Initial commit: Agentic OS troubleshooting platform

Self-hosted, Docker-based agentic troubleshooting platform: FastAPI backend +
LangGraph agent, Next.js UI, tiered LLM routing (local Ollama -> Gemini ->
DeepSeek -> OpenRouter), MCP server manager, encrypted device credentials,
RBAC, audit log, project-memory + Obsidian integrations, and editable
troubleshooting decision rules tuned for the GeneseasX vessel stack.

Co-authored-by: Cursor <cursoragent@cursor.com>

litellm/config.yaml

@@ -0,0 +1,45 @@
+# LiteLLM gateway: one OpenAI-compatible endpoint fronting local + cloud models.
+# The backend references the `model_name` aliases (local-triage / cloud-reason).
+
+model_list:
+  # Local triage model served by Ollama on the host (Apple Silicon).
+  - model_name: local-triage
+    litellm_params:
+      model: ollama_chat/qwen2.5:7b-instruct
+      api_base: os.environ/OLLAMA_BASE_URL
+
+  # Google Gemini (Google AI Studio free tier eligible).
+  - model_name: gemini-economy
+    litellm_params:
+      model: gemini/gemini-2.5-flash
+      api_key: os.environ/GEMINI_API_KEY
+
+  - model_name: gemini-premium
+    litellm_params:
+      model: gemini/gemini-2.5-pro
+      api_key: os.environ/GEMINI_API_KEY
+
+  # Cloud reasoning model (deep troubleshooting) via OpenRouter.
+  - model_name: cloud-reason
+    litellm_params:
+      model: openrouter/anthropic/claude-3.7-sonnet
+      api_key: os.environ/OPENROUTER_API_KEY
+
+  # Direct DeepSeek option.
+  - model_name: deepseek-reason
+    litellm_params:
+      model: deepseek/deepseek-chat
+      api_key: os.environ/DEEPSEEK_API_KEY
+
+  # DeepSeek via OpenRouter (alternative routing).
+  - model_name: openrouter-deepseek
+    litellm_params:
+      model: openrouter/deepseek/deepseek-chat
+      api_key: os.environ/OPENROUTER_API_KEY
+
+general_settings:
+  master_key: os.environ/LITELLM_MASTER_KEY
+
+litellm_settings:
+  drop_params: true
+  request_timeout: 120