# AI use cases for Watermarks Remover (guillaumemeyer/watermarks-remover) Repo: [guillaumemeyer/watermarks-remover](https://github.com/guillaumemeyer/watermarks-remover) | Use cases: 4 | Page: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/ A privacy-first Python service and agent skill that strips AI provenance marks from text and files you own. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Flag AI provenance marks in files Claude Code writes: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/auto-flag-ai-marks-in-agent-written-files/ - Clean AI marks from manuscripts and web copy: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/clean-ai-marks-from-manuscripts-and-web-copy/ - Strip AI marks from files inside a Codex project: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/run-ai-mark-removal-from-codex/ - Use the remove-ai-marks skill in Cowork sessions: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/clean-files-in-cowork-sessions/ ## How to flag AI provenance marks in files Claude Code writes with Watermarks Remover Repo: [guillaumemeyer/watermarks-remover](https://github.com/guillaumemeyer/watermarks-remover) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Get a PostToolUse hook that checks each file your agent writes for AI provenance marks. **Who it is for:** A developer using Claude Code who wants every file the agent writes checked before it reaches git. **The problem:** Agent-written files can carry provenance marks, and a skill alone depends on the model choosing to run it. **The outcome:** Each file the agent writes is scanned by a harness-run hook in check mode, and any findings are reported to the model. ### Steps 1. From the repository checkout, run `python3 install_skill.py --skill remove-ai-marks --target claude-code`. 2. Start the service with `make serve`. 3. Install the plugin in Claude Code with `/plugin marketplace add guillaumemeyer/watermarks-remover` and then `/plugin install watermarks-remover@watermarks-remover`. 4. Confirm Hook mode is set to check in `/plugin manage`, or add the PostToolUse hook from the README to ~/.claude/settings.json with --mode check. 5. Ask the agent to write a test file, then confirm the hook reports findings without changing the file. ### Prompt to paste into your agent ```text In the watermarks-remover repo, use the remove-ai-marks skill in Claude Code to write with . Report whether the PostToolUse hook in check mode flagged any provenance marks, and leave the file unchanged unless I ask you to clean it. ``` ### Caveat The hook covers files the agent writes, not the chat messages themselves, which the README says no hook can rewrite. Risk, reverse engineering: Check mode only reports marks, but clean mode strips them in place, so only use it on content you own. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/auto-flag-ai-marks-in-agent-written-files/ ## How to clean AI marks from manuscripts and web copy with Watermarks Remover Repo: [guillaumemeyer/watermarks-remover](https://github.com/guillaumemeyer/watermarks-remover) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README Run the text-only skill on your own documents with no service required. **Who it is for:** An author or content editor who has drafted manuscripts, docs, or website copy with AI help and wants to remove text-level marks before publishing. **The problem:** Invisible Unicode and other text-level marks can remain in drafts that you own and plan to publish. **The outcome:** Your text is cleaned by the self-contained skill, which runs its own Layer A scripts without the HTTP service. ### Steps 1. From the repository checkout, run `python3 install_skill.py --skill clean-user-facing-text --target claude-code` (or `--target cursor` for Cursor). 2. For Cursor projects that adopt this workflow, run `cp integrations/cursor/clean-user-facing-text.mdc /path/to/project/.cursor/rules` with your own project path. (inferred, check the README) 3. Open the manuscript, documentation, or web copy in your agent session. 4. Ask the agent to clean the text using the clean-user-facing-text skill. 5. Review the cleaned file and confirm the wording and meaning are unchanged. ### Prompt to paste into your agent ```text Using the clean-user-facing-text skill from the watermarks-remover repo, clean the text in . Report how many invisible Unicode or exotic space characters were removed, and confirm the visible wording is unchanged. ``` ### Caveat The README scopes this skill to authorized manuscripts, documentation, and web copy, and it does not cover statistical text watermarks handled by the full skill. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/clean-ai-marks-from-manuscripts-and-web-copy/ ## How to strip AI marks from files inside a Codex project with Watermarks Remover Repo: [guillaumemeyer/watermarks-remover](https://github.com/guillaumemeyer/watermarks-remover) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Install the full skill into one project and call the local service from Codex. **Who it is for:** A developer using Codex who needs to remove provenance marks from images, documents, or text in a specific project. **The problem:** Files and text in a project can carry C2PA, EXIF, XMP, or document properties that a plain text pass does not reach. **The outcome:** Codex calls the local service through the remove-ai-marks skill to inspect and clean files in that project. ### Steps 1. Run `python3 install_skill.py --skill remove-ai-marks --target codex-project --project-dir /path/to/project`. 2. From the repository checkout, start the service with `make serve`. 3. Open the project in Codex and select the remove-ai-marks skill with `/skills` or by typing `$`. 4. Ask the skill to inspect a file, then ask it to clean the file if findings are reported. 5. Open the cleaned file and confirm the marks are gone. ### Prompt to paste into your agent ```text In Codex, use the remove-ai-marks skill from the watermarks-remover repo to inspect through the local service at http://127.0.0.1:8765. Report the provenance marks found, then clean the file and report whether a second inspection shows none. ``` ### Caveat The skill depends on the local service being reachable at WATERMARKS_SERVICE_URL, so the service must be running before you use it. Risk, reverse engineering: The clean step removes provenance marks from files, so run it only on files you own. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/run-ai-mark-removal-from-codex/ ## How to use the remove-ai-marks skill in Cowork sessions with Watermarks Remover Repo: [guillaumemeyer/watermarks-remover](https://github.com/guillaumemeyer/watermarks-remover) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Upload the packaged skill to claude.ai and point it at a reachable service. **Who it is for:** A user of Cowork, claude.ai, or cloud sessions who needs to remove provenance marks from files during a session. **The problem:** Cowork sessions do not read skills from your local machine, so the skill must be uploaded to your account. **The outcome:** The packaged remove-ai-marks skill is available in your account and can call a service the session can reach. ### Steps 1. From the repository checkout, run `python3 install_skill.py --skill remove-ai-marks --target cowork`. 2. In the Claude Desktop app, open Customize → Skills → Add and upload dist/remove-ai-marks.zip. 3. Start the service with `make serve` for local Cowork sessions, or make a remote service URL reachable from the session. 4. Set WATERMARKS_SERVER_API_KEY on the service if it requires an API key, and set the matching key for the client. 5. Start a session, invoke the skill on a test file, and confirm it returns findings from the service. ### Prompt to paste into your agent ```text In a Cowork session with the uploaded remove-ai-marks skill, inspect through the watermarks-remover service at . Report the provenance marks found and confirm the service responded to the health check before inspection. ``` ### Caveat Cloud sessions and routines run remotely, so the service must be reachable from there, and a local make serve will not be enough. Risk, credentials: A remote service needs WATERMARKS_SERVER_API_KEY set on it and in the client, so store that key securely and do not paste it into chat. Dual-use notice: only use this on software, sites and systems you own or are authorized to test. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/guillaumemeyer-watermarks-remover/clean-files-in-cowork-sessions/