Integrate PR blader#94: quality scoring + core principles + 3 new patterns

Upstream blader/humanizer PR blader#94 (Charpup). Adds:
- Patterns 33-35 (renumbered from PR's 30-32 to avoid local conflict):
  Citation Fabrication, Broken Markup Artifacts, Era-Specific Vocabulary
  Clustering
- 5 Core Principles quick-reference block
- Quick Checklist (pre-delivery audit)
- Quality Scoring rubric (/50 across 5 dimensions, target >=40/50)
- Detection Confidence Framework (HIGH/MEDIUM/LOW table)

Renumber: local already has patterns 30-32 (Diff-Anchored, Conditional
Frame Stacking, Miscalibrated Epistemic Confidence) from local commit
c24f39c + 864a03a. PR's new patterns shifted to 33-35 and the
Detection Confidence HIGH row updated to include P33/P34/P35.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: spencergoldade

SKILL.md

@@ -579,6 +579,42 @@ Both are tells. The fix is not to replace one extreme with the other — it is t
 
 **Critical rule:** Do not fix over-assertion by adding hedges. Fix it by narrowing the claim.
 
+
+## RELIABILITY AND EVIDENCE PATTERNS
+
+### 33. Citation Fabrication / Hallucinated Sources
+
+**Problem:** LLMs sometimes generate citations that look legitimate but are off-topic, have invalid DOIs, or reference non-existent publications. Also watch for: broken external links, `utm_source=` parameters in URLs, named references declared but never used.
+
+**Fix:** Verify every citation. Remove unverifiable ones.
+
+**Before:**
+> According to Smith et al. (2023) in the Journal of Digital Innovation...
+
+**After:** [verify the source exists and says what you claim, or remove]
+
+
+### 34. Broken Markup Artifacts
+
+**Problem:** AI-generated text may contain markup from a different context: markdown syntax in Word documents, `oaicite` tags, `contentReference` spans, or `turn0search0` references.
+
+**Fix:** Remove all AI markup artifacts. Use the platform's native formatting.
+
+**Before:**
+> The results were **significant** (see `contentReference[oaicite:0]`)
+
+**After:**
+> The results were significant (see Table 2).
+
+
+### 35. Era-Specific Vocabulary Clustering
+
+**Problem:** AI vocabulary clusters by model era. When you spot one high-frequency AI word, scan the surrounding text. Others from the same era tend to cluster in the same paragraph. Finding 3 or more AI vocabulary words in one paragraph is a strong signal.
+
+**Fix:** When you spot a cluster, rewrite the entire paragraph, not just individual words.
+
+> Patterns 33-35 sourced from [Wikipedia: Signs of AI writing](https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing) and [WikiProject AI Cleanup](https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup).
+
 ---
 
 ## Process
@@ -610,6 +646,59 @@ Provide:
 **Absolute constraint on the final rewrite:** It must contain zero em dashes (—) and zero en dashes (–). If a dash slips through, the output is treated as a failed draft, not a final answer.
 
 
+## 5 Core Principles (Quick Reference)
+
+1. **Delete filler.** Remove filler words and emphatic crutches. "It is important to note" means nothing. Delete it.
+2. **Break formula.** Avoid binary contrasts ("not just X, but Y"), dramatic segmentation, rhetorical setup. Start with substance.
+3. **Vary rhythm.** Mix sentence lengths. Short. Then longer ones that develop the thought. Two items beat three. Vary paragraph endings.
+4. **Trust the reader.** State facts directly. Skip softening, excuses, hand-holding. If you wrote "to be sure" or "of course," delete it.
+5. **Kill the quotables.** If a sentence sounds like it belongs on a motivational poster, rewrite it. Real people don't speak in epigrams.
+
+> Adapted from [hardikpandya/stop-slop](https://github.com/hardikpandya/stop-slop).
+
+
+## Quick Checklist (Pre-Delivery Audit)
+
+- [ ] Three consecutive sentences same length? Break one up.
+- [ ] Paragraph ends with short single line? Vary the ending.
+- [ ] Em dash before a revelation? Delete it.
+- [ ] Explaining a metaphor? Trust the reader.
+- [ ] Used "Additionally" / "However" / "Furthermore"? Consider deletion.
+- [ ] Rule of three anywhere? Change to two or four.
+
+
+## Quality Scoring
+
+After rewriting, score your output across 5 dimensions (1-10 each, total /50):
+
+| Dimension | 10 = | 1 = |
+|-----------|------|-----|
+| **Directness** | Every sentence delivers info immediately | Every paragraph starts with setup |
+| **Rhythm** | Mix of short and long, varied endings | Mechanical uniformity |
+| **Reader Trust** | Concise, no over-explanation | Hand-holds throughout |
+| **Authenticity** | Has voice, opinions, specific observations | Obviously AI, formulaic |
+| **Conciseness** | Zero redundancy | Could be cut 40%+ |
+
+**Target: ≥ 40/50 before delivery.**
+
+> Scoring system adapted from [op7418/Humanizer-zh](https://github.com/op7418/Humanizer-zh), originally from [hardikpandya/stop-slop](https://github.com/hardikpandya/stop-slop).
+
+
+## Detection Confidence Framework
+
+Not all patterns carry equal weight. Individual LOW patterns may be normal human style.
+
+| Confidence | Patterns | Alone? |
+|-----------|----------|--------|
+| **HIGH** | P1, P3, P5, P6, P7 (clustered), P16, P18, P20, P21, P22, P25, P28, P30, P31, P32, P33, P34, P35 | Strong signal |
+| **MEDIUM** | P2, P4, P8, P9, P10, P11, P14, P15, P23, P24, P27, P29 | Suspicious if 2+ |
+| **LOW** | P12, P13, P17, P19, P26 | Normal unless clustered |
+
+**Combined Indicator Principle:** 3 or more categories matching in the same text = high probability AI origin.
+
+> Framework based on [Wikipedia: Signs of AI writing](https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing).
+
+
 ## Full Example
 
 **Before (AI-sounding):**