Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
63 lines (47 loc) · 3.72 KB

CONTRIBUTING.md

File metadata and controls

63 lines (47 loc) · 3.72 KB
Raw

Contributing to Humanizer

Thanks for helping make AI text more human. Here's how to contribute.

Adding a new pattern

  1. Follow the existing format (P1-P53 use this exact template — see SKILL.md for examples)
  2. Include:
    • Trigger words: what to search for (regex-friendly when possible)
    • What's happening: why AI does this — cite a source if you have one
    • Fix: how to replace it (prescriptive, not "consider...")
    • Before/after table: at least one real example, not invented
  3. Place it in the right category (Content P1-P8, Language & Style P9-P18, Communication P19-P21, Filler & Hedging P22-P30, Emerging P31-P43, Craft & Forensic P44+)
  4. Update three places in lockstep:
  5. Put the short entry (name, trigger, one-line fix, before/after) in SKILL.md; put any longer "what's happening" notes or sources in references/patterns.md so the core stays lean.

Improving an existing pattern

  • Add more trigger words from real AI text you've encountered
  • Add before/after examples
  • Refine the fix instructions

Reporting false positives

If the skill flags something that's genuinely human writing, open an issue with:

  • The flagged text
  • Which pattern triggered
  • Why it's a false positive

Pull request process

  1. Fork and create a branch (feat/pattern-31-whatever or fix/false-positive-p7)
  2. Make your changes in SKILL.md
  3. Keep SKILL.md lean: it is the always-loaded core, so push verbose deep dives, extra examples, and sources into references/ rather than growing the core. New patterns add a compact entry here and their depth there.
  4. If you touch the CLI (cli/), run cd cli && node --test and keep it green. New metrics or vocabulary entries should ship with tests (aim for a few positive and negative cases each, in the spirit of "every pattern earns its tests").
  5. Open a PR with a clear description

Code of conduct

Be direct. Be helpful. Skip the pleasantries. We're humanizers, not sycophants.

Spreading the word

If you found this skill useful, the highest-leverage way to help is submitting it to skill aggregators. Each row below is one PR away from listing.

Aggregator URL Submission method
awesome-claude-code hesreallyhim/awesome-claude-code Open issue using "[Resource]" template, then PR adding row to THE_RESOURCES_TABLE.csv
awesome-claude-skills ComposioHQ/awesome-claude-skills Fork, add to "Communication & Writing", PR
awesome-claude-skills (Travis) travisvn/awesome-claude-skills README PR
awesome-claude-code (jquery) jqueryscript/awesome-claude-code README PR
awesome-claude-code (jman) jmanhype/awesome-claude-code README PR
ClawHub docs.openclaw.ai/clawhub clawhub skill publish
LobeHub Skills lobehub.com/skills Web form
awesomeclaude.ai awesomeclaude.ai Submission form
claudepluginhub claudepluginhub.com Auto-indexed via repo topics

PRs in this direction are very welcome. If you submit and the listing goes live, drop a link in an issue so we can credit you here.