not a complete billing history.
get_job
Read an existing job with all task states and provide any finished export/url files. Pending and failed states remain explicit. Ended jobs and their data normally expire after 24 hours.
delete_job
Delete an existing job including its tasks and stored file data. Confirms HTTP 204 and then an independent not-found read. Deletion is not a refund of credits already consumed.
extract_metadata
Create a metadata extraction job for a file at a public URL. Returns the actual ExifTool metadata when complete; this operation creates jobs but does not consume conversion credits.
get_current_user
Read the account connected to the selected API environment and its remaining conversion credits. Requires user.read.
list_formats
Discover supported conversion paths from the current operations catalog. Filter by input format, output format, engine, or engine version before creating a conversion.
download_job_files
Download files from finished export/url tasks in an existing job. Reads the existing job and creates no new conversion or export. Files are limited to 32 MiB each and 20 files per call; provider files normally expire after 24 hours.
create_archive
Create a downloadable ZIP, RAR, 7Z, TAR, TAR.GZ, or TAR.BZ2 archive from files at public URLs.
create_job
Create a custom job from documented named tasks. Task inputs reference other names in this job. Processing can consume credits; use a unique tag for recovery and download results before their 24-hour lifetime ends.
manage_task
Get or list task states, explicitly cancel a waiting/processing task, create a new retry with a new ID, or delete a task and its data. Retry can consume credits and is never automatic. Cancellation and deletion do not refund consumed credits.
capture_website
Capture a website as PDF, PNG, or JPG and provide the resulting downloadable files. Optional page settings depend on the selected output format.
optimize_file
Optimize a PDF, PNG, or JPG from a public URL and provide the resulting downloadable files. Engine-specific options depend on the input format.
Every client uses the same Magic MCP URL, and so does any other MCP client. Add it once and sign in with SSO.
{"mcpServers": {"metorial": {"url": "<your-magic-mcp-url>"}}}
claude mcp add --transport http metorial <your-magic-mcp-url>
codex mcp add metorial --url <your-magic-mcp-url>codex mcp login metorial
One governed layer for every app your agents reach, not a separate connector per tool.
Convert, optimize, watermark, and process URL-based files with CloudConvert API v2. Capture websites, create thumbnails and archives, merge PDFs, extract metadata, and perform OCR, encryption, decryption, page splitting, extraction, rotation, and PDF/A conversion. Inspect account, jobs, and tasks; explicitly retry or cancel tasks, delete job data, and download existing results.
Common questions about connecting Cloudconvert to AI agents with Metorial.