arquitetar-oferta

Offer Architect

Most offer copy fails because it sells stuff instead of access, leads with elevated content language instead of mundane outcomes, and runs the same promotion until the audience tunes out. This skill builds offers that fix all three at once: a shell of access tiers, a Promise/Path/Pricing/Why-Now map written in the customer's words, a protagonist arc that conditions the audience to buy, and a daily Red Velvet rhythm so offers ship every day instead of every quarter.

the core job

Produce a complete, copy-pasteable offer package the user can drop into a sales doc, email, or landing page today. Includes: offer name, one-sentence promise, tier table, sales bullets, why-now hook, and CTA — all run through the anti-slop pass.

Output: A single offer doc structured per the Output Format section below, plus one suggestion for what to refine next session.

core philosophy

Make offers daily. The #1 KPI is Offers Made — not created, MADE.

Until you fix offer frequency, nothing else matters. Broken links, ugly design, simple systems — if you make an offer every day (sometimes more), you'll win.

before starting: gather context

STOP. Do not produce output until context is in hand AND the brief is clear. Cap questions at 2 max.

Step A: Gather context (fast)

1. Check CLAUDE.md first. Claude auto-loads ~/.claude/CLAUDE.md and any project-level CLAUDE.md. Scan for:
   • Business — what the user sells, current pricing tiers, delivery model.
   • ICP — pain points, objections, the words they use. Promise/Path/Why Now must speak to THIS person.
   • Voice — tone, do/don't, sample sentences. Offer copy must sound like the user wrote it.
2. If missing, check the project folder for Context/business.md, Context/icp.md, Context/voice.md. Read whichever exist.
3. If voice is missing after both: flag it — "I'll write the offer copy generic-tone and you can pass it through /anti-slop after."

Step B: Confirm the brief (≤2 questions, combined)

Skip anything context already covers. Combine into at most 2 questions:

1. Framework + offer — "Which framework are we starting with (Offer Shell, Offer Map, Protagonist, or 90 Days?) and what's the offer (or which existing one are we refining)?"
2. Why Now + pricing — "What's the deadline or Why Now angle, and the price point / tier structure?" (Skip the pricing half if business context covers it.)

Confirm the brief in 2-3 lines back to the user before writing.

the 4 frameworks

Framework	Purpose	When to Use
Offer Shell	Structure that monetizes access, not stuff	Building offer tiers, upsell paths
Offer Map	Craft compelling offer copy	Writing sales pages, Google Docs
Protagonist	Position as a leader people follow	Personal branding, content strategy
90 Days of Offers	Daily selling playbook	Email sequences, promotion planning

the process

[Context + Brief]
      |
      v
Step 1: STRUCTURE  — Offer Shell (tiers of access)
      |
      v
Step 2: CRAFT      — Offer Map (Promise/Path/Pricing/Why Now)
      |
      v
Step 3: POSITION   — Protagonist (Goal/Steps/Rules)
      |
      v
Step 4: SELL       — 90 Days of Offers (Red Velvet + daily rhythm)
      |
      v
Step 5: ANTI-SLOP  — mandatory pass
      |
      v
[Final offer package]

You can enter at any step depending on the user's brief. Most refinements start at Step 2 (Offer Map). Brand-new offers start at Step 1 (Shell). Personal-brand work starts at Step 3 (Protagonist). Promotion calendars start at Step 4 (90 Days).

structure: build the offer shell

Instruction: Decide which tiers serve the user at this stage. Everyone gets the same STUFF — the difference is how much ACCESS they get on top of it. Sell access to YOU, the most desirable thing you have.

Framework — the 4 tiers:

Level	What It Is	Example Pricing
One-Time	Workshop/Asset only	$100-500
Temporary	Limited time access	$500-3000
Ongoing	Continuous access	$100-400/week
Premium	1:1 access	2x your group rate

The Money Line: Workshop -> Temporary Access -> Ongoing -> Premium/1:1. The further down the line, more money and LESS selling. Stats to know: 80% of clients come from a workshop, 25% of workshop buyers become clients, 5% become 1:1.

