nosql-expert — Claude Code skill: what it does, how to install
Guides query-first schema design for Cassandra and DynamoDB, helping developers pick optimal partition keys and avoid hot partitions under heavy traffic.
npx claude-code-templates@latest --skill development/nosql-expertWhat nosql-expert does
The skill shifts schema architecture toward query-first thinking. In distributed wide-column and key-value systems, tables cannot serve ad-hoc queries without penalty. Ad-hoc joins do not exist here. Everything depends on read patterns defined before writing code. It targets systems like Cassandra, ScyllaDB, and DynamoDB instead of standard SQL or document stores.
Guidance covers single-table layouts, composite primary keys, and secondary index behavior. It helps pick high-cardinality partition keys so write throughput spreads evenly across nodes. It also checks if a partition threatens to exceed 10GB. When partitions grow indefinitely, sharding by date prefixes becomes necessary.
Using ALLOW FILTERING in production Cassandra queries signals an architectural flaw because it triggers full cluster scans. The skill also highlights write costs: while inserts append quickly to LSM trees, frequent deletes generate harmful tombstones.
What it helps with
- structure single-table schemas for Amazon DynamoDB
- select high-cardinality partition keys to prevent hot nodes
- map relational relationships into denormalized Cassandra tables
- audit database schemas for unbounded partition growth
Example prompts for the agent
- Design a DynamoDB single-table schema to store customer profiles alongside their orders
- Inspect this Cassandra table design for hot partition risks and unbounded growth
How to install nosql-expert
1claude-code-templates
npx claude-code-templates@latest --skill development/nosql-expert2Skills CLI (Vercel)
npx skills add davila7/claude-code-templates --skill nosql-expert -a claude-code3Manually: put the files in place
~/.claude/skills/nosql-expert/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/nosql-expert/
Manually: put the files in place
.agents/skills/nosql-expert/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/nosql-expert/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install davila7/claude-code-templates nosql-expert2Manually: put the files in place
.github/skills/nosql-expert/SKILL.mdPlatform CLI
gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/development/nosql-expertManually: put the files in place
.devin/skills/nosql-expert/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.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 nosql-expert
How to install nosql-expert in Claude Code?
Run npx claude-code-templates@latest --skill development/nosql-expert in your terminal. You can also install it using npx skills add davila7/claude-code-templates --skill nosql-expert -a claude-code.
Why does this skill focus on query-first design?
Distributed NoSQL engines like Cassandra and DynamoDB do not support cost-effective runtime joins. Data models must directly reflect application read patterns, often duplicating records to preserve single-lookup speeds.
What is wrong with using ALLOW FILTERING in Cassandra?
It triggers an exhaustive scan across the entire cluster rather than querying indexed partitions directly. Needing this directive indicates that primary and clustering keys do not match actual read requirements.