data-mesh-expert — Claude Code skill: data mesh design, how to install
data-mesh-expert walks the agent through data mesh work: domains own their data, products ship with contracts, governance is federated and the platform serves itself.
Apache-2.0 168 installs ★ 45 updated 2026-10-12
npx skills add personamanagmentlayer/pcl --skill data-mesh-expert -a claude-codeWhat data-mesh-expert does
The skill gives the agent a structured take on data mesh and covers five areas: domain design, data products, the self-serve platform, governance, and the cultural shift behind them. Rules stay short — tie domains to how the company is organised, give each product a written contract and a version.
A separate section lists anti-patterns as bad/good sketches: one central team owning everything, a single lake for the whole company, schema changes nobody announced, access requests that crawl through email and tickets. Deeper theory sits in references/CORE_CONCEPTS.md and is pulled in when needed.
Allowed tools are Read, Write, Edit, Bash, Glob and Grep, so it can edit files and run shell commands. Keep its work on a branch you can review. Where a central data team still owns every pipeline, expect the skill to push back on that arrangement.
What it helps with
- split data ownership across domains and take load off a central data team
- write a data product contract with a version and a deprecation window
- move one shared data lake to per-domain storage
- replace manual access approvals with automated policy checks
- review an architecture for undocumented schema changes
Example prompts for the agent
- Map our domains and list which data products each one should publish
- Check this S3 layout for a monolithic data lake and propose a split by domain
- Draft a data contract for the
ordersproduct, with versioning and deprecation rules
How to install data-mesh-expert
1Skills CLI (Vercel)
npx skills add personamanagmentlayer/pcl --skill data-mesh-expert -a claude-code2Manually: put the files in place
~/.claude/skills/data-mesh-expert/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/data-mesh-expert/
1Skills CLI (Vercel)
npx skills add personamanagmentlayer/pcl --skill data-mesh-expert -a codex2Manually: put the files in place
.agents/skills/data-mesh-expert/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add personamanagmentlayer/pcl --skill data-mesh-expert -a cursor2Manually: put the files in place
.cursor/skills/data-mesh-expert/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install personamanagmentlayer/pcl stdlib/data/data-mesh-expert2Manually: put the files in place
.github/skills/data-mesh-expert/SKILL.md1Skills CLI (Vercel)
npx skills add personamanagmentlayer/pcl --skill data-mesh-expert -a gemini-cli2Platform CLI
gemini skills install https://github.com/personamanagmentlayer/pcl --path stdlib/data/data-mesh-expertManually: put the files in place
.devin/skills/data-mesh-expert/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add personamanagmentlayer/pcl --skill data-mesh-expert -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about data-mesh-expert
How to install data-mesh-expert in Claude Code?
Run npx skills add personamanagmentlayer/pcl --skill data-mesh-expert -a claude-code. The skill file then lands at ~/.claude/skills/data-mesh-expert/SKILL.md.
What does data-mesh-expert do?
It supplies the agent with data mesh guidance: domain ownership, products with contracts, federated governance and a self-serve platform. Anti-patterns such as a single giant lake are covered too.
Which tools may the skill use?
Read, Write, Edit, Bash, Glob and Grep. Since it can change files and run commands, review the diff before merging.
Source text
SKILL.md · Apache-2.0
--- name: data-mesh-expert version: 1.1.0 description: >- Expert-level data mesh architecture, domain-oriented ownership, data products, federated governance, and self-serve platforms. Use when the user mentions architecture, domain driven, data products, governance, or platform, or when the task involves Data Mesh Principles, Domain-Oriented Data Ownership, Data as a Product, or Self-Serve Data Infrastructure Platform. category: data author: PCL Team license: Apache-2.0 tags: - data-mesh - architecture - domain-driven - data-products - governance - platform allowed-tools: - Read - Write - Edit - Bash - Glob - Grep --- # Data Mesh Expert You are an expert in data mesh architecture with deep knowledge of domain-oriented data ownership, data as a product, federated computational governance, and self-serve data infrastructure platforms. You design and implement decentralized data architectures that scale with organizational growth. ## Best Practices ### 1. Domain Design - Align domains with organizational structure - Clear bounded contexts for each domain - Domain teams own their data end-to-end - Cross-domain collaboration through well-defined interfaces - Avoid centralized data teams; embed in domains ### 2. Data Product Design - Treat data as a product with SLAs - Document data contracts explicitly - Version data products semantically - Implement comprehensive quality checks - Provide discoverability and self-service access - Monitor data product health continuously ### 3. Platform Design - Abstract infrastructure complexity - Provide self-serve capabilities - Automate repetitive tasks - Enable domain autonomy - Standardize common patterns - Invest in developer experience ### 4. Governance - Automate policy enforcement - Make governance policies executable - Balance autonomy with control - Federate decisions to domains - Global standards, local implementation - Continuous compliance monitoring ### 5. Cultural Transformation - Shift from centralized to federated model - Build data literacy across organization - Incentivize data product quality - Foster collaboration between domains - Celebrate data product owners ## Anti-Patterns ### 1. Centralized Data Team ``` // Bad: Central data team owns all data Central Team -> All domains (bottleneck) // Good: Domain teams own their data Sales Domain -> Sales data products Marketing Domain -> Marketing data products Product Domain -> Product data products ``` ### 2. Monolithic Data Lake ``` // Bad: Single giant data lake s3://data-lake/everything/ // Good: Domain-oriented storage s3://data-products/sales/ s3://data-products/marketing/ s3://data-products/product/ ``` ### 3. No Data Contracts ``` // Bad: Undocumented schema changes Breaking change deployed without notice // Good: Versioned contracts with deprecation v1: Deprecated (30 days notice) v2: Current v3: Beta ``` ### 4. Manual Governance ``` // Bad: Manual approval processes Email -> Ticket -> Manual review -> Access granted (weeks) // Good: Automated governance Request -> Policy check -> Auto-approval (minutes) ``` ## Reference Documentation Detailed material lives alongside this skill and is read on demand: - [Core Expertise](references/CORE_CONCEPTS.md) — Data Mesh Principles, Domain-Oriented Data Ownership, Data as a Product, Self-Serve Data Infrastructure Platform, Federated Computational Governance ## Resources - [Data Mesh by Zhamak Dehghani](https://www.oreilly.com/library/view/data-mesh/9781492092384/) - [Data Mesh Principles](https://martinfowler.com/articles/data-mesh-principles.html) - [ThoughtWorks Data Mesh](https://www.thoughtworks.com/en-us/what-we-do/data-and-ai/data-mesh) - [Data Mesh Architecture](https://www.datamesh-architecture.com/) - [Data Product Canvas](https://www.datamesh-architecture.com/data-product-canvas) - [Data Mesh Learning](https://datameshlearning.com/) - [Awesome Data Mesh](https://github.com/jhole89/awesome-data-mesh)
© PCL Team · Apache-2.0 · personamanagmentlayer/pcl