Inspect and remove supported C2PA Content Credentials.
Content Credentials can carry structured origin and editing records inside a digital asset. No Trace Share can inspect supported embedded C2PA structures, show the records it understands, distinguish explicit AI-related signals from generic provenance, and remove supported credentials during cleaning.
Opens the private browser-based cleaner. Supported embedded records are inspected locally before cleaning.
What are Content Credentials?
Content Credentials are the user-facing term for C2PA provenance data. Technically, C2PA manifests contain assertions, a claim and a claim signature, and can be stored in a structured JUMBF manifest store embedded in an asset or referenced externally.
No Trace Share focuses on the supported embedded records inside the file you select. It is not a full C2PA trust validator.
Technical reference: C2PA Content Credentials specification.
What No Trace Share can report
Credential presence
The app can identify supported embedded C2PA / Content Credentials structures and report that they are present.
Manifest details
Where parsing succeeds, the app can expose supported manifest counts, recorded actions, claim-generator names, titles, authors, identifiers, embedded preview indicators and assertion labels.
Explicit AI-related signals
If a supported C2PA action or source type explicitly points to trained-algorithmic or synthetic media, No Trace Share marks it as an AI Trace.
Generic provenance stays generic
A credential without an explicit AI signal is shown as Content Credentials rather than being mislabeled as proof of AI generation.
How removal works in supported image formats
For compatible JPEG, PNG and WebP files, Content Credentials are stored in format-specific containers. No Trace Share’s cleaning engine removes supported C2PA blocks as part of the format’s metadata-cleaning operation and then reinspects the generated copy.
| Format | Typical supported location | No Trace Share behavior |
|---|---|---|
| JPEG | APP11 / JUMBF data | Supported C2PA metadata segments are removed during the standard clean. |
| PNG | C2PA / JUMBF-compatible chunk structures | Supported provenance chunks are removed under the PNG cleaning policy. |
| WebP | C2PA chunk | The dedicated C2PA chunk is excluded from the cleaned output. |
| HEIF / HEIC | Container structures vary | Support is conservative; unsupported structures can remain or cause refusal rather than being rewritten unsafely. |
Removing credentials is not the same as disproving history
Deleting an embedded provenance record removes that supported record from the generated copy. It does not establish that the media was never AI-generated, never edited, or anonymous. External provenance, visible content, pixel-level marks and unsupported structures can exist independently of the embedded credential that was removed.
Content Credentials FAQ
Are Content Credentials the same thing as AI metadata?
No. Content Credentials are a C2PA provenance mechanism that can describe ordinary capture and editing history as well as AI-related workflows. Only explicit AI-related records are labeled as AI Traces.
Does No Trace Share verify C2PA signatures?
No. The current web app does not certify signature validity, certificate trust or authenticity.
What if I keep Content Credentials but change other metadata?
Changing the asset while retaining existing credentials can invalidate their signatures or claims. No Trace Share warns about this when supported C2PA fields are retained.
Check your own file, not just the explanation.
Open the No Trace Share workspace to inspect supported metadata, review the exact findings, and create a verified cleaned copy.
Clean a file nowYour selected files stay in your browser and are not uploaded to No Trace Share.