Generate and evaluate agent skills from traces before shipping them into repeatable production workflows with UPskill
Turn successful traces into reusable skills, then benchmark those skills across models before you trust them in production.
npx skills add agentskillexchange/skills --skill generate-and-evaluate-agent-skills-from-traces-before-shipping-them-into-repeatable-production-workflows-with-upskill
# or run without global install
uvx upskill
UPskill is publishable because the user-facing job is specific: distill a repeatable agent skill from successful traces, evaluate it, and compare model performance before rollout. Use it when a team keeps solving the same class of task and wants to convert that pattern into a tested reusable skill instead of relying on prompt memory or ad hoc coaching.
What this skill actually does
Invoke it instead of using a model or framework normally when the operator needs a generate-then-evaluate loop around skill creation itself. The scope boundary is clear: skill synthesis, evaluation, and benchmarking from traces. It is not a generic model framework listing and not just a broad eval product card.