remove-ai-watermarks
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| identifyA | Report which AI provenance signals an image or video carries: metadata (C2PA, IPTC, EXIF, TC260), visible AI-generator labels, and supported local invisible-watermark decoders. Read-only. Video is detected by file suffix. |
| removeA | Write a cleaned copy of an image or video the user generated or edited. mode=metadata strips AI metadata without touching pixels or re-encoding video; mode=visible also removes a known visible AI label; for an image, mode=all adds the invisible-watermark scrub where the installed extras and hardware allow it. The source is never overwritten; the default output is _clean beside it. |
| batchA | Clean every supported image (media=image) or video (media=video) in a directory the user owns, writing results to a separate directory (default _clean). Modes as for the remove tool. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
identify is a read-only detection tool, while remove operates on a single file and batch operates on a directory, so the boundaries are generally clear. The only mild overlap is that batch is essentially remove applied across a folder, but the descriptions make the scope distinction explicit.
All three tools use single lowercase verbs (identify, remove, batch) in a consistent, predictable style. It deviates from a strict verb_noun pattern, but there is no mixing of conventions or casing.
Three tools is on the lean side but well-matched to a narrow domain of detecting and stripping AI provenance signals. Each tool earns its place with no redundancy or filler.
The surface covers the core lifecycle: detect signals, clean a single file, and clean a whole directory, with modes controlling what is stripped. The only minor gap is a dedicated verification/preview step, though identify can be re-run after removal to confirm results.