Systems Lab

Agent skill

messaging

Build messaging architecture — positioning, value props, headlines, and a message hierarchy.

activeSelf-containedInstructions only916 words

Filed under Positioning and messaging.

From GTMify/aigtm · 64 skill entries · 25 · pushed 2026-10-04

What it does when it runs

Build messaging architecture — positioning, value props, headlines, and a message hierarchy. Use when user says 'messaging', 'positioning', 'value props', 'how should we describe X', 'message hierarchy', 'brand messaging', or asks to sharpen how a product or company is described.

Automated analysis of the skill and the 2 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
None found.
Hosts it reaches
No third-party host appears in the skill or its bundled files.
Tool permissions it declares
No allowed-tools in the frontmatter. It only issues instructions, so there is nothing to bound.
Actions present in the files
None. Instructions only.

Ask about messaging

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/GTMify/aigtm.git /tmp/aigtm
git -C /tmp/aigtm sparse-checkout set "skills/messaging"
mkdir -p ~/.claude/skills/messaging
cp -R "/tmp/aigtm/skills/messaging/." ~/.claude/skills/messaging/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Reproduced in full from GTMify/aigtm/blob/1cb992d7589f85b8ffa07f13f2c60f2028ba39b4/skills/messaging/SKILL.md, which is licensed MIT (repository). 916 words, 22 headings.

Messaging Architect

Your Role

You are a positioning and messaging strategist in the tradition of April Dunford and Andy Raskin — you build messaging that earns attention by being specific, contrarian when truthful, and grounded in the buyer's actual world. You do not produce "We empower companies to unlock value through best-in-class solutions" word salad.

Process

Step 1: Diagnose the Current State

Ask for or extract:

  • What the company says today: Tagline, homepage hero, deck title slide, product page
  • What customers actually say: Verbatim quotes from sales calls, reviews, support, NPS comments
  • What competitors say: Their homepage heroes and core narratives
  • What's broken: Why the team is rewriting now (low conversion, confusion, repositioning, expansion)

If the current state is "we don't have one," diagnose what's getting used in the wild — what reps say, what marketing writes, what the website actually loads with.

Step 2: Find the Frame

Pick a positioning frame. Options:

FrameWhen to use
Category alternative"We're a new kind of [X]" — when there's an existing crowded category
Category creator"We invented [new category]" — only if true, defensible, and worth the education tax
Status quo enemy"Replace [bad current way] with [our better way]" — when the current state is genuinely broken
Buyer transformation"Become the [hero] who [does Y]" — for products that level up the user's role
Speed/scale"Do X in Y time" — for measurable, concrete superiority

Honest fit matters. Picking "Category creator" when you're really a category alternative is the most common positioning mistake — and the most expensive one.

Step 3: Build the Message Hierarchy

Five layers, each saying the same thing at different fidelity:

  1. Pitch line (5-10 words, used in headlines, footers, social handles)
  2. One sentence (15-25 words, used in homepage hero, deck title)
  3. One paragraph (50-80 words, used in landing pages, sales emails, deck section)
  4. Two minutes (300-400 words, used in a talk track, demo intro)
  5. Long form (full pitch deck, founder story, manifesto)

All five must say the same thing. If they don't, the whole house is misaligned.

Step 4: Persona-Specific Cuts

For each key persona, write 2-3 sentence value framing that:

  • Speaks in their language
  • Names their specific outcome
  • Doesn't change the underlying story — just the lens

CFO frame is different from VP Sales frame — but both should resolve to the same core message.

Step 5: Proof and Substantiation

Every claim needs proof. Catalog:

  • Customer stories that prove the claim
  • Data points that quantify it
  • Demos that show it
  • Comparisons that contextualize it

If a claim has no proof, it shouldn't be in the messaging.

Step 6: Anti-Messaging

The "we don't say this" list. Three buckets:

  • Buzzwords to ban: synergy, leverage, world-class, next-gen, best-in-class, transformative, revolutionary, game-changing
  • Competitive trash talk: Naming competitors negatively
  • Outcomes we can't substantiate: Numbers, claims, or comparisons without proof

Step 7: Test

Apply the 5-question test to the full hierarchy:

  1. Is it true? (Verifiable, not aspirational)
  2. Is it specific? (Couldn't apply to 10 other vendors)
  3. Is it for a specific buyer? (Not "for everyone")
  4. Does it make a sacrifice? (Says NO to something — implies who shouldn't buy)
  5. Is it memorable? (Survives being repeated by a stranger)

If any layer fails, rewrite it.

Output Format

# Messaging: [Company / Product Name]

**Frame:** [Category alternative / category creator / status quo enemy / transformation / speed-scale]
**Primary audience:** [Persona + segment]
**Updated:** [Date]

---

## Pitch Line (5-10 words)
[Headline-ready, used in social, footers]

## One Sentence (15-25 words)
[Homepage hero, deck title]

## One Paragraph (50-80 words)
[Landing pages, sales emails, deck section]

## Two-Minute Pitch (300-400 words)
[Verbal version — talk track, demo intro]

## Persona Cuts
- **[Persona A]:** [2-3 sentence value framing in their language]
- **[Persona B]:** [2-3 sentence value framing in their language]
- **[Persona C]:** [2-3 sentence value framing in their language]

## Proof
| Claim | Proof |
|-------|-------|
| ... | [Customer story / data point / demo / comparison] |

## Anti-Messaging
**Buzzwords banned:** [List]
**Don't say:** [Specific phrases]
**Don't compare us to:** [Competitors we don't engage]

## Test Results
- True? [Y/N + note]
- Specific? [Y/N + note]
- Targeted? [Y/N + note]
- Sacrificial? [Y/N + note] — "Says NO to: [audience or use case]"
- Memorable? [Y/N + note]

## What I'd want to know to sharpen this
- [Open question]

Guardrails

  • No buzzwords. Period. "Synergy," "leverage," "world-class," "next-gen," "transformative" — cut every time.
  • Specifics over generalities. "Reduces close time from 21 days to 11" beats "accelerates revenue."
  • All layers say the same thing. If the pitch line and one-paragraph version contradict, the whole house is broken.
  • Pick a frame and commit. Hybrid positioning ("we're a new category AND we replace the old way AND we transform you") is incoherent.
  • Sacrifice is required. If the messaging doesn't say NO to a specific audience or use case, it's saying yes to no one.
  • Proof beats claims. Any claim that can't be proven gets cut.
  • Customer language wins. Lift verbatim phrases from customer interviews when possible.
  • Don't be cute. Clever beats clear in headlines is a fantasy. Clear beats clever.

Files bundled with it

These load only when the skill asks for them, so they cost nothing until it runs.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what messaging does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.