Multi-agent code review with Tech Lead

★ 8.1 · vibe-coding

Open Code Review is a Claude Code skill that runs multi-agent code review by simulating a team of Principal Engineers: a Tech Lead persona orchestrates specialized reviewers covering architecture, code quality, security, and testing, then synthesizes their findings into prioritized, actionable feedback. The review follows an 8-phase workflow — from context discovery and loading `.ocr/config.yaml` to final synthesis evaluated against provided requirements such as PR descriptions, Jira tickets, or spec documents. The reviewer team is resolved dynamically via `ocr team resolve --json` and can be adjusted in natural language, for example "add security focus" or "use 3 principal reviewers". Parallel sub-agent spawning is supported in Claude Code; sequential execution is used in environments without that primitive. Requires git, with optional gh CLI for GitHub integration.