# Luanmorenommaciel Agentspec Agentspec > Data Engineering Guide - Skill: `tomevault-io/luanmorenommaciel-agentspec-agentspec` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/luanmorenommaciel-agentspec-agentspec` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/luanmorenommaciel-agentspec-agentspec/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/luanmorenommaciel-agentspec-agentspec --- # Data Engineering Guide You have access to 23 specialized knowledge base domains and 15+ data engineering agents. Route the user to the right tool based on their task. ## Quick Routing | User Task | Command | Agent | |-----------|---------|-------| | Design a data pipeline / DAG | `/agentspec:pipeline` | pipeline-architect | | Design a schema / star schema / data model | `/agentspec:schema` | schema-designer | | Add data quality checks | `/agentspec:data-quality` | data-quality-analyst | | Review SQL performance | `/agentspec:sql-review` | sql-optimizer | | Choose table format (Iceberg/Delta) | `/agentspec:lakehouse` | lakehouse-architect | | Build RAG / embedding pipeline | `/agentspec:ai-pipeline` | ai-data-engineer | | Create a data contract | `/agentspec:data-contract` | data-contracts-engineer | | Migrate legacy ETL | `/agentspec:migrate` | dbt-specialist + spark-engineer | ## Knowledge Domains Available | Category | Domains | |----------|---------| | Core DE | dbt, spark, airflow, streaming, sql-patterns | | Data Design | data-modeling, data-quality, medallion | | Infrastructure | lakehouse, cloud-platforms, aws, gcp, microsoft-fabric, lakeflow, terraform | | AI & Modern | ai-data-engineering, genai, prompt-engineering, modern-stack | | Foundations | pydantic, python, testing | ## How Agents Use Knowledge 1. Agent reads KB index at `${CLAUDE_PLUGIN_ROOT}/kb/{domain}/index.md` 2. Loads specific pattern/concept file matching the task 3. Falls back to MCP if KB insufficient (max 3 MCP calls) 4. Calculates confidence from evidence matrix ## When to Suggest Commands - User mentions "dbt model" or "staging model" → `/agentspec:schema` or delegate to dbt-specialist - User mentions "pipeline" or "DAG" or "orchestration" → `/agentspec:pipeline` - User mentions "data quality" or "expectations" or "tests" → `/agentspec:data-quality` - User mentions "slow query" or "optimize SQL" → `/agentspec:sql-review` - User mentions "Iceberg" or "Delta Lake" or "table format" → `/agentspec:lakehouse` - User mentions "RAG" or "embeddings" or "vector" → `/agentspec:ai-pipeline` - User mentions "contract" or "SLA" or "schema governance" → `/agentspec:data-contract` - User mentions "migrate" or "legacy" or "SSIS" or "Informatica" → `/agentspec:migrate` --- > Source: [luanmorenommaciel/agentspec](https://github.com/luanmorenommaciel/agentspec) — distributed by [TomeVault](https://tomevault.io).