Anthropic's official agentic coding CLI. Terminal-native, tool-use-driven, with deep file system and shell access. #1 SWE-bench Pro standardized (45.89%), ~4% of GitHub public commits (SemiAnalysis), $2.5B annualized revenue (fastest enterprise SaaS to $1B ARR). 8M+ npm weekly downloads. Opus 4.6 with 1M context.
90/100
Trust
80K+ ↑121%
Stars
17
Evidence
26.7 MB
Repo size
Videos
Reviews, tutorials, and comparisons from the community.
Claude Code Tutorial #1 - Introduction & Setup
Claude Code Tutorial for Beginners
Claude Code - Full Tutorial for Beginners
Repo health
1d ago
Last push
6,874
Open issues
6,582
Forks
49
Contributors
Editorial verdict
The #1 coding CLI agent. Leads SWE-bench Pro standardized (45.89%), wins independent head-to-heads on reasoning depth, ~4% of GitHub commits. Rate limits are the #1 complaint. Costs 2-3x more per task than Codex CLI due to higher token consumption.
Source
GitHub: anthropics/claude-code
Public evidence
Claims Claude Code is best AI coding agent. Reference for feature claims only — Faros promoting Faros.
Ranks Claude Code highest for complex reasoning. Useful comparison structure but company blog, not independent review.
Major front-page HN thread on the Claude Code 2.0 release. Community consensus positioned it as the standard agentic coding CLI all others are measured against.
In-depth practitioner report from a real product team on daily Claude Code usage over 6 weeks. One of the highest-comment HN threads on any AI coding tool.
Tracks Claude Code VS Code extension installs surging from 17.7M to 29M since Jan 2026. Cites UC San Diego/Cornell survey where Claude Code (58 users) edged out GitHub Copilot (53 users) among 99 professional devs.
Swapping models changed scores ~1%. Swapping the harness changed them 22%. Claude Code's agent engineering (Agent Teams, parallel sub-agents) is the category's deepest scaffolding advantage.
Claude Code leads SWE-bench Pro standardized at 45.89% vs Codex CLI's 41.0%. SWE-bench Verified is saturated (top 5 within 1 point). Pro standardized is the fair comparison — custom scaffold scores (Codex 56.8%) are not comparable.
~4% of public GitHub commits, projected 20%+ by EOY 2026. 42,896x growth in 13 months. $2.5B annualized revenue. The hardest real-usage metric in the category.
56% baseline pass rate, no statistically significant degradation detected. No other tool in the category has this level of ongoing independent quality assurance.
One of the highest-engagement personal testimonials in the coding CLI category. Signals cultural mindshare beyond pure developer tooling.
$2.5B annualized revenue, 500+ customers at $1M+/year. Fastest enterprise SaaS to $1B ARR in history. 8M+ npm weekly downloads — 3x Codex CLI, 12x Gemini CLI.
8M+ weekly downloads is the hardest-to-game traction metric in the category. 3x Codex CLI (2.6M), 12x Gemini CLI (647K). Installs correlate with actual developer usage, not social media hype.
Maintainer merge rates are ~24pp lower than automated grading. Claude Code's 'deepest reasoning' positioning suggests it would benefit from a maintainer-merge-rate benchmark. Validates quality-over-speed thesis.
Claude Code completed a React project in 1h17m single-shot. Gemini CLI took 2h02m with multiple retries. Demonstrates Claude Code's reasoning advantage on practical tasks beyond synthetic benchmarks.
Morph's independent Tier 1 classification. Called Claude Code 'deepest reasoning on hard problems — best for complex refactors, unfamiliar codebases, architectural decisions.' Only Claude Code and Codex CLI made Tier 1.
High-engagement thread questioning whether Claude 4.5 has regressed in quality. MarginLab's independent monitoring found no statistically significant degradation (p<0.05), but community perception of quality regression is real and ongoing. The trust cost of this thread is itself a factor.
"4% of GitHub public commits are being authored by Claude Code right now... 42,896x growth in 13 months." (Dylan Patel, SemiAnalysis, Feb 2026). The single most cited real-usage metric in the category.
How does this compare?
See side-by-side metrics against other skills in the same category.
Growth
GitHub stars over time.
