evals-start
Find the right workflow for your AI evaluations.
What it does
Routes evaluation requests to the appropriate specialized skill and its workflow.
When to use it
Use for general evaluation help, users who do not know where to begin, or requests without a more specific matching skill.
How to use it
Give it an evaluation request; it explains which specialized skill fits and loads that skill to follow its workflow.
What you need · 8
Routes trace-based failure discovery without an established taxonomy to this skill, also after generating missing traces; an installation requirement is unstated.
Routes questions about whether an existing evaluation pipeline can be trusted to this skill; an installation requirement is unstated.
Routes known failure modes that code can check to this skill; an installation requirement is unstated.
Routes requests for an LLM judge of a known failure mode to this skill; an installation requirement is unstated.
Access · 0
No additional actions listed.