Santazuki

Routes images to Mimo/OpenAI vision API for text-only models. Use this skill when the user sends an image, asks "what's in this picture", says "analyze this screenshot", requests "OCR" or "extract text", reviews UI designs, reads charts, or uses Chinese triggers like 识别图片/看图. Self-healing setup on first run. Does NOT handle video, audio, or PDFs.

Use when starting any multi-step development task that benefits from structured multi-agent collaboration. Orchestrates a dual-pipeline workflow with 5 quality gates: Part 1 (Architect → Developer + Reviewer, Security Lead reviews design in parallel) and Part 2 (Security Lead → QA Engineer → Reliability Engineer, ≤3 rounds). Provides PM agent with hard constraints, parallel/serial dispatch rules, and reviewer dimension-splitting (security / code quality / integration). Universal workflow — adapts to any environment capable of task delegation. Triggers: "多agent", "devflow". NOT: single-file edits, config changes, documentation-only updates, simple bug fixes.