ECC

ECC adds reusable engineering workflows to AI coding assistants. It's for developers who want editable plans, test-first development and separate review contexts across sessions.

• Planning and testing: Workflows produce plans before implementation and guide a gated test-first development cycle backed by evidence.
• Independent review: Scoped agents separate planning, implementation and review, with a fresh review context to look for regressions and missed problems.
• Shared memory: Memory Vault keeps durable context and handoffs in local, inspectable Markdown documents that Claude, Codex, Hermes, OpenClaw and Kimi can share.
• Configuration checks: AgentShield scans the coding agent's configuration, while hooks can enforce checks outside the model's prompt.
• Component browser: A desktop dashboard lets you search agents, skills, commands and rules.

ECC is open source under the MIT license. It works through the selected agent's model connection, including cloud providers, compatible custom endpoints and self-hosted models. The agent and endpoint determine where inference runs; ECC doesn't serve the models itself.

Memory Vault separates project, team and personal context. Skill-only, minimal, manual and Claude plugin installations require a separate npm runtime for vault access. Stored memories remain unreviewed context, rather than executable policy. Adapter capabilities vary: the managed Kimi Code adapter doesn't configure hooks.