Short example: A $99 Claude Code workshop becomes One-Time. Add a $297 tier with two 1:1 calls = Temporary. Add a $497/month implementation cohort = Ongoing. Add $2k/month private coaching = Premium. Same skill, four levels of access.

Anti-pattern: Building a "Suite" of unrelated products (course + mastermind + community + program). Improvements in one don't bleed to others, attention splits, you stay miserable. Build ONE shell, not four products.

For the full framework (Front End cycling, Back End upsells, credit-forward language, life-based offers) see references/offer-shell.md.

craft: write the offer map

Instruction: Write the offer copy in 4 parts — Promise, Path, Pricing, Why Now. Good copy isn't written, it's assembled.

Framework:

1. PROMISE — One promise only. Must pass The Meaning Test (show two people, do they read it the same way?). Mundane > elevated. Outcome OR number.

   • Fails: "Transform your physique" / Passes: "See your six pack"
   • Fails: "Become a productivity master" / Passes: "Finish your workday by noon"
   • If intangible, use the "so you can" bridge: "Install the 3 AI Workflows so you can make more than last quarter."

2. PATH — Show HOW. 3 stages, each with Benefits (what they GET, not what they DO) + Checkpoint (straight line to Promise) + Timeframe.

   • Bad checkpoints (disconnected): Belief work -> Strategy -> Implementation
   • Good checkpoints (straight line to "Lose 30 lbs"): Lose first 10 -> next 10 -> final 10

3. PRICING — Weekly beats monthly. $100-200/week if early, $250-400/week if you can command it. Half your 1:1 rate if going group, double your group rate if going 1:1.

4. WHY NOW — Without urgency, people don't move. Pick scarcity (capped spots), urgency (deadline, price raise), or both.

Short example (for a $99 Claude Code workshop):

• Promise: "Have Claude Code installed and producing your first marketing asset by end of the 90-minute session."
• Path: Setup (15 min, working install) -> Skill files (30 min, plug-and-play config) -> First asset (45 min, one piece of content shipped)
• Pricing: $99 one-time
• Why Now: Doors close Friday, next cohort in 6 weeks, price goes to $149

Anti-pattern: Two promises in one offer ("Lose weight AND make money"). Halves effectiveness. Pick one. Or worse — using elevated content language ("transform your life") in the offer copy. Content = elevated, OFFER = mundane.

For the full framework (Meaning Test, Coaching/Consulting/Mentorship frames, Mundanity Rule, Promise Hierarchy, Path/Timeframe rules, all Why-Now tactics) see references/offer-map.md.

position: become the protagonist

Instruction: Write the user as a protagonist with a goal, steps, and rules — not a coach with credentials. People fall in love with characters, not outcomes. The "aura" comes from being willing to go first, not from special talent.

Framework — the Protagonist Post (3 parts, pinned for the year):

1. Your Goal — Must genuinely excite YOU. Money ("$100k/month profit"), Impact ("help 1000 people get a six pack"), Capture ("$5M/month for clients, $50k for me"), or Build ("Proyecto Casa"). The bravery to declare it matters more than the goal itself.

2. Your Steps — How you'll get there. This naturally promotes the Offer Shell and conditions the audience: they were warned they'd be sold to, so they can't complain (you can't get mad about being punched in a boxing gym).

3. Your Rules — What you will and won't do. Batman doesn't kill — that's why we feel things about him. Promises attract low-income people, philosophy attracts high-income people.

Short example (for the Claude Code workshop creator):

• Goal: "Help 1,000 non-technical marketers ship their first Claude Code asset in 2026."
• Steps: Run the $99 workshop monthly, open a 4-week cohort quarterly, document every win publicly.
• Rules: No webinars-while-eating-dinner. No fluff. If you don't ship something live in the session, you get a refund.

