Skillsstraight-prose
Writes and edits prose so it reads like a specific person wrote it and carries no stock AI patterns such as inflated significance, promotional wording, vague attribution, negative parallelisms, forced threes, em dashes, chatbot…
7 files and 506 lines, free under MIT. Built from arm-pz/straight-prose at 216bf85.
Writes and edits prose so it reads like a specific person wrote it and carries no stock AI patterns such as inflated significance, promotional wording, vague attribution, negative parallelisms, forced threes, em dashes, chatbot pleasantries and over-formatting. Applies the rules while drafting, edits existing text without changing the author's voice or dialect, or audits a draft and lists each pattern with its line, using a scanner script for the mechanical ones. Use whenever the user asks to remove AI slop or de-slop a draft, asks "does this sound like AI" or "why does this read like ChatGPT", wants writing "in my voice" or "less like a chatbot", or pastes anti-slop style rules to apply, even if they never say "slop".
Version, date and limits are read from the repo on every sync, and the check runs every night. Real-world runs are listed only with their evidence.
Measured from the packed archive at commit 216bf85. A skill is plain text your agent reads, so this list is all of it.
| File | Lines | Bytes |
|---|---|---|
| .gitignore | 4 | 35 B |
| SKILL.md | 160 | 11,615 B |
| evals/evals.json | 29 | 2,208 B |
| evals/trigger-cases.md | 47 | 4,467 B |
| references/sourcing.md | 43 | 2,506 B |
| scripts/slop_scan.py | 153 | 7,567 B |
| scripts/test_slop_scan.py | 70 | 3,871 B |
| 7 files | 506 | 13 KB |
Read it before you install it: the skill lives at . in arm-pz/straight-prose.