Cold Storage Geographic Distribution Strategy logo

Cold Storage Geographic Distribution Strategy

Community
HimanshuJ16
cold-storage-geographic-distribution-strategy

Use when auditing how M-of-N key shards are spread across facilities, countries and legal regimes, flagging both single points of failure and quorum concentrations. Generating the shards is shamir-secret-sharing-for-key-backup.

Overview

PublisherHimanshuJ16
RepositoryAlgo-Trading-Skills
Skill namecold-storage-geographic-distribution-strategy
Stars
50
Forks
7
Bundled files
5
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by HimanshuJ16 on GitHub. Read the source before you install it.

Installation

Install the Cold Storage Geographic Distribution Strategy AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/HimanshuJ16/Algo-Trading-Skills/tree/main/skills/cold-storage-geographic-distribution-strategy
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/HimanshuJ16/Algo-Trading-Skills.git /tmp/Algo-Trading-Skills
mkdir -p .claude/skills
cp -r /tmp/Algo-Trading-Skills/skills/cold-storage-geographic-distribution-strategy .claude/skills/cold-storage-geographic-distribution-strategy
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Cold Storage Geographic Distribution Strategy in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Cold Storage Geographic Distribution Strategy on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Cold Storage Geographic Distribution Strategy is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

When to Use

Use this skill when designing or auditing cold storage security architecture for institutional crypto assets held under an $M$-of-$N$ Shamir Secret Sharing scheme. Concentrating shards in one facility, one country, one legal regime, or one commercial custodian creates a Single Point of Failure in both directions:

  • Confidentiality SPOF - one group holds $\ge M$ shards and can reconstruct the key alone (seizure order, raid, insider collusion).
  • Availability SPOF - one group holds $> N - M$ shards and, by becoming unavailable, leaves fewer than $M$ shards reachable, permanently destroying access to the assets.

When NOT to Use

Do not use this skill to split or reconstruct secrets - it never touches key material and performs no cryptography. Shard generation belongs to shamir-secret-sharing-for-key-backup. It is also not a substitute for legal advice on where assets may lawfully be held: it audits a placement you supply, and no regulator surveyed prescribes a shard geography (see references/standards.md).

Prerequisites

  • $M$-of-$N$ threshold parameters (e.g. 3-of-5 or 5-of-7), with $M \ge 2$.
  • Inventory of vault locations with ISO 3166-1 alpha-2 country code, the legal jurisdiction that can compel each vault, and the custodian entity operating it.
  • Facility certification evidence. Note these standards cover different threats and none of them alone certifies a vault: ISO/IEC 27001 and SOC 2 Type II are organisational information-security assurances, EN 1143-1 grades burglary resistance of safes and strongrooms, and EN 1047-1 covers fire protection of data media only and confers no burglary resistance.

Workflow

  1. Vault Registration: Build a VaultShardLocation per shard (shard_id, vault_name, country_code, provider_name, optional jurisdiction, is_iso_27001). Country codes and provider names are upper-cased and stripped on construction so 'ch', 'CH' and ' Ch ' group as one country; free-text country names are rejected because they would split a single country into several groups and hide a concentration.
  2. Jurisdiction Mapping: Set jurisdiction explicitly whenever the compelling legal regime differs from the vault's country - a shared bloc, a parent entity's home regime, or a sub-custodian supervised elsewhere. When omitted it defaults to country_code, which is the optimistic assumption; if in doubt, map the broader regime.
  3. Placement Audit: Call audit_distribution. It rejects a malformed matrix outright - wrong shard count, shard ids outside $[1, N]$, or duplicate shard ids, since two copies of one shard are one shard and would otherwise inflate every safety margin.
  4. Interpret Violations: For each country, jurisdiction and provider group the audit reports a Confidentiality violation when the group holds $\ge M$ shards, otherwise an Availability violation when it holds $> N - M$. A placement can fail on availability while passing confidentiality (4-of-6 with 3 shards in one country) - do not treat "no reconstruction risk" as safe.
  5. Redundancy Reserve: The audit flags $N - M$ below min_redundancy_gap (default 2, an internal engineering default, not a regulatory figure). Raise or lower it deliberately; lowering it accepts that fewer lost shards destroy access.
  6. Report: AuditReport carries is_compliant, the per-group maxima, redundancy_gap, the country-level Shannon entropy and the violation list. Remediate every violation and re-audit; entropy is descriptive context only and never gates compliance.

