oh‑my‑kimi v1.1.0 – Parallel Coding Teams with Multi‑Agent Orchestration

Oh‑my‑kimi v1.1.0 ships a multi‑agent orchestration harness that wraps the Kimi Code CLI (K2.6) into parallel coding teams. The new CLI commands include `omk chat` for interactive Kimi sessions with resumable context and tmux support, `omk cockpit` for a real‑time dashboard that renders parallel TODOs and agent states, and `omk hud` for a full terminal dashboard that shows goal scoring and ETA estimation. A DAG scheduler has been added to handle retry, skip‑on‑failure, fallback roles, and evidence gates, while agents run in isolated git worktrees to prevent conflicts. Live quality gates enforce lint, typecheck, test, and build before a task can be marked complete, and optional local graph memory is available via Neo4j or Kuzu backends.

The release includes 234 passing tests and a green CI pipeline, and it requires Node 20+, a Kimi API key, and an npm install of @oh‑my‑kimi/cli. The project is MIT licensed and available on GitHub, with npm distribution at https://www.npmjs.com/package/@oh-my-kimi/cli. Users can install the CLI globally with `npm install -g @oh-my-kimi/cli`, run `omk init` to set up a workspace, and launch a parallel coding team with `omk run "Build a Next.js landing page with dark mode and contact form"`. The dashboard tools (`omk cockpit` and `omk hud`) provide real‑time visibility into the agents’ progress and goal scoring, making it easier to monitor and debug the workflow.