How to use the International Expansion Strategy agent skill

What is an agent skill and what does this one do?

An agent skill is a folder of instructions that an AI agent loads when a request matches the skill's description. This one lives in the c-level-advisor section of the alirezarezvani/claude-skills repository and is licensed MIT.

Once installed, the agent follows a fixed decision sequence when you ask about entering a foreign market: market selection, entry mode, regulatory assessment, localization plan, go-to-market strategy, team structure, then launch.

• Market selection: scores candidate markets on addressable market size (25%), competitive intensity (20%), regulatory complexity (20%), cultural distance (15%), existing traction (10%) and operational complexity (10%).
• Entry mode: compares export, partnership, local team, entity (full subsidiary) and acquisition on investment, control, risk and best use.
• Localization and operations: runs a checklist covering product, go-to-market and operations items.
• Regional notes: points to reference material for the EU, US, APAC and LATAM.

Install the intl-expansion skill

Download the ZIP from this page. Unzip it into one of two locations, depending on how you want to use it:

For all your projects, put the unzipped skill folder in ~/.claude/skills/. For a single project, put it in that project's .claude/skills/ folder.

Alternatively, in the Claude apps go to Settings → Capabilities → Skills and upload the skill there.

• Personal install: ~/.claude/skills/
• Project install: the .claude/skills/ folder inside your project
• Claude apps: Settings → Capabilities → Skills, then upload

Confirm what the skill needs before you install

The listing states that the skill requires nothing beyond the agent itself, runs no scripts and uses no network connection. It is made up of plain instruction and reference files: SKILL.md, references/market-entry-playbook.md and references/regional-guide.md.

Because it runs nothing and makes no network calls, it cannot fetch live market data. Everything it produces comes from the agent's own reasoning plus the frameworks in these files.

How do you trigger the skill with a natural request?

You do not need a special command. Describe the decision you are making in ordinary language. The skill's description covers requests about expanding to new countries, evaluating international markets, planning localization, and building regional teams, so phrase your request around one of those situations.

Give the agent your context: your product, your current market, your size, and what you already know about the target country. The framework works on specifics, so vague requests produce generic output.

• "We sell B2B invoicing software in the UK. Should we enter Germany in the next year?"
• "Help me plan the localization work needed before we open a Brazilian sales team."
• "Score Japan, Singapore and Australia for our expansion, using market size, competition and regulation."
• "Should we hire a local team in France or sell remotely first?"
• "We want to build a regional team in Southeast Asia. What structure should we consider?"

What output should you expect?

Expect the agent to walk through the framework rather than hand you a single answer. For a market-selection request it should rate each factor, explain the reasoning, and recommend an entry mode. The skill's default path is Export → Partnership → Local team → Entity, moving to the next stage as revenue justifies it.

The skill also lists common mistakes it tries to guard against, such as entering too many markets at once (its suggested limit is one to two new markets per year), copying go-to-market from the home market, and simply converting prices instead of researching local willingness to pay.

• Ask the agent to fill in the scoring matrix for each candidate market and explain every score.
• Ask it to compare two entry modes against your budget and risk tolerance.
• Ask for the localization checklist, then mark which items apply to your product.
• Ask the agent to list the questions to answer with a regulator or counsel, rather than the answers themselves.

Safety notes: what the skill does not do

The skill gives general checklists. It is not legal, tax or employment advice, and rules change by jurisdiction. Its own guidance for regulatory points is to check the requirement before committing, so treat any compliance item the agent raises as a question to verify with qualified counsel.

The skill contains no data sources. The agent may produce market size or competitor figures that sound plausible but are unverified. Check every number against a primary source before it goes into a board deck or budget.

• Do not rely on it for legal, tax or employment decisions.
• Do not treat its market size or competitor numbers as current facts.
• Do not use its output as a verified entry plan without expert review.

Tips for better results

Feed the agent real evidence. The playbook ranks signals of demand, from strong ones like inbound inquiries and existing customers using your product from that market, down to weak ones like market research saying a country is big. Tell the agent which signals you actually have.

Use the validation steps before the framework's heavy commitments. The playbook suggests a localized landing page with a waitlist, a small targeted ad test, around 20 sales calls in the market, and conversations with three to five potential local partners.

• Name your current market, company size and product type in the first message.
• Separate what you know from what you are guessing, so the agent flags the guesses.
• Ask for the regional notes that match your target (EU, US, APAC or LATAM).
• Re-run the scoring with different weights if your priorities differ from the defaults.