A
Claude CodeWrite trustworthy analytics SQL and dbt models
SQL Analytics and dbt Lab
Setuproll editorial@setuproll88.0Overall score
Connects the agent to a real warehouse so it validates queries against live schemas, then models transformations as tested dbt layers instead of one-off scripts. For data folks who want analytics that survive a schema change and a second reader.
88.0Score
5Components
Get this build
terminal
claude mcp add postgres -- npx -y @modelcontextprotocol/server-postgresWhat gets written
- CLAUDE.md
- .claude/agents/query-author.md
- .claude/agents/schema-validator.md
- .claude/agents/metric-reviewer.md
- .mcp.json
- HOOKS.md
Components
Model
- Claude Opus 4.8
MCP servers
- postgres
- filesystem
- github
Subagents
- query-author
- schema-validator
- metric-reviewer
Hooks
- PostToolUse: explain analyze
- pre-commit: dbt test
Rules
- Validate column names against the live schema
- Define metrics once, reuse everywhere
- Every model gets a test and a description
Summary
Connects the agent to a real warehouse so it validates queries against live schemas, then models transformations as tested dbt layers instead of one-off scripts. For data folks who want analytics that survive a schema change and a second reader.
88.0 score
0 Reviews
Your rating
Sign in to post
Loading discussion...