plan-canvas
Opens plans and HTML artifacts in a local browser canvas where the user annotates elements, chats, and approves or requests changes while the agent waits for structured JSON feedback.
What it does
Open plans and HTML artifacts in a local browser canvas where the human annotates elements, chats, and approves or requests changes without leaving the page. Use when presenting a plan for review, or when feedback like "move this, change that" is easier pointed at than typed.
Quality judgment
Exceptionally thorough: covers the full lifecycle (open/await/reply/end), failure modes (empty-chair listening, user-initiated end, offline Mermaid), environment variables, and anti-patterns with concrete commands and JSON schemas. Triggers are well-scoped with explicit exclusions; only minor deduction for assuming ECC-specific context that may not generalize.