Anti-pattern: "Coach mode" — always available, always nice, exists in circles, reacts to the market, no declared goal. Audience says "congrats" instead of "what goal?" and forgets you next week.

For the full framework (Netflix content strategy, Big/Small Updates, Minigames, the 5 Conditioning Tactics, Percentage Hits, public/private consequences) see references/protagonist.md.

sell: run 90 days of offers

Instruction: Set up a daily selling rhythm using Red Velvet (small change = looks new), insight-based emails (90%), and sales emails (10%). Email is the primary channel because people subscribe to solve problems.

Framework — the daily rhythm:

Cadence	Action
Every Day	1 Front End offer (email + social), Netflix-style update
Every Week	1 Back End offer to current clients
Every 2 Weeks Max	Red Velvet the offer (change something), cycle types
Every Month	Open/close an intake, Protagonist progress update
Every Quarter	Raise prices, Big Protagonist Update, launch a Minigame

Email mix: 90% insight-based (problem / outcome / idea / leadership), 10% sales (hand raisers, urgency, "Last Day:" subject lines). Insight attracts people who want to BUY. Value attracts people who want to TRY.

Red Velvet menu (rotate at least one every 2 weeks): limited spots, 1-on-1 add-on, bonus added, bonus removed, DFY work included, price raise scheduled, doors closing date, seasonal hook, "never available again."

Short example: Week 1-2 promote $99 workshop with "doors close Friday." Week 3-4 promote same workshop with new bonus (n8n automation pack). Week 5-6 promote the $297 tier (workshop + 2 calls). Week 7-8 back to workshop with price-raise warning. Same offer, four red-velvet variations, eight weeks of content.

Anti-pattern: Promoting the same offer the same way for 4+ weeks straight. Audience tunes out, opens drop, you start blaming the algorithm. Or the inverse — building a brand-new offer every week instead of red-velveting the one you have.

For the full framework (insight buckets, top-performing subject lines, recycling rhythm, where-to-sell rules, the 90-offer rotation plan) see references/90-days-offers.md.

anti-slop: mandatory pass

Before showing any offer copy to the user, run it through the anti-slop skill.

How: Invoke anti-slop with the full draft (offer name, positioning statement, sales bullets, pricing description, CTA) as input.

If anti-slop is not available, scan manually for: rhetorical questions ("Want to know..."), binary opposites ("it's not X, it's Y"), robot phrases (game-changer, supercharge, leverage, unlock, delve), throat-clearing, em dash overuse, zero contractions.

Rewrite anything flagged. Non-negotiable: no offer copy ships without this pass.

output format

Deliver the offer in this order so the user can copy-paste straight into their sales doc:

# [Offer Name]
[One-line tagline / promise statement]

## Promise
[One sentence — the outcome in the customer's words. Passes Meaning Test. Mundane.]

## Tiers
| Tier | Price | What's Included | Who It's For |
|------|-------|-----------------|--------------|
| [Name] | [$X] | [bullets] | [persona] |
| [Name] | [$X] | [bullets] | [persona] |

## What You Get
- [Outcome bullet 1 — what they GET, not what they DO]
- [Outcome bullet 2]
- [Outcome bullet 3]
- [3-7 total]

## Why Now
[Deadline, scarcity, or urgency hook in 1-2 sentences]

## CTA
[Button line — verb + outcome]
[Supporting line — risk reversal, deadline reminder, or social proof]

After the draft, list which framework references you pulled from (Shell / Map / Protagonist / 90 Days) and one suggestion for what to refine next session.

example: $99 Claude Code workshop offer

This is the worked example using the user's actual workshop, drawn from CLAUDE.md context.

