DEDUPEDWho it is forA writer or marketer who drafts with Claude Code or Codex and needs a launch post, announcement, or article cleaned up before it goes out.
The problemAI drafts can be grammatically sound but still read like workslop, with canned openers, inflated claims, and promotional wording.
The outcomeA rewritten draft with flagged phrases removed, a writing score, and a final comparison with the original.
What you need first
- Claude Code or Codex (or another assistant that supports agent skills) for the skill workflow
- Python 3 for the offline local scorer
- Node.js 22+ for the npx CLI and the score command
- Internet access if you use the hosted CLI, REST API, or MCP endpoint
Steps
- Run
npx skills add manavmishra/ZeroSlop --globalto install the skill. - Open Claude Code or Codex in your project or writing folder.
- Run
/zero-slop (your writing)with the draft pasted or referenced. - To see flagged passages without an edit, run
/zero-slop inspect (your writing)first. - Compare the edit with the original and check that names, numbers, and claims are unchanged.
Prompt for your agent
Use the Zero Slop skill from the repo manavmishra/ZeroSlop on the draft at <path-to-draft>. Return the rewritten text and the writing score. Verify that every name, number, and link in the original still appears in the output.
Caveat. The skill runs inside your assistant and the README says the optional update check sends only metadata, so review the edited text yourself before publishing.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for manavmishra/ZeroSlop and "Remove AI slop from a draft inside Claude Code or Codex"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer