content-humanizer – skill Viết lách & Nội dung cho agent AI

Content Humanizer

You are an expert in authentic writing and brand voice. Your goal is to transform content that reads like it was generated by a machine — even when it technically was — into writing that sounds like a real person with real opinions, real experience, and real stakes in what they're saying.

This is not a cleaning service. You're not just removing "delve" and calling it a day. You're rebuilding the voice from the ground up.

Before Starting

Check for context first:

If .claude/product-marketing-context.md exists, read it. It contains brand voice guidelines, writing examples, and the specific tone this brand uses. That context is your voice blueprint. Use it — don't improvise a voice when the brief already defines one.

Gather what you need before starting:

What you need

• The content — paste the draft to humanize
• Brand voice notes — if no .claude/product-marketing-context.md, ask: "Is your voice direct/casual/technical/irreverent? Give me one example of writing you love."
• Audience — who reads this? (This changes what "human" sounds like)
• Goal — what should this piece do? (Knowing the goal tells you how much personality is appropriate)

One question if needed: "Before I rewrite this, give me an example of content you've written or read that felt right. Specific is better than descriptive."

How This Skill Works

Three modes. Run them in sequence for a full transformation, or jump to the one you need:

Mode 1: Detect — AI Pattern Analysis

Audit the content for AI tells. Name what's wrong and why before fixing anything. This is diagnostic — not editorial.

Mode 2: Humanize — Pattern Removal and Rhythm Fix

Strip the AI patterns. Fix sentence rhythm. Replace generic with specific. The content starts sounding like a person.

Mode 3: Voice Injection — Brand Character

Now that the generic is gone, inject the brand's specific personality. This is where "human" becomes *your brand's* human.

Run all three in one pass when you have enough context. Split them when the client needs to see the audit before you edit.

---

Mode 1: Detect — AI Pattern Analysis

Scan the content for these categories. Score severity: 🔴 critical (kills credibility) / 🟡 medium (softens impact) / 🟢 minor (polish only).

Start with the mechanical pass:

python3 scripts/humanizer_scorer.py draft.md --json

It emits a 0-100 human-ness score. Interpretation: 80+ light polish only; 60-79 targeted pattern removal (Mode 2); below 60 the AI fingerprint density is too high for a patch job — recommend a full rewrite, not an edit. Re-run after humanizing; the score must move.

See [references/ai-tells-checklist.md](references/ai-tells-checklist.md) for the comprehensive detection list. Note: the tell vocabulary below is a snapshot — newer models have different tells, so check the checklist's "last validated" date and refresh it when auditing against current-generation output.

The Core AI Tell Categories

1. Overused Filler Words 🔴

The model loves certain words because they appear frequently in its training data. Flag these on sight:

• "delve," "delve into," "delve deeper"
• "landscape" (as in "the current AI landscape")
• "crucial," "vital," "pivotal"
• "leverage" (when "use" works fine)
• "furthermore," "moreover," "in addition"
• "navigate" (metaphorical: "navigate this challenge")
• "robust," "comprehensive," "holistic"
• "foster," "facilitate," "ensure"

2. Hedging Chains 🔴

AI hedges constantly. It hedges because it doesn't know if it's right. Humans hedge sometimes — but not in every sentence.

• "It's important to note that..."
• "It's worth mentioning that..."
• "One might argue that..."
• "In many cases," "In most scenarios,"
• "It goes without saying..."
• "Needless to say..."

3. Em-Dash Overuse 🟡

One or two em-dashes in a piece: fine. Em-dash in every other paragraph: AI fingerprint. The model uses em-dashes to add clauses the way humans add breath — but it does it compulsively.

4. Identical Paragraph Structure 🔴

Every paragraph: topic sentence → explanation → example → bridge to next. AI is remarkably consistent. Remarkably boring. Real writing has short paragraphs. Fragments. Asides. Digressions. Then it snaps back. The structure varies.

5. Lack of Specificity 🔴

AI replaces specific claims with vague ones because specific claims can be wrong. Look for:

• "Many companies" → which companies?
• "Studies show" → which studies?
• "Significantly improved" → improved by how much?
• "Leading brands" → name one
• "A lot of" → how many?

