Primary question: Do you need a standardized skill set and security layer across multiple AI coding harnesses?
RepoDaily adoption score
RepoDaily rates this as 91/100 (strong) for adoption: evidence, installation path, production risk, differentiation, license clarity, and AI/agent fit are scored from the article sources and adoption notes.
5 source(s) across 3 source category/categories, plus a RepoDaily-specific evidence module when available.
5 workflow step(s), 4 next-action step(s), and 3 command/install signal(s) were detected.
Trending momentum is +508 stars, with maintenance/release/issue signals counted when present.
Risk is marked medium, with 5 security note(s) and 5 explicit skip condition(s).
3 opportunity lens item(s), 3 alternative(s), and 3 type-specific section(s) support differentiation.
License source or license wording is present.
8 AI/agent-related signal(s) were detected in the article text and metadata.
Project overview
ECC (Everything Claude Code) is a harness-native operator system designed to optimize agent performance across multiple coding environments including Claude Code, Cursor, Codex, and OpenCode. Rather than simple configuration files, ECC provides a complete system comprising skills, instincts, memory optimization, continuous learning capabilities, security scanning, and research-first development protocols.
The project has gained significant traction with over 211,000 GitHub stars and 230+ contributors. It supports 12+ language ecosystems including TypeScript, Python, Go, Java, and more. ECC operates as a cross-harness substrate, enabling consistent workflows and skill reuse regardless of the underlying AI coding tool being used.
The system is distributed through multiple channels: the npm packages `ecc-universal` and `ecc-agentshield`, a GitHub App, a plugin system (slug `ecc@ecc`), and the official website ecc.tools. The project strictly enforces official distribution channels to prevent security risks from unofficial mirrors.
Why it is trending now
- Reached 211,900+ GitHub stars with 32,500+ forks, indicating massive community adoption
- Released version 2.0.0 stable with significant updates including Discord community integration and cross-harness orchestration
- Expanded to 156+ skills, 38 agents, and 72 commands covering diverse development workflows
- Added MCP inventory system and dynamic workflow team orchestration with the `orch-*` skill family
Problem it solves
- AI coding harnesses like Claude Code and Cursor lack standardized skill libraries and workflow definitions across different environments
- Security scanning and vulnerability detection are often missing or inconsistent in agentic coding workflows
- Memory optimization and context management for AI agents require manual configuration without unified tooling
- Research-first development and continuous learning patterns are difficult to implement consistently across AI-powered development tools
How it works
- Install ECC via the GitHub App, npm packages (`ecc-universal`, `ecc-agentshield`), or plugin slug `ecc@ecc` into your development environment
- Load specific skills from the library of 156+ available modules that cover languages, frameworks, workflows, and domain knowledge
- Activate security scanning through AgentShield (`ecc-agentshield`) to monitor for vulnerabilities in generated code and dependencies
- Utilize memory optimization and cross-harness workflows to maintain context across Claude Code, Cursor, Codex, and OpenCode sessions
- Implement orchestrator skills (`orch-*`) for dynamic workflow team orchestration and automated task delegation
Product demo and interface preview
Integration Surface
ECC integrates with multiple AI coding harnesses through a unified plugin and package system. The primary integration methods include the GitHub App installation, npm packages `ecc-universal` and `ecc-agentshield`, and the plugin slug `ecc@ecc` for harness-specific loading. The system supports Shell, TypeScript, Python, Go, Java, Perl, and Markdown environments.
The recently released 2.0.0 version introduces cross-harness compatibility specifically targeting Claude Code, Codex, Cursor, OpenCode, and Gemini. MCP (Model Context Protocol) connectors are supported but have been streamlined from six defaults to a single `chrome-devtools` connector per the new connector policy established in the June 2026 audit. Retired connectors like `github`, `context7`, `exa`, `memory`, `playwright`, and `sequential-thinking` are now covered by skills wrapping CLIs/REST APIs or harness-native features.
Command Surface
ECC exposes 72 commands across various workflow categories including deployment, testing, code generation, and security scanning. Commands are invoked through slash commands in supported harnesses and can be extended through the contribution system.
For the ECC 2.0 alpha control plane, the TUI interface (`ecc-tui`) currently exposes dashboard management (`dashboard`), session control (`start`, `stop`, `resume`), session listing (`sessions`), and status checking (`status`). The binary builds locally using `cargo build --manifest-path ecc2/Cargo.toml`.
