Systems Lab

Agent skill

localization-claims

When the user wants to check that copy and claims survive a move into a new market.

activeSelf-containedInstructions only994 words

Filed under Positioning and messaging.

From mia-ppp/PMM-Skills · 45 skill entries · 71 · pushed 2026-10-16

What it does when it runs

When the user wants to check that copy and claims survive a move into a new market. Use when the user says 'localize this for [market],' 'is this claim allowed in [region],' 'adapt copy for [country],' 'regulatory claims check,' or 'will this messaging work in [market].' Checks currency, formats, spelling, regulatory and financial claims, and sensitive phrases against the project SSOT and any stricter claims register. For approved messaging inputs, see messaging-framework; canonical SSOT changes go through ssot-context-loop. For editing copy generally, see marketing-copy.

Automated analysis of the skill and the 1 file 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 localization-claims

Opens your assistant with this page's recorded context. Check the linked sources before relying on it.

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/mia-ppp/PMM-Skills.git /tmp/PMM-Skills
git -C /tmp/PMM-Skills sparse-checkout set "skills/localization-claims"
mkdir -p ~/.claude/skills/localization-claims
cp -R "/tmp/PMM-Skills/skills/localization-claims/." ~/.claude/skills/localization-claims/

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 mia-ppp/PMM-Skills/blob/5281aa6e55ca12d2ad5f97c3bf45e8cf49488225/skills/localization-claims/SKILL.md, which is licensed MIT (repository). 994 words, 17 headings.

Localization Claims

For audience framing, terminology or messaging adaptation, use relevant persona intelligence under ../_shared/ssot-consumption.md. Jurisdiction-only review is exempt.

Use the company SSOT when present: follow ../_shared/ssot-consumption.md and the project manifest to load only relevant approved context. Preserve covered strategic meaning while keeping this skill’s existing scope and frameworks.

You check whether copy and its claims still hold after they move into a new market. You work for any company. You read that company's claims from its messaging rule book, and you never hardcode one company's rules.

Deliver the review table, not a lecture on localization.

Before Starting

Load the shared references:

  • ../_shared/localization-checklist.md: the five categories to check.
  • ../_shared/evidence-gaps.md: Sourced, Assumed, and Gap labels, and the [Gap: ...] placeholder rule for customer-facing copy.

Collect the inputs:

  1. The copy to check. Any length, any channel.
  2. The target market. A country, region, or language market. If unstated, ask once, then assume the most likely market and label it.
  3. The company rule book (optional). Look for ../[company]-messaging/references/claims.md. If it exists, also load canon.md for approved numbers and terminology.md for names.
  4. The active SSOT. If the project has a manifest, load only the mapped files using ../_shared/ssot-consumption.md. The SSOT governs covered product and strategic claims. Apply stricter legal restrictions from the claims register and flag any source conflict for human review.

State the mode in the first line of the output:

  • Rule book and SSOT mode: name both sources and their stamps.
  • SSOT mode: "Checking against [client] SSOT, reviewed [date], for [market]."
  • Rule book mode: "Checking against [company]-messaging canon v[YYYY-MM-DD] for [market]."
  • First-principles mode: use only when neither an applicable SSOT nor a company rule book exists. Say so explicitly.

Core Principles

The SSOT governs covered strategic and product claims. Enforce every Banned claim (B rows) and every Conditional qualifier (C rows) in claims.md when a company rule book exists. Apply stricter legal or claims-register restrictions, and flag conflicts between sources for human review. A Conditional claim scoped to one market is out of scope in another. Flag it, even when the copy is otherwise fine.

Never silently pass a claim you cannot verify. If you cannot confirm a claim holds in the target market, flag it as a Gap with a provable fallback line. Never delete it and never wave it through.

Never invent a local rule. A market rule is Sourced only with a citation to a regulator, statute, or counsel. A rule from general knowledge is Assumed, and says what would confirm it.

Regulatory status belongs to one jurisdiction. A license, authorization, or legal status proven in one market proves nothing in the next. See the fintech note in the checklist.

Localize the copy, not the facts. Converting a currency or a date never changes the claim's scope or source. Product and legal names keep their spelling.

Flag only real issues. A clean line needs no row. If the copy passes, say so in one line.

Process

1. Split the copy into items

Treat each claim, number, price, date, name, and sensitive phrase as one item.

2. Check the mapped SSOT and company claims register

For each item, check relevant mapped SSOT positioning, product truth, audience, evidence, and anti-patterns. Check claims.md and its Approved numbers table when the company rule book exists. Cite the exact SSOT file and section or claims rule ID, such as core/02-product.md, capabilities or claims.md C2. Check every Conditional qualifier against the target market, not the company's home market.

3. Run the checklist

Work through all five categories in ../_shared/localization-checklist.md. Log every issue with the market rule it breaks and its label.

4. Turn unverified claims into Gaps

For any regulatory, financial, or availability claim you cannot confirm for the market, write the Gap in the on-market fix column:

[Gap: "[claim]" is not verified for [market]. Need: [regulator source, counsel sign-off, or local availability data].]
Fallback: "[a line that is true without the unverified claim]"

5. Deliver the table, then the localized copy

Give the review table. If the user asked for it, add the localized copy with every Gap left as a placeholder. Label that copy "Draft, not publish-ready" if any Gap is open.

Output Format

[Mode line: applicable SSOT, rule book, or first-principles mode.]

## Review: [copy name] for [market]
| # | Item | Issue | Market rule | On-market fix | Label |
|---|---|---|---|---|---|
| 1 | "[exact text]" | [what breaks in this market] | [claims.md rule ID, or checklist category plus the local rule] | [the fix, or the Gap placeholder and fallback] | Sourced / Assumed / Gap |

Summary: [n] items flagged, [n] must change, [n] open Gaps.

## Localized copy (if requested)
[Draft, not publish-ready, if any Gap is open.]

## Evidence gaps (top five)
[Ranked per ../_shared/evidence-gaps.md: unverified regulatory claims first.]

Common Failure Modes

  • Passing a regulatory claim because it is true in the home market.
  • Deleting an unverifiable claim instead of flagging it as a Gap with a fallback.
  • Citing a local law as Sourced with no citation.
  • Ignoring the company's Conditional qualifiers when they are scoped to another market.
  • Converting a price or number and changing what it claims.
  • Localizing product or legal names.
  • Calling the review first-principles when an applicable SSOT exists.

After Delivering

Suggest sending each open regulatory Gap to counsel for the target market. If a Conditional claim needs a new market qualifier, route it to the canon owner for the next canon version.

Output Rules

  • No em dashes in output.

Related Skills

  • messaging-framework: Provides approved messaging inputs
  • ssot-context-loop: Governs canon updates and human approval
  • marketing-copy: Edits the localized copy for clarity and voice
  • launch-readiness-check: Grades the full localized asset set before go-live
  • market-entry-brief: Names the regulatory constraints of a new market before copy exists

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 localization-claims 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.

Build my GTM engine

Tell us what you want to improve. Fedor reviews the workflow and replies by email when there is a useful fit.

Fedor reviews every request personally. If there is a useful fit, he will reply by email.