Skip to content

Latest commit

 

History

History
58 lines (47 loc) · 1.97 KB

File metadata and controls

58 lines (47 loc) · 1.97 KB

🚀 S16 NetworkX: The Modern Agent Architecture

Welcome to the new NetworkX-First implementation. This version replaces the manual graph management with a robust DAG structure and integrates modern tools.

🏗️ Architecture

1. The Core (Graph Engine)

  • File: core/loop.py
  • Logic: Uses networkx.DiGraph to manage the plan.
  • Context: Strictly isolated. Node B only sees Node A's output if A -> B.
  • Summarizer: Special agent with Global Read Access to synthsize everything.

2. The Tools (Hybrid Stack)

  • Browser: Hybrid Mode.
    • Fast Search: duckduckgo (Text).
    • Deep Action: browser-use (Vision).
  • Memory: Dual Layer.
    • Short-term: Session Context (NetworkX attributes).
    • Long-term: mem0 (User Profile, Local).
  • Sandbox: e2b capable (currently wrapped in tools/sandbox.py).

3. Folder Structure

16_NetworkX/
├── agents/             # AgentRunners (Logic)
├── config/             # YAML Configs + Models
├── core/               # Main Loop & Utils
├── memory/             # Context & Mem0 Store
├── mcp_servers/        # Tool Servers (Browser, RAG)
├── prompts/            # System Prompts (Markdown)
├── tools/              # Helper Scripts
└── app.py              # Entry Point

🏃‍♂️ How to Run

Interactive Mode

uv run 16_NetworkX/app.py

Type your query (e.g., "Plan a 3 day trip to Tokyo").

Automated Test

uv run 16_NetworkX/test_run.py

🛠️ Configuration

  • Agents: config/agent_config.yaml maps Agents -> Prompts -> Tools.
  • Models: config/models.json defines LLMs (Gemini, Ollama).
  • Prompts: Edit prompts/*.md to change agent behavior.

🔮 Next Steps (Roadmap)

  1. UI: Connect ui/visualizer.py to a React Flow frontend.
  2. E2B: Fully replace tools/sandbox.py with E2B SDK.
  3. Mem0: Enable active learning in SummarizerAgent.