c25. AI Ecosystem Inventory Map

Create a new qmap cycle for an inventory/index view of the company-wide AI ecosystem. This is not a single-session surface. The center of the map should be the active agents within a selected timeframe, with surrounding repositories, then files per repository, plus MCP servers, tools, and sub-agents.

Build a dedicated wrapper component, QMapInventoryIndex, and start with qmap's existing actor types rather than inventing new meshes immediately:

• AI agents -> agent
• repositories -> app
• files -> file
• MCP servers -> mcp-server
• tools and sub-agents -> skill

Use tags to distinguish semantic families:

• entityType: 'AI Agent' | 'Repository' | 'File' | 'MCP Server' | 'Tool' | 'Sub-Agent'

Prove the view with deterministic synthetic data first. The scenario pack should include both tiny demo conditions and very large enterprise stress cases. Validate at least these grouping modes:

• Entity Type
• Repository
• Team
• Business Unit
• Flat

Create a runnable experiment, add compact scenario stats, and call out whether repository/file density forces summarization or filtering before new visual primitives are warranted.