terse-cli 0.1.5

Python CLI for scaffolding and managing Terse projects.

terse-cli

Python CLI for scaffolding, generating, testing, and deploying Terse jobs.

Installation

pipx install terse-cli

The CLI installs the published terse-sdk automatically.

Don't have pipx? Install it first:

# macOS
brew install pipx && pipx ensurepath

# Ubuntu/Debian
sudo apt install pipx && pipx ensurepath

# Generic
python3 -m pip install --user pipx && python3 -m pipx ensurepath

Then restart your shell.

Requirements

• Python >=3.11
• uv for working with scaffolded projects

The CLI is installed via pipx, but the generated project scaffold uses uv sync and uv run.

Quick Start

terse init my-terse-job
cd my-terse-job
uv sync
terse test

If you connect integrations later, regenerate the project helpers:

terse generate

Commands

• terse init [PROJECT_NAME]
• terse generate
• terse integrate
• terse test [JOB_NAME]
• terse deploy

Supported Generated Helpers

The current Python codegen surface is:

• Schedule.cron(...)
• Attio.skill(...)
• Snowflake.skill(...)
• deterministic wrappers on agent.tools.attio and agent.tools.snowflake

terse generate writes these helpers into src/terse_generated.py inside your project.

For example, after generating Snowflake helpers and registering skills=[Snowflake.skill()], a job can call:

from terse_sdk import CronJobInputEvent
from terse_generated import TerseAgent

def handle(event: CronJobInputEvent, agent: TerseAgent) -> None:
    result = agent.tools.snowflake.execute_query(query="select current_date()")
    print(result.rowCount)

Environment Variables

• TERSE_API_KEY: required for terse generate, terse deploy, and any agent/tool calls made by your job

Typical Workflow

terse init my-terse-job
cd my-terse-job
uv sync
terse test
terse deploy

Source

• Homepage: useterse.ai