Adoption Checklist
- Verify official installation channels only: GitHub repository, npm packages `ecc-universal`/`ecc-agentshield`, GitHub App, plugin slug `ecc@ecc`, or ecc.tools website
- Check Node.js version compatibility (plugin hooks fix addresses Node 21+ `require.main` undefined issues)
- Review the 156+ available skills and identify which match your team's workflow patterns
- Configure MCP connectors via `mcp-configs/mcp-servers.json` if opting into connectors beyond the default `chrome-devtools`
- Set up Discord integration for community updates and release notifications using the dependency-free community bot
- Review security policy and vulnerability reporting process for the supported versions (2.x/rc and 1.10.x)
Who should pay attention?
Good fit if
- Development teams actively using Claude Code, Cursor, Codex, or OpenCode in production environments
- Organizations requiring standardized security scanning and vulnerability detection for AI-generated code
- Teams managing multiple AI coding harnesses that need consistent skill libraries and workflow definitions
- Projects benefiting from research-first development patterns and continuous learning integration
- Development shops needing cross-language support (TypeScript, Python, Go, Java, etc.) in AI-assisted coding
Skip for now if
- Teams not using any of the supported AI coding harnesses (Claude Code, Cursor, Codex, OpenCode, Gemini)
- Projects requiring complete isolation from external package ecosystems or npm dependencies
- Organizations with strict policies against community-maintained security tools in AI workflows
- Teams preferring minimal configuration over comprehensive workflow orchestration systems
- Developers working exclusively with traditional IDEs without AI coding assistant integration
Risks and cautions
ECC requires careful attention to official distribution channels to avoid security risks from unofficial mirrors. The project maintains active support for version 2.x/rc and 1.10.x, with critical fixes only for 1.9.x. The recent MCP connector policy change may require workflow adjustments for users dependent on the retired default connectors.
- Security policy explicitly warns against unofficial packages and mirrors, listing `@chil_ntl/ecc-cli` and `ecc-100xprompt-plugin` as unverified
- Version 2.0.0 introduced significant changes including MCP connector reduction from six defaults to one, requiring configuration updates
- Plugin hooks required fixes for Node 21+ compatibility, indicating potential environment-specific issues
- ECC 2.0 alpha control-plane remains in development state (`ecc2/` directory) separate from the stable release surface
- Security fixes land on `main` first with backports on a best-effort basis for supported release lines (2.x/rc and 1.10.x)
- Vulnerability reporting should use GitHub private vulnerability reporting or email to `affaan@ecc.tools` (not `security@ecc.tools`)
- Expected response times: acknowledgment within 48 hours, initial assessment within 7 days, critical fix target within 14 days
- Official distribution surfaces are strictly limited to GitHub repo, npm packages `ecc-universal`/`ecc-agentshield`, GitHub App, plugin slug `ecc@ecc`, and ecc.tools website
- Supply-chain rules require GitHub Actions to use pinned commit SHAs for third-party actions to prevent dependency confusion attacks
Alternatives to compare
| Approach | When to use | Trade-off |
|---|---|---|
Native Harness Configurations | When minimal setup is preferred and cross-harness consistency is not required | Free |
Custom Prompt Libraries | When teams have highly specialized requirements not covered by general skill libraries | Development time |
AgentShield (Standalone) | When only security scanning is needed without the full ECC workflow system | Free |
What this trend reveals
Standardization Across Teams
ECC provides a shared vocabulary and skill library that can standardize AI-assisted coding practices across large development organizations, reducing inconsistency between teams using different AI tools.
The project supports 12+ language ecosystems and provides 156+ skills covering language standards, framework patterns, workflows, and domain knowledge.
Security Layer Integration
The integration of AgentShield provides automated vulnerability scanning for AI-generated code, addressing a significant gap in current agentic development workflows.
The `ecc-agentshield` npm package provides dedicated security scanning capabilities with documented vulnerability reporting processes.
Cross-Harness Portability
Skills and workflows defined in ECC can be reused across Claude Code, Cursor, Codex, OpenCode, and Gemini, reducing lock-in to any single AI coding platform.
Version 2.0.0 introduced cross-harness substrate documentation and adapters for codex-worktree and opencode sessions.
RepoDaily verdict
ECC represents the most comprehensive effort to standardize and optimize agentic coding workflows across multiple harnesses. With its extensive skill library, integrated security scanning, and cross-platform support, it addresses critical gaps in AI-assisted development tooling. Teams should evaluate ECC if they use Claude Code, Cursor, Codex, or OpenCode and need consistent workflows, security scanning, and reusable skill definitions across environments. However, careful attention must be paid to official installation channels to avoid security risks from unofficial mirrors.