mrkeyoor.comAsk

guillaumemeyer/watermarks-removertested 28 Sept 2026commit 258cf20

watermarks-remover review

Our watermarks-remover run installed 55 packages in 6 seconds and passed its 96-second test run, making the core a credible trial for authorized file hygiene.

  • Build 1 s
  • Tests pass
6 sto install
  • 55 packages
AI Toolsevaluationupdated 29 Sept 2026
+438stars / 7d

watermarks-remover inspects and removes provenance signals from text and files you own. It covers deterministic Unicode cleanup, file metadata, and optional attempts against statistical text, image, audio, and video watermarks, while stating that no universal clean result is possible.

Verdict

Our watermarks-remover run installed 55 packages in 6 seconds and passed its 96-second test run, making the core a credible trial for authorized file hygiene. Use it when you can keep originals, inspect detailed detector status, and validate each result independently. Skip it when provenance must survive or when a clean response must amount to proof that no watermark remains.

We ran it

Lab card: what happened when we ran watermarks-remover
Lab card from our sandbox run
Screenshot of watermarks-remover (github.com/guillaumemeyer/watermarks-remover)
github.com/guillaumemeyer/watermarks-remover
Install✓ · 6s55 packages · 74 MB
Build✓ · 1s
Tests✓ · 96sran, no count parsed
Known vulns0(pip-audit)
Repo292 files~48,621 lines of source · 2.1 MB · 3 CI workflows · tests dir

Answers from our run

Does watermarks-remover build from source?

Dependencies installed in 6 seconds (55 packages), and the build succeeded in 1 seconds. We cloned commit 258cf20 into a clean Debian container with 3 CPUs and no project-specific setup.

Do watermarks-remover's tests pass?

Yes. The project's own test command exited successfully in 96 seconds, although its output did not expose an individual test count.

Does watermarks-remover have known vulnerabilities in its dependencies?

pip-audit found none in the dependency tree at the time of our run.

Who should not use watermarks-remover?

Archives, newsrooms, or compliance teams that must preserve provenance: the tool can remove C2PA and generator metadata they may need as evidence.

What are the alternatives to watermarks-remover?

MAT2, C2PA Rust, MarkLLM. Our watermarks-remover run installed 55 packages in 6 seconds and passed its 96-second test run, making the core a credible trial for authorized file hygiene.

Setup4/56-second install and passing build and tests
Docs5/5Setup, format limits, detector scope, and quality costs are explicit
Community4/523,272 stars, an October 2 push, and active issue work
Maturity4/5v0.7.0 is broad, but several advertised paths still have open bugs

Discussed on

  1. hnShow HN: Watermarks Remover: Clean LLM watermarks from text and files5 points
  2. hnWatermarks-remover: Strip multi-vendor AI provenance marks3 points

Who it’s for

Publishers removing private metadata or invisible characters from content they own.
Claude Code users who want file checks after an agent writes or edits a file.
Privacy teams that can retain untouched originals and validate every cleaned output.
Researchers testing a known watermark scheme with the matching configuration and detector.

Who it’s NOT for

Archives, newsrooms, or compliance teams that must preserve provenance: the tool can remove C2PA and generator metadata they may need as evidence.
Anyone seeking proof that a vendor detector will fail: the README calls statistical text removal best-effort and says its research harnesses are not vendor oracles.
Hosted agent users unwilling to expose or self-host the service: issue 356 reports that the thin skill cannot reach its default local service inside a hosted sandbox.
Audio pipelines centered on Ogg, Opus, or AAC: issue 327 says those formats currently reach the unknown-file refusal before the audio cleaner.
In-place publishing jobs that cannot tolerate provenance-bearing backup files: issue 369 says --in-place still leaves a sibling .bak file.

Setup reality

Our sandbox installed commit 258cf20 in 6 seconds, adding 55 packages and using 74 MB. The build passed in 1 second, the tests passed in 96 seconds, and pip-audit found 0 known vulnerabilities.

The basic service needs Python and can run on loopback without a secret. Statistical text cleaning needs a configured rewrite backend; remote use may require an API key. PDF and provenance work improves when qpdf, ExifTool, c2patool, or Ghostscript are installed.

The 2.1 MB checkout had 292 files and about 48,621 source lines. We found 3 CI workflows, a compose file, and a tests directory, but no Dockerfile at the repository root. Optional research backends bring separate models, checkouts, licenses, and hardware requirements.

The tool separates verifiable cleanup from removal attempts

watermarks-remover handles several jobs that are easy to confuse. Its deterministic layer removes invisible Unicode, exotic spaces, bidirectional controls, and similar text carriers. File handlers strip C2PA, EXIF, XMP, document properties, and related metadata from supported containers. A separate rewrite layer attacks statistical text marks by changing wording, while optional research backends try known image, audio, or video schemes.

