IndexMapAbout
Tool / RuntimeLLM

Claude Code

Also known as: Claude Code CLI

Plain English

Claude Code is Anthropic’s coding assistant for working directly in a codebase. It can read and edit files, run commands, and check the result with the project’s own tooling.

Technical Overview

The harness connects Claude models to repository, shell, and MCP tools. Permission rules and hooks govern tool calls, and sessions can run locally or in an isolated cloud environment.

Chronological Entries (4)

#4Release
AI-Assisted

Claude Code 2.1.328 added a fail-closed option for MCP tool hooks. If a hook configured this way fails, the harness blocks the tool action, giving teams a way to require that a policy check completes before a tool runs.

#3Release
AI-Assisted

Anthropic released Claude Code on the web. Each session runs in an isolated cloud environment, and a separate service handles access to the source repository so repository credentials stay outside the sandbox.

#2Release
AI-Assisted

Alongside Claude 4, Anthropic graduated Claude Code to general availability and shipped native VS Code, JetBrains, and GitHub Actions integrations so the agent could run in both interactive editors and background CI tasks.

#1Release
AI-Assisted

Anthropic introduced Claude Code as a limited research preview for terminal-based coding. The preview let Claude navigate a codebase, change files, and run tests from the developer’s working environment.

Direct Relations (2)