Add Detection Guidance: false positives, human-writing signs, LLM idiolects#113
Open
philippdubach wants to merge 1 commit intoblader:mainfrom
Open
Add Detection Guidance: false positives, human-writing signs, LLM idiolects#113philippdubach wants to merge 1 commit intoblader:mainfrom
philippdubach wants to merge 1 commit intoblader:mainfrom
Conversation
…olects Most of this skill tells the editor what to remove. This adds the inverse — what to leave alone, and how to decide. Sourced from Wikipedia: Signs of AI writing (revision fetched 2026-05-01), specifically the "Ineffective indicators", "Signs of human writing", and "Differences between LLMs" sections. Three subsections, no new patterns: - "What NOT to flag (false positives)" — the indicators that look AI-coded but are actually neutral (perfect grammar, em dashes alone, curly quotes alone, formal vocabulary, common transition words). The over-editing risk is real: if the skill is applied too aggressively, it strips legitimate prose. Closes with the "clusters matter, isolated signs don't" rule. - "Signs of human writing (preserve these)" — positive markers that should be left untouched: specific detail, mixed feelings, era-bound references, sentence-length variation, parenthetical self-corrections, and the November 30, 2022 cutoff for ruling out AI involvement entirely. - "LLM Idiolects" — quick triage notes per model family (ChatGPT/Grok verbose with artifacts; Gemini/Claude concise, no curly quotes by default). Tendencies, not rules. No pattern-count change. No README changes (the README's pattern table is unaffected since this section is meta-guidance, not new patterns). No version bump.
philippdubach
added a commit
to philippdubach/humanizer
that referenced
this pull request
May 1, 2026
Brings the fork's main branch in line with the maintained local v2.6.0, consolidating the changes that are also opened as focused PRs against blader/humanizer (blader#111, blader#112, blader#113): - §7 expanded with era-specific AI vocabulary clusters (GPT-4 / GPT-4o / GPT-5 eras), plus 'bolstered' and 'meticulous' added to the master list and a literal-vs-figurative caveat. - §21 renamed to "Knowledge-Cutoff Disclaimers and Speculative Gap-Filling"; covers the retrieval-augmented "maintains a low profile" / "keeps personal details private" speculation pattern. - New patterns §30-34: reference-markup artifacts (turn0search0, oaicite, utm_source=chatgpt.com, etc.), placeholder leftovers, Markdown/wikitext contamination, formal "Conclusion" closers, didactic disclaimers. - New Detection Guidance group: what NOT to flag (false positives), signs of human writing to preserve, and per-model LLM idiolects. Frontmatter version bumped to 2.6.0. README pattern table updated (29 → 34 patterns) with a new Artifacts and Contamination section and a pointer to Detection Guidance. WARP.md count corrected from the stale "25 patterns" to 34. Sourced from Wikipedia: Signs of AI writing (revision fetched 2026-05-01).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Most of this skill tells the editor what to remove. This adds the inverse — what to leave alone, and how to decide.
Sourced from Wikipedia: Signs of AI writing (revision fetched 2026-05-01), specifically the "Ineffective indicators", "Signs of human writing", and "Differences between LLMs" sections.
Three subsections, no new patterns and no version bump:
The new section sits between §29 (Fragmented Headers) and the Process section. README and WARP.md are unchanged — the README pattern table is unaffected since this is meta-guidance, not new patterns.
Why this matters
The current skill is a strong "what to fix" guide but doesn't push back against over-editing. In practice, applying any pattern-based rewriter too aggressively is the failure mode — humans get their voice flattened. A short "what NOT to flag" sidebar plus a positive-signals list keeps the editor honest.
This PR is independent of #111 (era vocab + gap-filling) and #112 (patterns 30–34), and can land on its own without rebasing.
Test plan
Source: Wikipedia:Signs of AI writing.