A model-driven product end to end: chat workbench/sidebar/search, the generation studio, autonomous agents, the agent workspace, the AI admin console, plus the backend router, vector search, and MCP scaffolding that power them.
npx rahman-resources add <slug>A model-driven app surface, env-free in the demo: the ChatGPT-style chat workbench, the generation studio, and the local-first assistant with user-built agents & skills.
npx rahman-resources add ai-workspacenpx rahman-resources add assistantOrchestration and the operator console: autonomous multi-step agents and the AI admin (providers, skills, models, moderation).
npx rahman-resources add ai-adminFramework-neutral compile-time contract for a future AI administration surface. Ships provider/model/skill/tool/agent/budget/audit types plus tab, icon-name, and permission taxonomy; intentionally no renderer, persistence backend, or runtime framework dependency.
The infra behind the surfaces — each previews env-free but needs a key to go live: the tiered model router, semantic vector search, and the MCP token scaffolding that exposes your tools to ChatGPT/Claude.
npx rahman-resources add ai-routerTry with your key — add the slice, set OPENROUTER_API_KEY, and the live action runs in your own backend.
npx rahman-resources add vector-searchFramework-neutral semantic-search adapter contract. The host injects VectorSearchCtx for query/index/reindex and owns the actual vector backend, embeddings, credentials, persistence, authorization, and reindex policy; RR ships no renderer or Convex schema for this slice.
Try with your key — add the slice, set OPENAI_API_KEY + a Convex vector index, and the live action runs in your own backend.
npx rahman-resources add create-your-mcpTry with your key — add the slice, set MCP_API_KEY + an OAuth host allowlist, and the live action runs in your own backend.