GitHub Stars
Where it wins
#1 SWE-bench Pro standardized (45.89%) — the authoritative benchmark now that Verified is saturated
~4% of GitHub public commits — strongest real-usage signal (SemiAnalysis), $2.5B annualized revenue (fastest enterprise SaaS to $1B ARR — Constellation Research)
8M+ npm weekly downloads — 3x Codex, 12x Gemini. Installs are the hardest metric to game
Opus 4.6 with 1M context — highest-capability model available in a CLI
HN peak engagement 2,127 pts — unmatched community mindshare
Wins independent blind code quality tests 67% of the time across independent evaluations
Independent daily quality monitoring with no degradation detected (MarginLab, p<0.05)
Real-world daily cost ~$6 avg, <$12 at p90
Where to be skeptical
Rate limits are the #1 complaint (1,085 pts HN thread dedicated to this)
Costs $200+/month at heavy usage — 2-3x more expensive per task than Codex CLI due to 3-4x higher token consumption
Terminal-Bench 2.0: 58.0% (#38) — genuinely weaker on terminal-native/systems tasks than Codex CLI (62.9%)
Tied to Anthropic models only
Anthropic blocked third-party subscription use, angering users (625 pts HN)
Quality regression perception: 'Claude Code is being dumbed down?' HN thread (1,085 pts, 702 comments, Feb 2026) — MarginLab monitoring shows no statistical regression but community trust is a live issue
Ranking in categories
Featured in bundles
Andrej Karpathy
Karpathy Three-Tier Coding Stack
2 skills · 2025-12
swyx (Shawn Wang)
swyx Research-First Agent Stack
2 skills · 2026-01
McKay Wrigley
McKay Wrigley Shipping Stack
1 skills · 2026-02
Simon Willison
Simon Willison Terminal Agents Stack
3 skills · 2025-06
Pieter Levels
Levelsio Server-First Coding Stack
2 skills · 2025-08
Yam Peleg
Yam Peleg CLI Agents Field Report
3 skills · 2025-12
Know a better alternative?
Submit evidence and we'll run the full pipeline.
Similar skills
Aider
91Open-source AI pair programming CLI. The only tool in the coding CLI category with a fully verifiable, independent download number: 191,828/week PyPI installs, 5.7M lifetime, 15B tokens/week (homepage stat). Multi-model, git-native, no vendor lock-in. v0.86.2 released 2026-02-12.
OpenHands
88Category leader in multi-agent orchestration — 69,352 stars (verified), $18.8M Series A, AMD hardware partnership, 455 contributors, 1M downloads/month PyPI (3.4M all-time). SWE-Bench Verified 72% with Claude 4.5 Extended Thinking (updated 2026-03-19), Multi-SWE-Bench #1 across 8 languages. Gap to #2 is enormous on every axis.
Gemini CLI
86Google's open-source terminal agent with Gemini 3 models, 1M token context, built-in Google Search grounding, and the best free tier in the category (1K req/day). 97.9K stars, 444 contributors. SWE-bench Pro standardized 43.30% (#2 behind Claude Code).
Codex CLI
85OpenAI's open-source coding agent built in Rust. Terminal-Bench 77.3% (GPT-5.3-Codex), SWE-bench Pro standardized 41.04%. 3-4x more token-efficient than Claude Code, 60-75% cheaper per task. Free with ChatGPT subscription, sandbox-first execution, 619 releases in 10 months.
Raw GitHub source
Constrained peek so you can sanity-check the source material without leaving the site.
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows -- all through natural language commands. Use it in your terminal, IDE, or tag @claude on Github.
Learn more in the official documentation.
<img src="https://raw.githubusercontent.com/anthropics/claude-code/main/demo.gif" />[!NOTE] Installation via npm is deprecated. Use one of the recommended methods below.
For more installation options, uninstall steps, and troubleshooting, see the setup documentation.
Install Claude Code:
MacOS/Linux (Recommended):
curl -fsSL https://claude.ai/install.sh | bash
Homebrew (MacOS/Linux):
brew install --cask claude-code
Windows (Recommended):
irm https://claude.ai/install.ps1 | iex
WinGet (Windows):
winget install Anthropic.ClaudeCode
NPM (Deprecated):
npm install -g @anthropic-ai/claude-code
Navigate to your project directory and run claude.
This repository includes several Claude Code plugins that extend functionality with custom commands and agents. See the plugins directory for detailed documentation on available plugins.
We welcome your feedback. Use the /bug command to report issues directly within Claude Code, or file a GitHub issue.
Join the Claude Developers Discord to connect with other developers using Claude Code. Get help, share feedback, and discuss your projects with the community.
When you use Claude Code, we collect feedback, which includes usage data (such as code acceptance or rejections), associated conversation data, and user feedback submitted via the /bug command.
See our data usage policies.
We have implemented several safeguards to protect your data, including limited retention periods for sensitive information, restricted access to user session data, and clear policies against using feedback for model training.
For full details, please review our Commercial Terms of Service and Privacy Policy.