FiddleCube Compliant LLM

FiddleCube's Compliant LLM is a toolkit focused on ensuring the security and compliance of AI systems, which directly aligns with the 'Safety & Alignment Framework' feature. It offers 'Security Testing' against various attack strategies and 'Compliance Analysis' against frameworks like NIST, ISO, GDPR, and HIPAA. The platform supports 'multiple LLM providers via LiteLLM', implying API access. While it doesn't explicitly offer conversational AI or fine-tuning, its focus on testing AI agents and MCP servers suggests an underlying capability to interact with and evaluate LLMs. The GitHub page also mentions 'GitHub Copilot' and 'GitHub Models' which are code generation and AI model management tools, respectively, offered by GitHub itself, not FiddleCube. However, FiddleCube's tool is designed to test AI agents and MCP servers, which are often built using such models and APIs. The 'Enterprise Solutions' feature is supported by their offering of 'Self-hosted or hosted cloud' options and focus on enterprise compliance. The 'Research & Publications' and 'Security & Red Teaming' features are supported by their focus on security testing and compliance, which often involves research into vulnerabilities and red teaming exercises.