Brief confirmed: $99 live workshop teaching non-technical marketers to set up Claude Code, with a $297 upsell tier (workshop + 2x 1:1) and a possible 4-week cohort upsell. Audience: marketers on the 350-person waitlist who feel FOMO and overwhelm. Voice: direct, casual, no fluff, sounds like Ole.

Step 1 — Offer Shell (structure)

Tier	Level	Price
Workshop only	One-Time	$99
Workshop + 2x 1hr 1:1	Temporary	$297
4-Week Implementation Cohort	Ongoing	$497-997 (TBD)
Private 1:1 (Ole's calendar)	Premium	$2k+/month

Money Line: Workshop -> Cohort -> Private 1:1. 80% of cohort buyers should come from the workshop. 5% of workshop buyers become 1:1.

Step 2 — Offer Map (copy)

Offer name: Claude Code for Marketers (Live Workshop)

Promise (passes Meaning Test): "Have Claude Code installed, configured for your marketing work, and producing your first asset by the end of a 90-minute live session — even if you've never touched code."

Path (3 stages, straight line to Promise):

1. Setup (0-15 min) — Working Claude Code install on your machine. Checkpoint: it runs.
2. Configure (15-45 min) — My personal skill files plugged in for content, ICP, voice. Checkpoint: Claude knows your business.
3. Ship (45-90 min) — Produce one real marketing asset live. Checkpoint: you have something to post tomorrow.

Pricing: $99 workshop / $297 with two 1:1 calls.

Why Now: Doors close Friday at midnight. Next cohort isn't for 6 weeks. Price goes to $149 after this run.

Step 3 — Protagonist Positioning

Goal: Help 1,000 non-technical marketers ship their first Claude Code asset in 2026.

Steps: Monthly $99 workshop, quarterly cohort, document every win and L on X and email.

Rules: No webinars-while-eating-dinner. No fluff intros. If you don't ship something live in the session, full refund.

Step 4 — 90-Day Sell Plan (Red Velvet rotation)

• Weeks 1-2: Promote workshop + "doors close Friday" urgency
• Weeks 3-4: Same workshop, add bonus (n8n automation pack)
• Weeks 5-6: Promote $297 tier (workshop + 2 calls), 5 spots only
• Weeks 7-8: Back to $99 workshop, "price goes to $149 next month"
• Weeks 9-10: Cohort waitlist open
• Weeks 11-12: Recycle best workshop email + open next cohort

Daily emails: 90% insight-based (e.g., "The skill file that 10x'd my newsletter," "I tried Claude Code and got stuck. Here's what fixed it"), 10% sales ("Last Day: Workshop closes tonight").

Final Output Block (copy-pasteable)

# Claude Code for Marketers — Live Workshop
Get Claude Code set up and shipping marketing work in 90 minutes.

## Promise
Have Claude Code installed, configured for your marketing, and producing your first
asset by the end of a 90-minute live session — even if you've never touched code.

## Tiers
| Tier | Price | What's Included | Who It's For |
|------|-------|-----------------|--------------|
| Workshop | $99 | 90-min live session, my skill files, n8n automation pack | Marketers on the waitlist who tried Claude Code and got stuck |
| Workshop + 1:1 | $297 | Everything above + 2x 1hr 1:1 calls with me | People who want a custom setup for their specific stack |

## What You Get
- Working Claude Code install on your machine in 15 minutes
- My personal skill files for content, ICP research, and voice — plug-and-play
- One marketing asset shipped live during the session (newsletter, post, or page)
- The n8n automation pack ($99 value) — bonus
- Replay access if you can't make it live

## Why Now
Doors close Friday at midnight. Next workshop runs in 6 weeks. Price goes to $149
after this run. 50 spots, first-come.

## CTA
Save your seat — $99
Doors close Friday. Refund if you don't ship something live.

References pulled from: offer-shell.md (Money Line + tiers), offer-map.md (Promise/Path/Why Now), protagonist.md (Goal/Steps/Rules), 90-days-offers.md (Red Velvet rotation).

