v1.0.0 - Multi-LLM Provider Support

What's New

Multi-LLM Provider Support

Rustbucket now supports multiple LLM providers, giving deployers flexibility to choose their preferred AI backend:

Provider	Environment Variable	Default Model
OpenAI	OPENAI_API_KEY	gpt-4o-mini
Anthropic Claude	ANTHROPIC_API_KEY	claude-3-haiku-20240307
Google Gemini	GEMINI_API_KEY	gemini-1.5-flash
Ollama	(none required)	llama3.2

Configuration

[llm]
provider = "openai"  # or "claude", "gemini", "ollama"
model = "gpt-4o-mini"

# For Ollama only:
# ollama_host = "http://localhost:11434"

Additional Features

• Protocol-specific prompts: Customize LLM behavior per protocol (SSH, HTTP, FTP, SMTP)
• Tarpit mode: Progressive response delays to waste attacker time
• Dynamic fingerprinting: Randomized server banners for each instance
• S3 log uploads: Automatic hourly log uploads to S3
• Registry integration: Central management of multiple honeypot instances

Docker

docker pull jamesbinford/rustbucket:latest

Breaking Changes

• CHATGPT_API_KEY still works but OPENAI_API_KEY is now preferred
• Default model changed from gpt-3.5-turbo to gpt-4o-mini

---

Full documentation: README.md