ECC Tools (Everything Claude Code)

Everything Claude Code (ECC) is a comprehensive, open-source performance optimization system for AI agent harnesses — the infrastructure that guides, controls, and optimizes AI coding agents so they produce reliable output without constant human supervision. Evolved over 10+ months of intensive daily use building real products, ECC ships production-ready agents, skills, hooks, rules, MCP configurations, and legacy command shims that work across Claude Code, Codex, Cursor, OpenCode, Gemini, and other AI coding tools. The system won the Anthropic x Forum Ventures hackathon and has grown to 140K+ stars with 170+ contributors across 12+ language ecosystems.

The core problem ECC solves is that every codebase contains implicit knowledge — architectural decisions, naming conventions, testing patterns, deployment workflows — buried in commit history. Without a harness, each AI coding session starts from scratch. ECC makes this knowledge explicit and machine-readable, giving agents computed context at session start so they follow your conventions from the first line of code.

• 47 Specialized Agents — Delegate tasks to purpose-built subagents for planning, architecture, TDD, code review, security analysis, build error resolution, documentation syncing, and more; invoke via /ecc:plan, /tdd, /code-review, etc.
• 181 Skills — The primary workflow surface covering coding standards, backend and frontend patterns, TDD, security review, continuous learning, eval harnesses, and domain-specific frameworks including Django, Spring Boot, Laravel, Go, Rust, Swift, Perl, and more.
• Automated Skill Generation — Install the GitHub App, comment /ecc-tools analyze on any issue, and receive a PR with SKILL.md and instincts extracted from up to 5,000 commits of git history.
• Continuous Learning v2 — Instinct-based learning system that automatically extracts patterns from sessions with confidence scoring; use /instinct-status, /instinct-import, /instinct-export, and /evolve to cluster related instincts into full skills.
• Hook-Based Automations — Trigger-based hooks (PreToolUse, PostToolUse, Stop, SessionStart) for memory persistence, strategic compaction, secret detection, auto-formatting, and TypeScript checks; runtime-controlled via ECC_HOOK_PROFILE and ECC_DISABLED_HOOKS.
• AgentShield Security Auditor — Run npx ecc-agentshield scan to audit CLAUDE.md, settings.json, MCP configs, hooks, and agent definitions across 102 static analysis rules and over 1,200 tests; supports red-team/blue-team/auditor pipeline.
• 79 Legacy Command Shims — Familiar slash-command entrypoints (/plan, /build-fix, /e2e, /orchestrate) that map to the underlying skills system during migration.
• Cross-Platform Harness Support — Full support for Windows, macOS, and Linux with tight integration for Claude Code, Cursor (DRY adapter pattern), OpenCode (plugin system), Codex (TOML config + AGENTS.md), and Antigravity IDE.
• Token Optimization — Recommended settings for model selection, MAX_THINKING_TOKENS, and strategic compaction to reduce costs and prevent agent context overflow.
• Multi-Agent Orchestration — PM2 process management and parallel execution using git worktrees for complex multi-service harness coding workflows.
• MCP Configurations — Pre-built server configs for GitHub, Supabase, Vercel, Railway, Context7, Exa, Playwright, Sequential Thinking, and Memory.
• Selective Install Architecture — Manifest-driven install pipeline with full or per-language targeting (TypeScript, Python, Go, Swift, PHP); state store tracks installed components.
• Dashboard GUI — Desktop application with tabbed interface for Agents, Skills, Commands, Rules, and Settings with dark and light theme toggle.

To get started, install globally with npm i -g ecc-universal, add the plugin in Claude Code with /plugin marketplace add affaan-m/everything-claude-code, or install the GitHub App at github.com/apps/ecc-tools for automated repository analysis.