Next session suggestion: write the 12-email launch sequence using the insight buckets in references/90-days-offers.md.

quick reference: the daily rhythm

Every Day:

• Make 1 Front End offer (email + social)
• Post Netflix-style content

Every Week:

• Make 1 Back End offer to current clients

Every 2 Weeks Max:

• Red Velvet your offer (change something)

Every Month:

• Open/close an intake
• Update Protagonist progress

how to use this skill

User Wants To...	Start With...
Create/refine an offer	Offer Map (Promise first)
Build offer tiers	Offer Shell
Position/brand themselves	Protagonist
Sell more often	90 Days of Offers
Write sales emails	90 Days (insight-based emails)
Plan a launch	Red Velvet + Protagonist

how this connects to other skills

Feeds in (run these first if available):

• secondbrain:setup — populates Context/business.md, icp.md, voice.md before this skill runs
• vibe2-positioning-angles — finds the angle before you commit to a Promise
• entrevista-negocio — fills business / ICP / voice context (into CLAUDE.md or Context/*.md) if it's empty

Feeds out (use these after to ship the offer):

• landing-page — turns the offer package into a full sales page
• sequencia-email — builds the 12-email launch sequence around the Why Now hook
• escrever-newsletter — writes the daily insight emails for the 90-day rotation
• post-longo-x / reaproveitar-conteudo — turns the Protagonist Post into social content
• anti-slop — MANDATORY pass on every output before delivery

the test

Before delivering the offer package, verify each:

1. Context check: Did I pull business / ICP / voice from CLAUDE.md or Context/*.md before writing? (If anything critical was missing, did I flag it to the user?)
2. Promise meaning test: Would two different people read the Promise the same way? Is it mundane (outcome or number), not elevated?
3. Path checkpoints: Do all 3 stages connect in a straight line to the Promise, or are they disconnected (belief work -> strategy -> implementation)?
4. Tiers sell access, not stuff: Does the price difference between tiers reflect access to the user, not just more deliverables?
5. Why Now is specific: Is there a real deadline, price raise, or spot cap? "Limited time" alone doesn't count.
6. Protagonist voice: Does the copy sound like the user has a goal, steps, and rules — or like a generic coach who's "always there"?
7. Anti-slop pass: Did the draft go through the anti-slop skill (or manual scan) and get cleaned of robot phrases, rhetorical questions, em dashes, throat-clearing?

Failure condition: If any check fails — especially #1, #2, or #7 — DO NOT deliver. Fix the failure and re-run the test. Shipping a generic-promise, slop-laden offer is worse than asking the user one more clarifying question.

source documents

These frameworks synthesized from:

• 90 Days Of Offers Swipe File
• Google Doc 314
• The Offer Shell
• Protagonist Workshop

Feedback Log — progressive learning

This skill gets sharper over time. Every correction the user gives becomes a permanent rule in feedback.log next to this SKILL.md.

At the start of every session

1. Read feedback.log in this skill's folder. Apply every rule to this session's output.
2. If it doesn't exist, create it with header: # Offer Architect Workshop — Learned Rules

Triggers that require a log write

When the user says any of the following, immediately append a rule before continuing:

• "Don't do X anymore" / "Stop doing X" / "Never do X"
• "Always do Y" / "From now on, do Y"
• "I hate when you..." / "I don't like..."
• "That's perfect, keep doing that" (log positive patterns — don't drift)
• Silent rewrites where the user fixes your draft — infer the rule from the diff

Format for each rule

## YYYY-MM-DD — [short title]
**Rule:** [imperative sentence]
**Why:** [reason]
**How to apply:** [when this kicks in — which offer type, which section]

What to log vs skip

Log: pricing rules, positioning preferences, structural choices, voice calibrations. Skip: one-off task fixes, preferences already in Context/voice.md.

Cleanup every ~10 entries. Consolidate, delete reversed rules, keep under 100 lines.