What problem does it solve? Producing a video that spans multiple styles and tools (ASCII, Manim, generative art, AI image-to-video, voiceover, editing) requires coordinating many specialized agents, and doing it ad hoc leads to inconsistent briefs, mismatched teams, and stalled tasks. This Skill turns any video request into a structured, monitored multi-agent production pipeline backed by Hermes Kanban. ## Core Features & Use Cases - Adaptive discovery and briefing: Asks targeted per-style intake questions (narrative, marketing, music video, explainer, ASCII, installation) and produces a confirmed brief.md contract before any work starts. - Team design and automated setup: Composes a role-based team (director, writer, renderers, voice-talent, editor, reviewer) and generates a setup.sh that creates Hermes profiles, SOUL.md files, workspace, and the initial kanban task via scripts/bootstrap_pipeline.py. - Execution monitoring and intervention: Polls the kanban with scripts/monitor.py to detect stuck, overtime, or flapping tasks and provides intervention recipes (comments, re-run tasks, dependency rewiring). - Use Case: A user asks for a 3-minute music video synced to a provided track; the Skill scopes the brief, designs a team with a music-supervisor producing beats.json plus ASCII and p5.js renderers, bootstraps the kanban, and helps monitor it to a finished final.mp4. ## Quick Start Ask the agent to plan and set up a kanban-based production pipeline for your video idea, starting with its length, style, and target platform.