That division is the project's best judgment call. A byte removed from a file is observable. A detector result only speaks for the configured detector, model, and key. The README says MarkLLM checks are same-configuration experiments and cannot certify that a vendor's detector will fail. It also says rewriting can flatten tone and precision. If your interface reduces all of that to one green badge, you have thrown away the most useful honesty in the project.

What happened when we ran it

Our sandbox installed commit 258cf20 in 6 seconds, pulling 55 packages and occupying 74 MB. The build finished successfully in 1 second. Tests also passed, taking 96 seconds, and pip-audit reported 0 known vulnerabilities. We ran the repository in an unprivileged Debian container with Python 3.12, 3 CPUs, 8 GB of RAM, and no secrets.

The checkout contained 292 files, about 48,621 lines of source, and used 2.1 MB before installation. We found 3 CI workflow files, a compose file, and a tests directory. There was no Dockerfile at the repository root. Those results cover the commit and commands we ran. They do not measure removal accuracy, visual or audio damage, rewrite quality, or success against a private vendor detector.

The small core grows when a format needs real cleaning

The basic local path is approachable. Python starts a standard-library HTTP service on 127.0.0.1:8765, and the command-line tools can inspect or clean individual files. The published core container adds qpdf, ExifTool, and c2patool. PDF handling explains why these extras matter: ExifTool edits can leave old bytes recoverable, so qpdf performs a structural rewrite, while Ghostscript is used for metadata inside embedded images.

Layer B text work has a different bill. The /clean service needs a rewrite backend for text, such as local Ollama or an explicitly allowed remote endpoint. Optional MarkLLM, MarkDiffusion, CtrlRegen, and SynthID components add external repositories, model weights, licensing conditions, or GPU demands. Our 74 MB environment is a fair description of the tested Python setup. It is not the footprint of every research backend running together.

Hosted skills still depend on a reachable service

The Claude Code plugin can run a post-write hook in check or clean mode, and a pre-commit path can inspect files before they enter Git. The skill itself is deliberately thin. It calls the HTTP service instead of carrying the cleaning implementation. That is sensible on a developer machine where the service is on loopback and the hook can reach it.

Hosted agents expose the boundary. Open issue 356 reports that a claude.ai sandbox saw connection refused at 127.0.0.1:8765, could not start Docker there, and did not receive the referenced service files in the plugin package. The README now tells remote Cowork or cloud sessions to use a reachable service URL and bearer key. If you cannot operate that endpoint, use the self-contained text-only skill or choose a local workflow.

Current audio and backup bugs block two unattended workflows

Version 0.7.0 added destructive audio watermark removal and per-frame video purification. Open issue 327 identifies a routing gap: Ogg, Opus, and AAC appear in the audio cleaner's extension list, yet the main classifier sends them to the unknown-file refusal. The issue includes an OggS reproduction and says /clean returns 400 for the same inputs. A browser-recorder or Discord pipeline using those formats should wait for a merged fix and release.

Open issue 369 catches a quieter operational problem. clean_file.py --in-place still creates <file>.bak, with no documented --no-backup option in the reported revision. In a static site or deploy directory, that backup may be published and still contain the provenance just removed from the original filename. Keep backups outside scanned and deployed trees, or write cleaned output to a separate staging directory.

October activity is healthy, but a clean result has a narrow meaning

GitHub showed 23,272 stars and 15 combined issues and pull requests on October 3, 2026. The API list separated those into 9 issues and 6 pull requests. The repository was pushed on October 2, while v0.7.0 shipped on September 3. Recent work and the 3 CI workflows show active maintenance, though users still need to check whether a fix has reached the image or tag they deploy.

Use the detailed report as evidence of what a named cleaner attempted and what a named detector observed. Keep the source file. Reopen the output with an independent parser, compare page count or media duration, and review rewritten prose for changed meaning. The passing 96-second suite earns this project a trial. Its own documentation supplies the stopping rule: no tool can promise that an unknown or private detector sees nothing.

Alternatives

ProjectWhat it isPick it when
MAT2A Python metadata cleaner for common documents and media files.pick this instead when ordinary metadata removal is the whole job and AI-specific rewrite or detection paths would add needless scope.
C2PA RustA Rust SDK and CLI for reading, creating, and validating C2PA manifests.pick this instead when you need to preserve and verify provenance rather than remove it.
MarkLLMA research toolkit for generating and evaluating language-model watermarks.pick this instead when controlled watermark experiments matter more than cleaning everyday files.

Sources

  1. watermarks-remover README
  2. watermarks-remover repository facts
  3. watermarks-remover v0.7.0 release
  4. Hosted agent service issue
  5. Ogg, Opus, and AAC routing issue
  6. In-place backup issue

More ai tools reviews