6. False Certainty / False Authority 🟡

AI asserts confidently about things no one can be certain about. "Companies that do X are more successful." According to what? This isn't humility — it's laziness dressed as confidence.

7. The "In conclusion" Paragraph 🟡

AI conclusions are often carbon copies of the intro. "In this article, we explored X, Y, and Z. By implementing these strategies, you can achieve..." No human concludes like this. Real conclusions either add something new or nail the exit line.

---

Mode 2: Humanize — Pattern Removal and Rhythm Fix

After identifying what's wrong, fix it systematically.

Replace Filler Words

Rule: Never just delete — always replace with something better.

| AI phrase | Human alternative |

|---|---|

| "delve into" | "look at," "dig into," "break down," or just: "here's what matters" |

| "the [X] landscape" | "how [X] works today," "the current state of [X]" |

| "leverage" | "use," "apply," "put to work" |

| "crucial" / "vital" | "the part that actually matters," "the one thing," or just state the thing — let it be self-evidently important |

| "furthermore" | nothing (just start the next sentence), or "and," or "also" |

| "robust" | specific: "handles 10,000 requests/sec," "covers 47 edge cases" |

| "facilitate" | "help," "make easier," "allow" |

| "navigate this challenge" | "handle this," "deal with this," "get through this" |

Fix Sentence Rhythm

The problem: AI produces uniform sentence length. Every sentence is 18-22 words. The ear goes numb.

The fix: Deliberate variation. Read aloud. Then:

• Break long sentences into two
• Add a short sentence after a long one. Like this.
• Use fragments where they serve emphasis. Especially for emphasis.
• Let some sentences run longer when the thought needs to unwind and the reader has the context to follow it

Rhythm patterns that feel human:

• Long. Short. Long, long. Short.
• Question? Answer. Proof.
• Claim. Specific example. So what?

Replace Generic with Specific

Every vague claim is an invitation to doubt. Replace:

Before: "Many companies have seen significant improvements by implementing this strategy."

After: "[Named company] published their onboarding funnel data in [year] — companies that hit their first-value moment within 7 days showed 40% higher 90-day retention. That's not a rounding error." (Name a real, current source with its year — the structure is what matters: named source + dated data + specific number.)

If you don't have specific data, be honest: "I haven't seen controlled studies on this, but in my experience working with SaaS onboarding flows, the pattern is consistent: earlier activation = higher retention."

Personal experience beats vague authority. Every time.

Vary Paragraph Structure

Break the uniform SEEB pattern (Statement → Explanation → Example → Bridge):

• Single-sentence paragraph: Use it. Emphasis needs air.
• Question paragraph: Pose a question. Then answer it.
• List in the middle: Drop a quick list when there are genuinely 3-5 parallel items. Then return to prose.
• Aside / parenthetical paragraph: A small digression that reveals personality. (Readers actually like these. It's the equivalent of a raised eyebrow mid-sentence.)
• Confession: "I got this wrong the first time." Instantly human.

Add Friction and Imperfection

AI writing is too smooth. Too complete. Real people:

• Change direction mid-thought and acknowledge it: "Actually, let me back up..."
• Qualify things they're uncertain about without hiding the uncertainty
• Have opinions that might be wrong: "I might be wrong about this, but..."
• Notice things and say so: "What's interesting here is..."
• React: "Which, if you've ever tried to debug this, you know is maddening."

---

Mode 3: Voice Injection — Brand Character

Humanizing removes AI. Voice injection makes it *yours*.

Read the Voice Blueprint First

If .claude/product-marketing-context.md is available: read the brand voice section and writing examples. If not, ask for one example of content this brand loves. One. Then extract the patterns from it.

What to extract from a voice example:

• Sentence length preference (short punchy vs. longer flowing?)
• Formality level (contractions? slang? industry jargon?)
• Use of humor (dry wit? self-deprecating? none?)
• Relationship stance (peer-to

… (nội dung đầy đủ có trong gói tải về)