# Agent

Trawler agent can currently extract metadata from relational databases. Support for additional systems will come in the near future.

To get started with configuring trawler-agent using docker, check out the Getting Started guide. You can also install the agent via pip.

pip install trawler-python
export TRAWLER_TOKEN=<token>
export TRAWLER_ENDPOINT=http://localhost:9090

# Configuration

The agent's tasks can either be configured directly using the command line or from a supplied YAML file.

To run a single task from a configuration file, example.yml:

trawler run --now example.yml

This mode may also be useful if you want to use your own job scheduler. To make life easy, the agent comes with a very simple scheduler which uses the Python schedule package. To run the agent in daemon mode using any supplied job schedules:

trawler run example.yml