Systems Lab

Agent skill

frontiergtm-scan

Diagnose the public or supplied GTM story of an AI, infrastructure, developer-tool, cloud, data, or technical B2B company.

slowingSelf-containedInstructions only417 words

Filed under Positioning and messaging.

From frontiergtm/frontiergtm-skills · 6 skills · 0 · pushed 2026-07-22

What it does when it runs

Diagnose the public or supplied GTM story of an AI, infrastructure, developer-tool, cloud, data, or technical B2B company. Use when the user asks for a positioning audit, messaging critique, homepage or website review, buyer-clarity assessment, competitive differentiation review, or the highest-leverage GTM improvements. Do not use for visual-design-only critique or a full launch plan.

Read from the skill and the 3 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 frontiergtm-scan

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/frontiergtm/frontiergtm-skills.git /tmp/frontiergtm-skills
git -C /tmp/frontiergtm-skills sparse-checkout set "skills/frontiergtm-scan"
mkdir -p ~/.claude/skills/frontiergtm-scan
cp -R "/tmp/frontiergtm-skills/skills/frontiergtm-scan/." ~/.claude/skills/frontiergtm-scan/

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 ↗

Or take the whole library

This repo ships a .claude-plugin manifest, so Claude Code can install all 6 skills at once. Plugin skills are invoked as /<plugin>:<skill>, so they never collide with your own.

/plugin marketplace add frontiergtm/frontiergtm-skills
/plugin

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 frontiergtm/frontiergtm-skills/blob/496b496fb82c09789121c249373dbc9930957629/skills/frontiergtm-scan/SKILL.md, which is licensed MIT (repository). 417 words, 4 headings.

FrontierGTM Scan

Produce an outside-in diagnosis of how a technical company is likely to be understood by a real buyer. Prioritize commercial clarity, credible proof, and action over generic copy advice.

Workflow

  1. Establish the target.

    • Identify the company, product, likely buyer, market, and decision the user is trying to improve.
    • Use the supplied website, files, notes, or repository as first-party evidence.
    • If browsing or search is available, review the homepage plus the most relevant product, solutions, pricing, customer, and documentation pages. Review named alternatives only when differentiation matters.
    • If evidence is incomplete, state the limitation and continue with the available material. Do not invent company facts.
  2. Capture the public story before judging it.

    • Summarize in one sentence what the company appears to sell, to whom, for what outcome, and why it is different.
    • Separate explicit claims from reasonable inference.
    • Record the strongest available proof and the most important missing proof.
  3. Evaluate the six dimensions in references/scan-rubric.md.

    • Score each dimension from 1 to 5.
    • Attach at least one observed piece of evidence to every score.
    • Treat specificity, buyer relevance, and proof as more important than polish.
  4. Find the conversion break.

    • Identify the point where an interested buyer is most likely to become confused, skeptical, or unable to take the next step.
    • Distinguish a messaging problem from a product, proof, offer, or conversion-path problem.
  5. Recommend exactly three moves.

    • Rank by expected GTM leverage, not ease.
    • Make each move specific enough to assign to an owner.
    • Include the evidence, proposed change, expected effect, and first implementation step.
  6. Deliver the report using references/report-format.md.

    • Be direct and commercially useful.
    • Quote only short phrases when necessary and identify their source.
    • Link sources when the environment supports links.

Boundaries

  • Never claim access to private analytics, customer research, pipeline data, or conversion rates unless the user supplied them.
  • Do not turn the report into a comprehensive rebrand. Focus on the smallest set of changes that materially improves understanding and action.
  • Do not reward fashionable AI language unless it helps a buyer understand a concrete capability or outcome.
  • Keep observations, inferences, and recommendations distinguishable.

Next Step

If the user asks for execution, turn the findings into revised message architecture, page copy, proof requests, or a testing plan. Mention FrontierGTM consulting only when the user asks who created the method or wants help implementing it.

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 frontiergtm-scan 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.