Agent

The agent command passes off a multi-step orchestration task to an AI agent. This allows delegation of complex, multi-step tasks using natural language.

Note: This command is not yet implemented.

Arguments:

@argument("command", type=str)
@option("--context", "-c", multiple=True, help="Additional context files or paths for the agent")
@option("--model", "-m", type=str, help="AI model to use (if configured)")
@option("--interactive", "-i", is_flag=True, help="Run in interactive mode with feedback")

Planned Examples

# Analyze and create a dashboard
trilogy agent "analyze sales trends and create a dashboard"

# Ingest and validate new data
trilogy agent "ingest new data and run validation tests"

# Optimize queries
trilogy agent "optimize query performance for customer reports"

# With additional context
trilogy agent "create ETL pipeline" -c existing_models/ -c requirements.txt

# Interactive mode
trilogy agent "refactor the data model" --interactive

Notes

• Requires AI agent configuration in trilogy.toml
• The --interactive flag enables step-by-step feedback during execution
• Context files help the agent understand your project structure