#cli

11 articles

Google Open-Sources Workspace CLI: Empowering AI Agents for Next-Gen Office Automation
AI 人工智慧

Google Open-Sources Workspace CLI: Empowering AI Agents for Next-Gen Office Automation

Google Workspace team has open-sourced the Workspace CLI, a command-line tool integrating operations across Gmail, Calendar, and more. This move unlocks unprecedented automation possibilities for developers and AI agents, despite its unofficial status, signaling a shift in enterprise productivity and AI integration.

Vercel Introduces New API CLI Command: A Modern Terminal Tool for the AI Agent Era
軟體開發

Vercel Introduces New API CLI Command: A Modern Terminal Tool for the AI Agent Era

Vercel version 50.5.1 debuts the new `vercel api` command, granting developers and AI agents direct terminal access to the full suite of Vercel APIs. This update streamlines infrastructure management and features permission inheritance, enabling tools like Claude Code to integrate seamlessly into dev workflows, signaling a shift toward 'Agent-First' developer tools.

The Era of Agentic Workflows: How GitHub Copilot CLI is Transforming the Terminal
軟體開發

The Era of Agentic Workflows: How GitHub Copilot CLI is Transforming the Terminal

Since its public preview launch in September 2025, GitHub Copilot CLI has evolved into a powerful agentic AI tool. It moves beyond simple chat, enabling developers to automate complex terminal tasks like repository setup, visual bug fixing via screenshots, and headless scripting, all while integrating deeply with the broader GitHub ecosystem.

The Best Code is No Code: Composing CLIs vs. Bespoke Tools in the LLM Era
軟體開發

The Best Code is No Code: Composing CLIs vs. Bespoke Tools in the LLM Era

As we move into 2026, a paradigm shift is occurring in how LLM agents interact with tools. Instead of defining granular tool calls like MCP, developers are finding immense power in teaching models to compose Unix-style shell commands. By leveraging tools like Restish and oauth2c, we can reduce maintenance overhead and token costs while creating more transparent and scriptable workflows.