How to use PDF Inspector
- Choose a supported PDF within the file and page limits.
- Run the inspection and read the document metadata.
- Compare each page's visible dimensions, rotation and declared resources.
- Use the report as a structural reference and review the PDF in a viewer when appearance matters.
Example: PDF Inspector
Inspect a cropped and rotated page.
Options
- File-size investigation
- Image resources can explain a large scan or photo-heavy PDF, but resource counts do not measure exact savings. Use Compress PDF to test supported image recompression, or PDF Attachment Manager to review attached files separately.
Supported inputs and limits
- One unencrypted PDF up to 100 MiB and 1,000 pages.
- At most 500 font resources and 500 XObject resources per page are inspected.
- Metadata values shown in the report are shortened to 500 characters.
- Page resources may be shared, unused or nested, so counts do not equal visible placements.
- The report observes supported dictionaries and page boxes without rendering or executing scripts.
More details
It does not validate every PDF structure, establish authenticity or certify security.
Where your input is processed
This tool does not upload your inputs. See the limits for anything saved on your device.
Visible pages and document resources
The CropBox and rotation describe the area a viewer displays. Resource dictionaries describe objects that page content can refer to. Their presence alone does not show where or how often an object is drawn, so use both the report and a rendered view for a complete visual check.
Questions about PDF Inspector
Why is an image count different from what I see?
The report counts supported page resource entries. An entry can be unused or reused several times, and content inside nested structures may need another inspector.
Does a font resource prove the font is embedded?
A declared font entry is not the same as an embedding or rendering guarantee. This report describes its supported resource scope rather than validating every font stream.
Does a JavaScript observation mean the file is malicious?
No. A structural field or name-tree observation is not a malware verdict. The page does not execute the document's scripts or certify its safety.
Can it inspect password-protected contents?
Encrypted documents are outside the supported content inspection. A refusal should be distinguished from a malformed file; it does not recover a password.