github:edwinhu/workflows · first committed 2026-01-07 · source · badge
This skill should be used when reviewing AI-generated text, checking for AI writing patterns, detecting undisclosed AI content, or before finalizing any written content. Covers 12 categories of AI writing indicators from Wikipedia's comprehensive guide.
Skill Health: not yet scored
Listed, not yet scored. This skill has no public model test yet. It is in the scoring queue; submitted skills go first, then the most-installed.
Version history
22 commits to this file on GitHub (newest first).
2026-10-15 · commit d4475f3 · docs: state the rule, drop the story (wc-probe P24) across 28 skills
2026-09-15 · commit 2eebd58 · constraints out of references/, one loader name, and a skill-shape probe
2026-09-15 · commit 407b2bb · skills: the TOC bang degrades instead of taking the skill down
2026-09-15 · commit 0ca2c7d · skills: trim the TOC preamble to one line
2026-09-15 · commit 751037e · skills: every skill lists its own references and scripts at load
2026-09-14 · commit 243c22e · prose: stop enforcing eleven rules the repo's own measurements refute
2026-08-22 · commit fb01e2c · feat: writing splits into a base and two domain registers, with a blocking guard
2026-08-22 · commit c0fd70b · refactor: ship the agents in user-agents/, symlinked into user scope
2026-08-21 · commit 09d66e6 · feat!: custom subagents per workflow, one register, one output style (v6.8.0)
2026-08-06 · commit 82bd2d0 · refactor: skill names say which beat they serve
2026-08-05 · commit 70c633d · feat: bound-abstraction tic — the object discriminates, not the word
2026-08-05 · commit 83e9735 · feat: two corpus-gated tics, and stop calling unenforced preferences a lint
2026-08-05 · commit 12a9bc0 · feat: make emphasis a first-class deterministic signal in the prose audit
2026-06-19 · commit c096b5c · skills(ai-anti-patterns): flag 'the cut' / 'X cut Y' for rule changes