feat: Update README and overview for Atlas UI 3; add Cline CLI installation and CLI alias setup

- Changed project title in README from "Chat UI" to "Atlas UI 3".
- Added installation instructions for Cline CLI in the agent setup script.
- Implemented CLI alias setup for 'dclaude' in the agent setup script.
- Removed deprecated agent_setup.sh and agent_update.md files.
- Introduced new agent update plan and detailed architecture changes for the strict Reason–Act–Observe loop.
- Updated file storage manager to format last modified date in ISO format.
- Added comprehensive overview of Atlas UI 3, highlighting its features and architecture.

Author: garland3

README.md

@@ -1,4 +1,4 @@
-# Chat UI
+# Atlas  UI 3
 
 A modern LLM chat interface with MCP (Model Context Protocol) integration.
 

_agent_setup.sh

@@ -19,6 +19,20 @@ if ! command_exists node || ! command_exists npm; then
 fi
 echo "Prerequisites met."
 
+# --- Cline CLI ---
+echo "--- Installing Cline CLI ---"
+if command_exists cline; then
+  echo "Cline CLI is already installed."
+else
+  echo "Installing via npm..."
+  if npm install -g cline; then
+    echo "Cline installed. Run 'cline' to get started."
+  else
+    echo "Failed to install Cline CLI." >&2
+  fi
+fi
+
+
 # --- OpenAI Codex CLI ---
 echo "--- Installing OpenAI Codex CLI ---"
 if command_exists codex; then
@@ -66,4 +80,24 @@ claude mcp add --transport http context7 https://mcp.context7.com/mcp
 echo "--- Installation Complete ---"
 echo "Remember to configure API keys and restart your shell for changes to take effect."
 
+# --- Setting up CLI aliases ---
+echo "--- Setting up CLI aliases ---"
+ALIAS_COMMAND="alias dclaude='claude --dangerously-skip-permissions'"
+
+# Add alias to common shell configuration files
+for shell_config in ~/.bashrc ~/.zshrc ~/.profile; do
+  if [ -f "$shell_config" ]; then
+    if ! grep -q "alias dclaude=" "$shell_config"; then
+      echo "" >> "$shell_config"
+      echo "# Claude CLI shortcut" >> "$shell_config"
+      echo "$ALIAS_COMMAND" >> "$shell_config"
+      echo "Added dclaude alias to $shell_config"
+    else
+      echo "dclaude alias already exists in $shell_config"
+    fi
+  fi
+done
+
+echo "Alias 'dclaude' has been added. Restart your terminal or run 'source ~/.bashrc' (or appropriate shell config) to use it."
+
 

agent_setup.sh

@@ -164,7 +164,7 @@ def organize_files_metadata(self, file_references: Dict[str, Dict[str, Any]]) ->
                 'source': source_type,
                 'source_tool': source_tool,
                 'extension': filename.split('.')[-1] if '.' in filename else '',
-                'last_modified': file_metadata.get("last_modified"),
+                'last_modified': file_metadata.get("last_modified").isoformat() if file_metadata.get("last_modified") else None,
                 'content_type': file_metadata.get("content_type", "application/octet-stream"),
                 'can_display_in_canvas': self.should_display_in_canvas(filename)
             }

backend/modules/file_storage/manager.py

@@ -0,0 +1,5 @@
+# Overview
+
+**Atlas UI 3** is a modern chat interface that lets you interact with multiple AI models (OpenAI GPT, Anthropic Claude, and Google Gemini) through a single, unified interface. What makes it unique is its integration with the Model Context Protocol (MCP), which allows AI assistants to connect to external tools and data sources in real-time. Whether you're looking to have conversations enhanced by database queries, file system access, or custom APIs, this application provides a flexible foundation for building AI-powered workflows. The interface updates dynamically as the AI uses tools, with support for custom HTML rendering and real-time streaming responses.
+
+Under the hood, the project employs clean architecture principles with a FastAPI backend communicating via WebSockets to a React 19 frontend. The backend uses protocol-based dependency injection for loose coupling, implements two distinct agent loop strategies (ReAct and Think-Act) for different reasoning patterns, and integrates LiteLLM for unified model access across providers. The system supports RAG capabilities, S3-compatible file storage, group-based authorization for MCP servers, and comprehensive observability through OpenTelemetry. Configuration management follows a layered approach using Pydantic models with YAML/JSON configs, while the frontend leverages Context API for state management without Redux. The project is actively developed and welcomes contributions - whether you're interested in adding new MCP integrations, improving agent reasoning patterns, enhancing the UI, or expanding LLM provider support, pull requests are encouraged and appreciated.