# AI use cases for Sloptrim (seyedehsanhadi/sloptrim) Repo: [seyedehsanhadi/sloptrim](https://github.com/seyedehsanhadi/sloptrim) | Use cases: 4 | Page: https://dedupedai.com/use-cases/seyedehsanhadi-sloptrim/ Sloptrim is a local command-line tool and agent plugin that scores prose files for AI-writing patterns, using only the Python standard library with no network and no model. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Score a prose draft for AI-writing tells: https://dedupedai.com/use-cases/seyedehsanhadi-sloptrim/score-draft-before-publishing/ - Guard prose an agent saves in Claude Code: https://dedupedai.com/use-cases/seyedehsanhadi-sloptrim/guard-agent-prose-on-save/ - Add a prose contract to AGENTS.md and Cursor rules: https://dedupedai.com/use-cases/seyedehsanhadi-sloptrim/apply-writing-contract-to-repo/ - Screen Office, EPUB and notebook files for tells: https://dedupedai.com/use-cases/seyedehsanhadi-sloptrim/screen-office-and-notebook-files/ ## How to score a prose draft for AI-writing tells with Sloptrim Repo: [seyedehsanhadi/sloptrim](https://github.com/seyedehsanhadi/sloptrim) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README Run a local 0-100 score on a .docx, .md or other prose file before you publish it. **Who it is for:** An editor or content writer who has a finished draft and wants a local check for AI-writing patterns before sending it out. **The problem:** Reading a draft line by line to find machine-like phrasing is slow, and uploading the draft to a web service is not acceptable for unpublished work. **The outcome:** A JSON report with a 0-100 score, one of five bands (clean, light tells, mixed, heavy tells, pervasive tells), the named patterns and the metrics, all produced on your machine. ### Steps 1. Clone the repository with `git clone https://github.com/seyedehsanhadi/sloptrim.git ~/.claude/skills/sloptrim` 2. Change into the cloned folder, for example `cd ~/.claude/skills/sloptrim` (inferred, check the README) 3. Run `python scripts/detect.py draft.docx` on your file, replacing draft.docx with its path 4. Read the score and band in the JSON, then check the listed patterns and metrics 5. In Claude Code, run `/sloptrim check ` to get a score and the named tells without a rewrite (inferred, check the README) ### Prompt to paste into your agent ```text Using the sloptrim repo at , run `python scripts/detect.py ` on and report the 0-100 score, the band and the named patterns. Verify that the output is valid JSON and includes a score between 0 and 100. ``` ### Caveat The score describes the writing only and must not be used to judge whether a person or a model wrote the text, as the README warns. 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/seyedehsanhadi-sloptrim/score-draft-before-publishing/ ## How to guard prose an agent saves in Claude Code with Sloptrim Repo: [seyedehsanhadi/sloptrim](https://github.com/seyedehsanhadi/sloptrim) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Install the plugin so each prose file the agent saves is scored and flagged for rewrite. **Who it is for:** A developer or writer who uses Claude Code to produce documents and wants a check on prose the agent writes to disk. **The problem:** An agent can save long prose with machine-like patterns, and reviewing every saved file by hand is slow. **The outcome:** Each prose file saved through Write or Edit gets a score, and the guard asks for flagged spans to be fixed when the score passes the threshold, with the default full mode. ### Steps 1. In Claude Code, paste `Install the sloptrim plugin from https://github.com/seyedehsanhadi/sloptrim` and restart 2. Run `/sloptrim doctor` and confirm it prints four [OK] lines 3. Save a prose file through the agent, then run `/sloptrim show` and look for its entry 4. Pick a mode with `/sloptrim full`, `/sloptrim strict`, `/sloptrim lite` or `/sloptrim off` (inferred, check the README) ### Prompt to paste into your agent ```text Install the sloptrim plugin from https://github.com/seyedehsanhadi/sloptrim, then run `/sloptrim doctor` and report whether it prints four [OK] lines. Save one prose file and verify that it appears in `/sloptrim show`. ``` ### Caveat The README says a file written by a Bash command reaches disk without passing Write or Edit, so the guard will not see it. 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/seyedehsanhadi-sloptrim/guard-agent-prose-on-save/ ## How to add a prose contract to AGENTS.md and Cursor rules with Sloptrim Repo: [seyedehsanhadi/sloptrim](https://github.com/seyedehsanhadi/sloptrim) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Write the Sloptrim contract into a project so Codex or Cursor agents follow it. **Who it is for:** A team lead or developer whose project uses Codex or Cursor to write documentation and wants the same prose rules applied in that repo. **The problem:** Each agent session starts without the team's prose rules, so the output varies from one run to the next. **The outcome:** The project gets an AGENTS.md file with the Sloptrim contract and a Cursor rule in .cursor/rules/, so agents read the same guidance. ### Steps 1. Open the project folder where the agent will work 2. Run `/sloptrim init` in an agent session that has the Sloptrim plugin installed 3. Open the new `./AGENTS.md` file and confirm the contract is present 4. Check that a Cursor rule was written to `./.cursor/rules/` 5. Commit the new files so the whole team gets the same contract ### Prompt to paste into your agent ```text In the project at , run `/sloptrim init` and report which files were written. Verify that AGENTS.md exists and that a Cursor rule exists under .cursor/rules/. ``` ### Caveat The README describes the contract as writing advice, and 12 of the 62 detectors count for nothing in the score, so the contract does not guarantee a low score. 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/seyedehsanhadi-sloptrim/apply-writing-contract-to-repo/ ## How to screen Office, EPUB and notebook files for tells with Sloptrim Repo: [seyedehsanhadi/sloptrim](https://github.com/seyedehsanhadi/sloptrim) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Score the prose in .docx, .pptx, .xlsx, .epub and .ipynb files from the command line. **Who it is for:** A research or operations person who keeps reports, slide decks or notebooks in Office or Jupyter formats and wants a local check before sharing them. **The problem:** Office and notebook files do not show their prose clearly in a plain text view, and the check has to skip code, notes and comments. **The outcome:** A score for the selected prose in each file, with an explicit list of what the detector reads and what it skips, so the result is read with the right scope. ### Steps 1. Clone the repository and change into the folder 2. Run `python scripts/detect.py report.docx` on a Word file, replacing report.docx with its path (inferred, check the README) 3. Run `python scripts/detect.py slides.pptx` and `python scripts/detect.py analysis.ipynb` for other formats (inferred, check the README) 4. Read the README limits for each format, since DOCX skips headers, footers and comments and notebooks read Markdown cells only 5. Do not redirect `--clean` output over an Office archive, EPUB or notebook; use a format-aware editor for revisions ### Prompt to paste into your agent ```text Using the sloptrim repo at , run `python scripts/detect.py ` on each of and report the score and the skipped areas for each. Verify that no output was redirected over an Office, EPUB or notebook file. ``` ### Caveat PDF and RTF are recorded as unreadable rather than scored, and embedded images and OCR are unsupported, so those parts of a file are not checked. 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/seyedehsanhadi-sloptrim/screen-office-and-notebook-files/