prose-writing

What this skill does

# Prose Writing

Founders write a lot of prose: blog posts, essays, build-in-public updates, About pages, newsletter intros. Hand the task to AI and you get slop — hedged, generic, three-item-list cadence, throat-clearing openings, "this highlights" tails. This skill is the proactive composition workflow: voice first, structure second, drafting third, edit last.

> **If `ABOUT-ME.md` exists in the project root**, read it before writing — it carries the founder's voice. **If `MY-ICP.md` exists**, read it for who the prose is talking to. The founder's voice delivers the customer's frame.

## Three principles of prose that doesn't sound like AI

- **Lead with a specific moment, claim, or number — not a category.** "I shipped a feature on Tuesday and lost two customers by Friday" beats "Building products comes with tradeoffs."
- **Trust the reader.** State what you mean. Don't gesture at significance ("This highlights an important point about…"). The reader can see the point if you make it.
- **Structure invisibly.** Every paragraph earns its place by moving the argument or story forward. If a paragraph could be deleted with no loss, delete it.

---

## Workflow

```
- [ ] Read ABOUT-ME.md and MY-ICP.md if they exist
- [ ] Build a voice card (once, reusable)
- [ ] Pick the format and state the one takeaway
- [ ] Write the opening line first
- [ ] Outline as 3–5 plain-English beats
- [ ] Draft fast, in voice. Don't edit while drafting.
- [ ] Edit pass: throat-clearing, "this highlights," ending
- [ ] Run humanize if it still sounds AI
```

---

## Voice discovery (do once, reuse forever)

Before writing in your voice, you need to know what your voice is. This is a 15-minute exercise.

1. Find 3 pieces of prose you wish you'd written. Blog posts, essays, threads. Not "good writing" in the abstract — writing that sounds like you on your best day.
2. For each piece, note:
   - **Sentence length pattern.** Mostly short? Mixed? Long sentences with short punches?
   - **Vocabulary register.** Plain ("we shipped this thing")? Mid ("we released the feature")? High ("we deployed the capability")?
   - **Opinion stance.** Stated directly? Hedged? Self-aware about uncertainty?
   - **Self in the prose.** First person constantly? Occasionally? Never?
   - **Humor.** Dry? Earnest? None?
3. Distill into a 5-line **voice card** you paste into every prompt:

```
Voice card:
- Sentence rhythm: [e.g., short punches with one longer setup per paragraph]
- Register: [e.g., plain, casual, no jargon unless naming a real thing]
- Opinions: [e.g., stated directly, qualified once if needed, never hedged twice]
- First person: [e.g., I/we present throughout — this is a person talking]
- What I never do: [e.g., listicles, "in conclusion," motivational closers, three-item lists]
```

Save this in `ABOUT-ME.md` so every skill in this plugin can use it.

**Tell AI:**
```
Help me build my voice card. I'll paste 3 pieces of writing I wish I'd written.
For each: identify sentence rhythm, register, opinion stance, first-person presence, humor.
Then distill into a 5-line voice card I can paste into prompts.
Don't flatten the differences — find what's consistent across all three.

Piece 1: [paste]
Piece 2: [paste]
Piece 3: [paste]
```

---

## Format: Blog post (800–1500 words)

**Shape:** hook → premise → 2–4 beats with examples → resolution that earns a takeaway.

- **Hook (first 1–3 sentences):** a specific moment, claim, or number. Not a definition. Not a question. Not "Imagine if…"
- **Premise (next paragraph):** the one thing this post argues or shows. State it.
- **Beats:** each beat = one point, one example, one paragraph (or two). Examples are concrete. If you find yourself writing "for instance, consider a scenario where…" delete and use a real one.
- **Resolution:** the post lands somewhere. Not a recap. Not "in conclusion." The strongest line you've got, placed last.

**Anti-pattern:** the listicle that's actually a content-marketing skeleton — H2 / paragraph / H2 / paragraph, all same-length, same-cadence, no argument moving through.

**Tell AI:**
```
Write a [800/1200/1500]-word blog post on [topic].
Voice: [paste voice card]
The one takeaway: [in one sentence, what the reader walks away believing]
Structure: hook with [specific moment / claim / number] → premise → [3–4] beats with concrete examples → resolution that lands on the strongest line.
Constraints:
- No "this highlights / underscores / demonstrates"
- No -ing analysis tails ("...marking a shift toward...")
- No three-item lists unless they're actually three things
- No "in conclusion" or "to summarize"
- End on a strong line, not a recap
- Read aloud test: cut anything I wouldn't say
```

---

## Format: Founder essay (1500–3000 words)

**Shape:** personal stake → tension or question → exploration → arrived-at view.

- **Personal stake:** why are you the one writing this? A specific experience, decision, mistake. Not a credential.
- **Tension:** the thing you didn't know, or got wrong, or are still working out. The whole essay turns on this.
- **Exploration:** you working through it. Multiple beats. It's allowed to wander a little — that's what makes it an essay and not a blog post.
- **Arrived-at view:** what you think now. Not a tidy thesis. Closer to "here's where I've landed, with these open questions."

**Anti-pattern:** the LinkedIn-influencer "lessons learned" cadence. "Lesson 1. Lesson 2. Lesson 3." If you find yourself bullet-pointing the takeaway, you've stopped writing an essay.

**Tell AI:**
```
Write a [1500/2000/2500]-word founder essay.
Voice: [paste voice card]
Personal stake: [the specific experience this essay turns on]
Tension: [what I didn't know, got wrong, or am still working out]
Where I've landed (roughly): [your current view, with open questions allowed]
Constraints:
- This is an essay, not a blog post. It can wander. It doesn't need a tidy thesis.
- No "lessons learned" or numbered takeaways
- No motivational closer ("the journey continues...")
- Open with the specific experience, not a setup
- End where my thinking actually ends — open questions are fine
```

---

## Format: Build-in-public update (200–500 words)

For threads, single posts, weekly recaps, monthly reviews.

**Shape:** the number or shipped thing → what was hard → what changed in your view.

- **Open with the specific.** "We hit $4K MRR last week" or "I shipped the new onboarding on Tuesday and pulled it back on Wednesday." Not "Here's an update on what we've been up to."
- **What was hard:** the actual hard part, not the polished version. If the hard part is "I almost gave up on Thursday," say that.
- **What changed in your view:** the small reframe. The thing you'll do differently. Specific.

**Anti-pattern:** the progress brag with no specifics ("Made tons of progress! Excited for what's next!"). Or the formulaic three-emoji-bullet recap.

**Tell AI:**
```
Write a build-in-public [thread / single post / weekly recap], [200/300/500] words.
Voice: [paste voice card]
The number / shipped thing: [specific]
What was hard: [the actual hard part]
What changed in my view: [the reframe or new decision]
Constraints:
- Open on the specific. No "Here's what I've been up to."
- Name the hard part honestly, not the polished version
- One reframe at the end, not three takeaways
- No celebratory emoji-bullets
```

---

## Format: About page / origin story (300–600 words)

**Shape:** the problem you ran into → why existing answers didn't work → what you're building → who it's for.

- **The problem you ran into:** a specific situation, not a market. "I was running invoicing for my agency in three spreadsheets" beats "Small businesses face complex billing challenges."
- **Why existing answers didn't work:** for *you*. Be specific about which ones and why. This is also positioning.
- **What you're building:** plain language. What it does, not what it represents.
- **Who it's for:** named, not "anyone who needs…"

**Anti-pattern:** the compa