Everything Claude Code hits 163K stars: what developers should know

I've watched the Everything Claude Code repo grow from a hackathon project into one of the most-starred repositories on GitHub, and the trajectory makes sense. Everything Claude Code (ECC) solves a problem every team hits once they start taking AI coding agents seriously: the defaults are generic, the configuration is per-project, and the work doesn't carry over.

Created by Affaan Mustafa, ECC packages 48 agents, 183 skills, rules, hooks, and security scanning into a portable, version-controlled system that works across Claude Code, Cursor, Codex, and OpenCode. At 163K stars and 25.3K forks, it's the most visible attempt at a shared configuration layer for AI coding agents I've seen.

What Happened

The Everything Claude Code repository shipped v1.10.0 in April 2026 and has seen rapid adoption since. The project started as a personal config pack by an Anthropic hackathon winner and has grown into a cross-platform system with 161 contributors and support for 12 language ecosystems.

What I'd highlight specifically about v1.10.0: it's not an incremental release. It adds a desktop dashboard GUI, an ECC 2.0 alpha written in Rust, expanded operator workflows, and richer Codex and OpenCode installation surfaces. The Rust control-plane prototype in particular is a sign that this project is thinking beyond configs toward a more durable architecture.

The repo now has 48 agents, 183 skills, and 79 legacy command shims built over 10+ months of daily use on real production projects.

Key Features

• Cross-platform agent configs. Ships ready-made configurations for Claude Code, Cursor, Codex (app and CLI), OpenCode, Gemini, and Antigravity, all from one repo. The fact that it covers this many tools from a single install is what makes it worth the setup time.
• 183 skills across 12 language ecosystems. Covers TypeScript, Python, Go, Java, Kotlin, Rust, Swift, PHP, Perl, C++, and more. Each skill is a structured SKILL.md file with YAML front matter, which means they're readable, modifiable, and portable.
• AgentShield security scanner. Runs 1,282 tests against 102 static analysis rules. The --opus flag launches three Claude Opus agents in a red-team/blue-team/auditor pipeline for adversarial analysis of your agent configs. This is the feature I'd evaluate first for any team running MCP configs in production.
• Hook runtime controls. Environment variables (ECC_HOOK_PROFILE, ECC_DISABLED_HOOKS) let teams gate hook strictness without editing config files. Useful for teams with different risk tolerances across environments.
• Selective install architecture. Manifest-driven pipeline with install-plan.js and install-apply.js for targeted component installation. A Python shop installs only Python rules. A Go team installs only Go agents. No noise.
• Continuous learning v2. An instinct-based system that extracts patterns from sessions, scores confidence, and clusters related instincts into reusable skills via /evolve. The agent gets better the more your team uses it.

Why It Matters

AI coding agents ship with generic defaults. Teams spend hours writing custom CLAUDE.md files, Cursor rules, and MCP configurations, and then repeat that work on the next project. ECC packages those patterns into a portable, version-controlled system that travels with your team.

The practical result: a team using Claude Code and Cursor can install a single repo and get consistent TDD workflows, security-scanning hooks, and code-review agents across both tools. That's not a minor convenience. That's hours of setup time recovered on every new project.

The broader pattern I'm seeing is that as AI coding tools proliferate, developers need a shared layer to manage agent behavior, enforce quality gates, and prevent context-window bloat. ECC is the most visible attempt at that layer, and the 163K stars suggest it's landing.

Example Use Case

A team building a Next.js and Supabase SaaS app installs ECC with ./install.sh --profile full, then runs /ecc:plan "Add user authentication with OAuth". The planner agent produces an implementation blueprint. They switch to /tdd for test-first development, then /security-scan to run AgentShield against their CLAUDE.md and MCP configs.

The beforeSubmitPrompt hook catches any sk- or ghp_ tokens before they reach the model. When context runs long, the strategic-compact skill suggests /compact at logical breakpoints instead of waiting for the 95% auto-compaction threshold.

This is the workflow I'd walk through with a team that's been spending too much time on agent setup and not enough time shipping. ECC compresses that setup into a single install.

Competitive Context

Claude Code is the primary target. ECC ships 48 agents, 183 skills, and 8 hook event types. Cursor receives 15 hook event types via a DRY adapter pattern that reuses Claude Code's hook scripts without duplicating them. The .cursor/hooks/adapter.js module transforms Cursor's stdin JSON format to match Claude Code's, so the same scripts/hooks/*.js files work across both tools.

Codex support is newer and more limited. According to the repository's feature parity table, Codex lacks hook-execution parity with Claude Code, so enforcement relies on AGENTS.md instructions and sandbox permissions rather than on programmatic hooks. OpenCode actually leads in some areas: it offers 11 hook event types and 6 native custom tools compared to Claude Code's 8 event types.

The gap ECC fills is standardization. Each tool has its own config format: settings.json for Claude Code, hooks.json and rules/ for Cursor, config.toml for Codex, opencode.json for OpenCode. ECC abstracts over all four, with AGENTS.md at the root serving as the universal cross-tool file that all supported tools read. That's the architectural bet worth paying attention to.

My Take

Everything Claude Code is the largest open-source attempt to standardize AI coding agent configuration across tools. If your team uses Claude Code, Cursor, or Codex and spends time writing custom agent rules, security hooks, or TDD workflows, this repo packages those patterns into a single installable system.

At 163K stars, 25.3K forks, and 161 contributors, the community weight behind it is real. The Rust-based ECC 2.0 alpha in-tree tells me this project is planning for the long term, not just accumulating stars. Worth evaluating for any team running AI agents in production.