ECC

ECC is an agentic engineering toolkit that acts as an operating layer for AI coding agents. It gives agents a shared way to work across tools through skills, hooks, a planning canvas, unified memory, compute workflows and a security scanner called AgentShield. The toolkit is open source and installs natively as a plugin in Claude Code and Codex, through managed adapters in Cursor and OpenCode, and through project setup in Kimi Code. It is aimed at developers and engineering teams that run several coding agents and want consistent practices, memory and security checks across them. A guided installer sets it up from the command line. The ECC GitHub App adds repository level analysis: it reads repository history, opens pull requests triggered from issue comments, runs audits on pull requests and large pushes, and uses AgentShield to scan agent configuration and harness files for security problems. The open source toolkit stays free. The GitHub App has a Free plan at $0 covering public repositories only, with 10 analyses a month and 200 commits per run. Pro costs $19 per active developer seat per month or $190 per seat per year, and adds private repository analysis, 50 analyses per seat pooled across the installation, optional metered overage, 1,000 commits per run, AgentShield scanning, automatic triggers and priority support. Enterprise is quote based and adds unlimited analyses, 5,000 commits per run, SSO ready governance, audit logs, custom rules and policy packs, and dedicated onboarding with an SLA. The site publishes its acquisition figures with a documented methodology and notes where counts overlap or cannot be verified, which is useful context when judging adoption claims.