Slopkit skill
Two anti-slop skills: slopbeth edits writing you publish and slopgent keeps the agent's own replies honest and plain.
- 113 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Installation runs code from GitHub through `npx` and writes into the skills folders of several agents.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/slopkitDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Installation runs code from GitHub through `npx` and writes into the skills folders of several agents.
This is third-party code. Review the repository files before installing.
What it does
Slopkit installs two independent skills for an agent. `slopbeth` rewrites and reviews a finished artifact (a README, a postmortem, a release note), cutting AI tells without changing what the text can honestly claim. `slopgent` shapes the agent's own replies: status reports describe only what actually ran, the action comes first and the language stays plain. One install command puts both skills into the Codex, Claude Code, Hermes, OpenClaw, OpenCode and Pi directories that already exist in your home folder. A plugin route for Claude Code and Codex through a marketplace exists, along with a PowerShell 7 edition of `slopbeth` that needs neither Node nor Python. Either skill can be deleted and the other keeps working.
Who it is for. Authors of documentation, release notes and posts who edit agent drafts, and anyone tired of "all fixed" replies with nothing verified behind them.
Good fit when
- A README or post drafted by an agent is full of smooth phrases with no content.
- A release note should keep every verified fact and lose only filler.
- An agent reports "done" at the end without saying which commands it ran.
- Status messages are long and the one command you need gets lost in the middle.
Not a fit when
- A guarantee of passing AI-text detectors is wanted, which the authors do not promise.
- The text is legal or academic where passive voice and formal style fit.
- A plain "write shorter" rule is enough, while these skills aim to preserve caveats and facts.
Example request
Rewrite this README draft with `slopbeth`, keeping all facts and versions, and show a list of changes.Limitations
The authors state that the skills promise no immunity from AI-text detectors. README comparisons use designed case sets scored by a small panel and do not prove a general edge. `slopbeth` needs Node and Python (or PowerShell 7), `slopgent` needs Python and, for its memory helper, Node. Installation runs through `npx github:ehmo/slopkit`, which fetches code from GitHub. The rules are tuned most finely for English text. The last commit is from July 2026.
How to disable. Delete the `slopbeth` and `slopgent` folders from the `skills/` directories of the relevant agents. Remove plugins through the Claude Code or Codex plugin manager.
FAQ
How does `slopbeth` differ from `slopgent`?
`slopbeth` edits text you publish. `slopgent` sets the manner of the agent's own replies: honesty about what was done, action first, plain language.
Can I install just one skill?
Yes. They sit in separate folders `skills/slopbeth` and `skills/slopgent`, and deleting either leaves the other working.
Which agents are supported?
Codex, Claude Code, Hermes, OpenClaw, OpenCode and Pi. The install command skips agents whose directories are missing and `--all` installs into all of them.
How do I install as plugins?
Run `npx github:ehmo/slopkit install-plugin` for both agents or add `claude` or `codex`. Through the Claude Code marketplace: `/plugin marketplace add ehmo/slopkit`.
Are Node and Python required?
For `slopbeth`, yes, though on Windows `pwsh` 7 running `bin/slopkit.ps1` is enough. `slopgent` needs Python.
Discussion
No questions or reviews yet.
Related
SkillsA free skill that finds AI writing patterns in a draft, guides the edit and checks that the author's facts and wording survived
A skill and CLI that strip AI tells from Korean, English, Chinese and Japanese text and check that meaning and numbers stay unchanged
A library of 64 skills for founders: GTM, growth, SEO, email, graphics and video, installed with one npx command for Claude Code, Codex and more
A skill for Claude Code, Codex and Cursor: strips signs of AI writing and checks the result against human-writing ranges.
A skill that finds AI tells in English prose and edits them surgically; three Python scanners run on their own and the teach command learns your voice
A prose-checking plugin that runs a local detector after supported file saves and gives the agent editing feedback
README in short
The README describes the two skills and how they differ, installation from the command line and as plugins, the PowerShell edition, comparison results against other packs and before and after examples.