Local privacy inventory

Metadata & Privacy Inspector

Find supported EXIF, PNG text and PDF document properties before a file is published, emailed or handed to a client.

✓ Deterministic local baseline✓ Evidence per finding✓ JSON + CSV report✓ Versioned method
Interactive tool

Analyse your own file

Local by default

Choose a PDF, JPEG and PNG file

Drop it here or browse this device

No file selected

Your original file is analysed in this browser and is not uploaded. This tool has no AI or server-processing step.

Processing manifest

What happens to the file

Original stays on this device
Default processing
Browser-only JavaScript, PDF.js, pdf-lib and Tesseract where the task needs them
Accepted input
PDF, JPEG and PNG
Report contract
File Intelligence JSON schema 1.0 plus a flattened CSV view
Optional AI
Not used by this tool
Example analysis report

Sample PDF exposes an author name, producer and creation date. The report masks values in the overview and lists exact evidence only on the user’s device.

Download metadata-sample.pdfRun it with “Load published sample” above to reproduce the report.

How the result is calculated

The browser scans JPEG APP1 markers, PNG ancillary chunks and PDF Info/catalog dictionaries. Findings show the field, a bounded preview and the container that supplied the evidence.

Published formula and threshold

Privacy risk uses evidenced field categories: location/identity markers are high, author/device/time/software context is medium, and container-only markers are low. Confidence is parser coverage, never a guarantee of absence.

Supported and unsupported cases

Input or situationWhat version 1.0.0 checksWhat it does not claim
JPEGAPP1/EXIF presence, common camera/software/date/GPS markersComplete maker-note decoding or authenticity
PNGtEXt, iTXt, zTXt, eXIf and tIME chunk inventoryDecompression of every vendor-specific compressed field
PDFTitle, author, creator, producer, dates, language and XMP markerAll compressed object streams or attachment metadata
Other filesBasic identity can be shown by the health toolMetadata inspection on this page
Advertisements

What the confidence score means

Confidence describes how much of the file the published parser could inspect and how specific the evidence was. It is deliberately separate from the result score. A 95/100 confidence value does not mean “95% likely to be true,” and a high health/readiness score does not certify safety, compliance or factual correctness.

Comparison with other approaches

ApproachUseful strengthImportant limitation
Local metadata inspectorBrowser-only, evidence source, exportable inventoryFocused parser coverage
EXIF command-line utilitiesMuch deeper tag coverageInstallation and command-line knowledge required
Removing file properties in an OSConvenient for supported typesMay not show or remove every field
Screenshot/export copyOften strips document structureCan reduce quality and still retain new metadata
Anonymous synthetic release dataset

4/4 fixtures passed (100%)

Dataset: FH-INT-SYNTH-2026-07. Scope: PDF Info fields, PNG text chunks, JPEG EXIF marker and clean-file control. Fixtures contain generated data only; no visitor file or personal record is collected.

Open method and raw CSV

When this result can be wrong

Metadata can exist in compressed, encrypted, malformed or proprietary structures the lightweight parser cannot decode. “No supported metadata found” means no match in the published coverage—not proof of zero metadata.

Keep the original file, test the output in the receiving application and escalate high-stakes legal, financial, medical, accessibility or security decisions to a qualified reviewer.

Advertisements

Version and accuracy change log

ReleaseDateCalculation/coverage changeRe-test requirement
1.0.02026-07-29Initial public engine, report schema, thresholds, synthetic fixtures and sample files.All published fixtures and one browser sample per tool.

Downloadable evidence, not a black-box verdict

After analysis, use Download JSON report for structured evidence or Download CSV report for review in a spreadsheet. Both include engine version, timestamp, score, confidence, metrics, findings, recommendations and limitations. Sensitive-data reports mask candidate values and no report includes the complete extracted document text.

Related tools and independent checks

Method or accuracy correction? Read the testing methodology and contact info@brothers.net.tr. Material scoring changes are published in the change log.