Glm 4.7 testing

.env.example

@@ -0,0 +1,38 @@
+# =============================================================================
+# Simple Agent Configuration
+# =============================================================================
+# Copy this file to .env and fill in your values
+
+# -----------------------------------------------------------------------------
+# LLM API Configuration
+# -----------------------------------------------------------------------------
+# Base URL for the OpenAI-compatible API endpoint
+API_BASE_URL=http://localhost:8080/v1
+
+# API key (set to "not-needed" if your endpoint doesn't require authentication)
+API_KEY=not-needed
+
+# Model name to use
+MODEL_NAME=gpt-4
+
+# -----------------------------------------------------------------------------
+# Tavily API (Web Search & Extraction)
+# -----------------------------------------------------------------------------
+# Get your API key from https://tavily.com
+TAVILY_API_KEY=tvly-your-api-key-here
+
+# Optional: Custom Tavily endpoints (defaults shown)
+# TAVILY_SEARCH_URL=https://api.tavily.com/search
+# TAVILY_EXTRACT_URL=https://api.tavily.com/extract
+
+# -----------------------------------------------------------------------------
+# Agent Settings
+# -----------------------------------------------------------------------------
+# Maximum number of iterations before stopping (default: 40)
+MAX_ITERATIONS=40
+
+# Command execution timeout in seconds (default: 60)
+COMMAND_TIMEOUT=60
+
+# Maximum content length for webpage extraction (default: 5000)
+MAX_CONTENT_LENGTH=5000

.gitignore

@@ -1,10 +1,89 @@
-# Python-generated files
+# =============================================================================
+# Simple Agent - Git Ignore
+# =============================================================================
+
+# -----------------------------------------------------------------------------
+# Environment
+# -----------------------------------------------------------------------------
+.env
+.env.local
+.env.*.local
+
+# -----------------------------------------------------------------------------
+# Python
+# -----------------------------------------------------------------------------
 __pycache__/
-*.py[oc]
+*.py[cod]
+*$py.class
+*.so
+.Python
 build/
+develop-eggs/
 dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
 wheels/
-*.egg-info
+*.egg-info/
+.installed.cfg
+*.egg
 
 # Virtual environments
-.venv
+.venv/
+venv/
+ENV/
+env/
+
+# -----------------------------------------------------------------------------
+# IDE
+# -----------------------------------------------------------------------------
+.idea/
+.vscode/
+*.swp
+*.swo
+*~
+.DS_Store
+
+# -----------------------------------------------------------------------------
+# Project Specific
+# -----------------------------------------------------------------------------
+# Generated assets (uncomment to track)
+# assets/
+
+# Output logs
+output/
+
+# Batch run logs
+batch_output*.log
+
+# LaTeX auxiliary files
+*.aux
+*.log
+*.out
+*.toc
+*.lof
+*.lot
+*.fls
+*.fdb_latexmk
+*.synctex.gz
+
+# -----------------------------------------------------------------------------
+# Testing
+# -----------------------------------------------------------------------------
+.pytest_cache/
+.coverage
+htmlcov/
+.tox/
+.nox/
+
+# -----------------------------------------------------------------------------
+# Misc
+# -----------------------------------------------------------------------------
+*.bak
+*.tmp
+*.temp

README.md

@@ -7,7 +7,8 @@ An experimental AI agent leveraging state-of-the-art LLM models with a focus on
 Agent S demonstrates that modern LLM models can achieve sophisticated agentic capabilities through:
 - Carefully crafted prompting
 - Direct Unix tool integration
-- Console-based I/O instead of native function calling
+- Function calling with persistent Python REPL
+- Web search and content extraction
 - Continuous learning loops
 
 ## Philosophy
@@ -19,19 +20,35 @@ The project is built on the hypothesis that reasoning-focused LLM models, when p
 This project uses [uv](https://docs.astral.sh/uv/) as its package manager.
 
 ```bash
-uv run main.py
+uv run python main.py
+```
+
+Run with predefined tasks:
+```bash
+uv run python main.py --list              # List available tasks
+uv run python main.py --task ai_industry  # Run specific task
+uv run python main.py --custom "Your task here"
 ```
 
 ## Environment Setup
 
-Create a `.env` file in the root directory with the following API keys:
+Create a `.env` file in the root directory (see `.env.example`):
 
 ```env
-OPENAI_API_KEY=your_openai_key
-DEEPSEEK_API_KEY=your_deepseek_key
-SERPER_API_KEY=your_serper_key
+API_BASE_URL=http://localhost:8080/v1
+API_KEY=your_api_key
+MODEL_NAME=gpt-4
+TAVILY_API_KEY=your_tavily_key
 ```
 
+## Assets
+
+The `assets/` folder contains example outputs from pipeline runs, including:
+- Generated charts and visualizations (PNG)
+- Data files (CSV)
+- LaTeX reports and compiled PDFs
+- Industry analysis reports (AI, Clean Energy, Pharma, etc.)
+
 ## License
 
-MIT
+MIT