playground UI to test different settings and simulate failures

[gitcommitshow]

a playground UI can help with

Testing and simulation

• testing if everything works as expected
• testing different settings
• simulate and understand behaviour in case of complex failures
• the difference between calling LLM Api with vs without Resilient LLM
• uncovering unpredictable failures and errors

Planning the right prompt and LLM settings

• understanding capabilities and impact of setting changes of an LLM+prompt
• testing different system prompt input vs output
• testing output formats

Testing orchestration and experiencing the new use cases

• testing tool calls
• testing CoT
• choosing the right agentic workflow approach based on the outcomes
• group chat

[gitcommitshow]

This has to be served as a different tool either as a CLI or web server to avoid adding dependencies to the core project