Full procedure: see references/workflows.md. Standards reference: see references/standards.md. Printable pre-flight checklist: see assets/checklist.md.

Common Pitfalls

  • Auditing Confidentiality Only: A 4-of-6 scheme with 3 shards in one country passes a reconstruction-risk check (3 < 4) yet is one country-level outage away from permanent, unrecoverable loss (6 - 3 = 3 < 4). Both directions must be checked on every group.
  • Single Jurisdiction Concentration: Three shards in three different cities of the same country - one domestic court order or tax seizure reaches all three. Multi-city is not multi-jurisdiction.
  • Country-Level Grouping Only: Vaults in different countries can still sit under one effective legal regime through a shared bloc or a sub-custodian's home supervisor. Country counts of 1 each hide this; the jurisdiction field is what catches it.
  • Provider Collusion Risk: Different physical vaults operated by the same commercial custodian, or by subsidiaries of one parent, are one adversary.
  • Duplicated Shards Counted as Coverage: Copying shard 3 into a second vault raises the count of places the key can leak from without raising the number of distinct shards; it weakens confidentiality and adds no reconstruction margin.
  • Threshold Imbalance: A 2-of-3 scheme tolerates only one lost shard; the second loss is terminal. Choose $N - M$ against the real probability of losing a vault, not against convenience.
  • Treating Entropy as a Safety Score: High Shannon entropy does not imply threshold safety - two placements with identical entropy can differ in whether a single group reaches $M$. Only the violation list is authoritative.
  • Assuming a Certificate Covers the Threat: An EN 1047-1 data safe resists fire and provides no defined burglary resistance; ISO 27001 certifies a management system, not the steel around the shard.

Verification

  • Instantiate ColdStorageGeographicDistributor(threshold_m=3, total_shards_n=5) and place 3 shards in Switzerland (CH). Verify the report is non-compliant with a Country Confidentiality SPOF Violation. Redistribute across CH, SG, IS, US and JP and verify a compliant report with max_shards_in_single_country == 1.
  • Instantiate ColdStorageGeographicDistributor(threshold_m=4, total_shards_n=6) with 3 shards in the US and one each in CH, SG, JP. Verify it is non-compliant with a Country Availability SPOF Violation and no confidentiality violation.
  • Run python -m unittest discover -s skills/cold-storage-geographic-distribution-strategy/scripts (16 tests).

Related Skills

  • shamir-secret-sharing-for-key-backup
  • air-gapped-signing-workflow-for-cold-storage
  • regulatory-custody-requirements-by-jurisdiction

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Cold Storage Geographic Distribution Strategy AI skill do?

Use when auditing how M-of-N key shards are spread across facilities, countries and legal regimes, flagging both single points of failure and quorum concentrations. Generating the shards is shamir-secret-sharing-for-key-backup.

Why use Cold Storage Geographic Distribution Strategy on TypingMind?

Because you install it once and use it with any model. Cold Storage Geographic Distribution Strategy is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Cold Storage Geographic Distribution Strategy in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/HimanshuJ16/Algo-Trading-Skills/tree/main/skills/cold-storage-geographic-distribution-strategy. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Cold Storage Geographic Distribution Strategy?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Cold Storage Geographic Distribution Strategy?

As many as you like. As long as a model supports skills, you can use Cold Storage Geographic Distribution Strategy with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Cold Storage Geographic Distribution Strategy AI skill free?

Yes. It is published on GitHub by HimanshuJ16 under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