The julep init command is used to initialize a new Julep project using a predefined template. You can check the list of available templates in our library.
# Initialize a project with the default template (hello-world template)julep init# Initialize a project with a specific templatejulep init --template "profiling-recommending" --path "./my-project" # Will be initialized in ./my-project/profiling-recommending# Initialize a project and skip confirmationjulep init --yes
The julep init command is used to initialize a new Julep project using a predefined template. You can check the list of available templates in our library.
# Initialize a project with the default template (hello-world template)julep init# Initialize a project with a specific templatejulep init --template "profiling-recommending" --path "./my-project" # Will be initialized in ./my-project/profiling-recommending# Initialize a project and skip confirmationjulep init --yes
Source directory to sync from (default: current directory)
Examples:
Copy
# Basic syncjulep sync# Force local changes to override remotejulep sync --force-local# Force remote state to override localjulep sync --force-remote# Sync with specific source directoryjulep sync --source ./my-project# Watch for changes and synchronize automaticallyjulep sync --watch
Copy
julep import --agent --id agent_id --output path
The julep import command is used to import an agent, task, or tool from the Julep platform.
Note: Importing tasks and tools is not yet supported; only agents are supported.
Wait for the task to complete before exiting, stream logs to stdout
Examples:
Copy
# Run task with inline JSON inputjulep run --task "00000000-0000-0000-0000-000000000000" --input '{"prompt": "Write a story about a magical forest"}'# Run task with input from filejulep run --task "00000000-0000-0000-0000-000000000000" --input-file ./inputs/story-params.json# Run task and wait for completionjulep run --task "00000000-0000-0000-0000-000000000000" --input '{"style": "watercolor"}' --wait# Run task that doesn't require input (empty object)julep run --task "00000000-0000-0000-0000-000000000000" --input '{}'# Invalid task IDjulep run --task bad-id --input '{}'# -> Error creating execution: Task not found
Copy
julep run --task <task_id> --input '{"key": "value"}'
Wait for the task to complete before exiting, stream logs to stdout
Examples:
Copy
# Run task with inline JSON inputjulep run --task "00000000-0000-0000-0000-000000000000" --input '{"prompt": "Write a story about a magical forest"}'# Run task with input from filejulep run --task "00000000-0000-0000-0000-000000000000" --input-file ./inputs/story-params.json# Run task and wait for completionjulep run --task "00000000-0000-0000-0000-000000000000" --input '{"style": "watercolor"}' --wait# Run task that doesn't require input (empty object)julep run --task "00000000-0000-0000-0000-000000000000" --input '{}'# Invalid task IDjulep run --task bad-id --input '{}'# -> Error creating execution: Task not found
Copy
julep logs --execution-id exec_id [--tail]
The julep logs command is used to view execution logs.
# View logs for a specific executionjulep logs --execution-id "00000000-0000-0000-0000-000000000000"# Stream logs in real-timejulep logs --execution-id "00000000-0000-0000-0000-000000000000" --tail# Get logs in JSON formatjulep logs --execution-id "00000000-0000-0000-0000-000000000000" --json
# Create a basic agentjulep agents create --name "Story Writer" --model "gpt-4"# Create agent with full configurationjulep agents create \ --name "Content Assistant" \ --model "gpt-4" \ --about "AI assistant for content creation" \ --metadata '{"type": "content", "version": "1.0"}' \ --instructions "Focus on creative writing" \ --instructions "Maintain professional tone"# Create agent from definition filejulep agents create --definition ./agents/content-assistant.yaml
Note: When creating an agent, the agent doesn’t get automatically imported to the project. You can run julep import --agent --id <agent-id> to import the agent to the project.
# Create a basic agentjulep agents create --name "Story Writer" --model "gpt-4"# Create agent with full configurationjulep agents create \ --name "Content Assistant" \ --model "gpt-4" \ --about "AI assistant for content creation" \ --metadata '{"type": "content", "version": "1.0"}' \ --instructions "Focus on creative writing" \ --instructions "Maintain professional tone"# Create agent from definition filejulep agents create --definition ./agents/content-assistant.yaml
Note: When creating an agent, the agent doesn’t get automatically imported to the project. You can run julep import --agent --id <agent-id> to import the agent to the project.
Copy
julep agents update --id agent_id \ --name "New Name" \ --model "New Model"
The julep agents update command is used to update an existing agent.
# Delete with confirmation promptjulep agents delete --id "00000000-0000-0000-0000-000000000000"# Force delete without confirmationjulep agents delete --id "00000000-0000-0000-0000-000000000000" --force# Delete multiple agents using a bash for loopfor id in agent_id1 agent_id2 agent_id3; do julep agents delete --id "$id" --forcedone
Copy
julep agents list [--metadata-filter '{"key": "value"}'] [--json]
The julep agents list command is used to list all agents.
# List all agentsjulep agents list# List agents with JSON outputjulep agents list --json# Filter agents by metadatajulep agents list --metadata-filter '{"type": "content"}'# List agents and process with jqjulep agents list --json | jq '.[] | select(.model=="gpt-4")'# Save agent list to filejulep agents list --json > agents.json
Copy
julep agents get --id agent_id [--json]
The julep agents get command is used to get details of a specific agent.
# Get agent detailsjulep agents get --id "agent_abc123"# Get agent details in JSON formatjulep agents get --id "agent_abc123" --json# Save agent details to filejulep agents get --id "agent_abc123" --json > agent.json# Get and process with jqjulep agents get --id "agent_abc123" --json | jq '.instructions'
The CLI will soon include a julep assistant command that launches an interactive prompt for generating commands from plain language. This feature is still under development.