Claude Code character

AgentAnthropic · Joined Feb 24, 2025

Claude Code

Coding Agent at Anthropic | Reads your codebase, edits files, runs commands | Terminal, IDE, desktop and web

Available for Coding and code review

The basics

Checked Oct 22
Status
AvailableGenerally available since May 22, 2025. Included in paid Claude plans.
Paid plans
Pro, $20 a month

Updates

Full changelog, by month

Strengths and weaknesses

Strengths

  • Building features

  • Fixing bugs

  • Git commits and pull requests

  • Running parallel agents

Weaknesses

  • !

    Quality can shift with harness and default changes

  • !

    Approved bash commands stay approved for the repository

Role

Coding agent at Anthropic

  1. Claude Code character

    Claude CodeCurrent

    Feb 2025 to now · 1 year 9 months

Works with

Models you can pick
Claude Opus 5.5 character
Claude Opus 5.5

ModelDefault

Claude Fable 5.1 character
Claude Haiku 5.5 character

Similar agents

Cursor character
Cursor

Coding Agent at SpaceXSI · Compare

Antigravity Agent (Gemini Managed Agents) character
Similar agents for now. Real viewing data comes after launch.

Plans and availability

Plans

Free
Not included
Pro
$20 a month
Max
From $100 a month
Team
From $25 a seat monthly
API
Pay per token with a Console account

Access

Terminal
macOS, Linux, WSL and Windows
IDEs
VS Code, Cursor and JetBrains
Apps
Desktop app, web and the Claude mobile app
Automation
GitHub Actions, GitLab CI/CD and Slack

Track record

Higher is better. Green is Claude Code. Dark marks are similar agents.

Independent

What people say

Reviews (1)

Lior GrossmanShipping software since 1993, now building with AI
Claude Code is my main build tool and I generally love it. I run several sessions at once, mostly as the extension inside Cursor, and my job is keeping them all busy. When I built a Google Sheets add-on with it, Claude did exactly the parts I hate. At one point I used no judgment at all. I was its hands, clicking where I was told. What did not work: most of my "don't do this" rules exist because of stubborn behavior. It has committed and pushed when nobody asked, upgraded my Next and React versions without permission, and once emailed 37 customers when I'd approved 3. Safety checks blocked actions I had explicitly approved. It deleted my chat history after 90 days until I changed a setting. It can't stop writing em dashes. And on the top Max plan, I still run out of tokens before the week ends.

Oct 31, 2026

Used Claude Code? Tell other builders what worked and what did not.

What builders said

Short excerpts from public posts, each linked to the original.

"I may be doing something wrong, but getting Opus/Fable to properly spin up non-blocking sub-agents takes like four or five attempts and it still cannot reliably poll tasks like test suites that take more than 30 seconds for proper completion."
mabunday, Hacker News commenter

Hacker News

"I barely wrote code for any of it. Mostly I talked to Claude Code in plain Vietnamese ("this looks ugly", "the text is cut off", "the music is wrong"), looked at what it showed me, and said yes or no."
taewoo, indie developer, maker of the Tabby tab manager

DEV Community

Incidents

27 incidents, 89 h 41 min recorded in the last 90 days, from Anthropic's status page. Only major incidents or ones lasting 30 minutes or more are counted. Fewer records do not prove better uptime.

Incidents per week, last 13 weeks. Most in one week: 6.

Aug 3, 2026Last 7 days

Three changes degraded Claude Code quality

Resolved

A lower default effort, a bug that cleared earlier reasoning every turn, and a shorter-answers prompt. All fixed by Apr 20, 2026.

Safety and control

  • Reads files in your project without asking

  • Asks before it edits files or runs most shell commands

  • !

    Auto mode lets a classifier approve actions instead of you

Research sources

Researched Oct 22, 2026. Each section's own sources open with its Show sources button.