StevenMPhillips/dremio-skill

Dremio AgentSkill - Query and explore Dremio data lakehouse via AI agents

★ 0Forks 0PythonGitHub ↗Compare

README

Dremio Skill for AgentSkills.io

Query and explore Dremio data lakehouse using natural language. This skill enables AI agents to run SQL queries, discover tables/views, inspect schemas, analyze query performance, and explore data lineage.

Installation

  1. Extract the skill directory to your agent's skills folder
  2. Configure the required environment variables

Required Environment Variables

Variable Description
DREMIO_URL Your Dremio instance URL (e.g., https://dremio.example.com)
DREMIO_PAT Personal Access Token for authentication

Getting a Personal Access Token

  1. Log into your Dremio instance
  2. Go to Account Settings → Personal Access Tokens
  3. Click Generate Token and save it securely

Quick Usage

# Set environment variables
export DREMIO_URL="https://your-dremio-instance.com"
export DREMIO_PAT="your-personal-access-token"

# Search for tables
./scripts/search_views_and_tables.py "customer orders"

# Get table schema
./scripts/get_table_or_view_schema.py "Sales" "Orders"

# Run a query
./scripts/run_sql.py 'SELECT * FROM "Sales"."Orders" LIMIT 10'

# Analyze query performance
./scripts/get_recent_jobs_summary.py --hours 24
./scripts/analyze_query_profile.py <job-id>

Features

  • SQL Execution: Run SELECT, DML, and EXPLAIN queries
  • Discovery: Semantic search for tables and views
  • Schema Inspection: Get column definitions and metadata
  • Performance Analysis: Analyze query profiles and job history
  • Data Lineage: Trace data dependencies

Documentation

See SKILL.md for complete documentation including:

  • All available tools and their usage
  • Task-specific reference guides
  • Best practices and core rules

License

See LICENSE file for terms.

Contributors

StevenMPhillips

Issues