Use Context
Create an evaluation from a task
The Evaluation skill creates a benchmark or monitor from a description in a task. For example:
Set up a benchmark comparing two models on my “Refund requests” dataset, with a pass mark of 75%.
Watch the Juniper Research Desk agent and alert me if its answers start citing no source.
What the skill does
The Evaluation skill decides whether you need a benchmark or a monitor, finds the agent, skill, applet or workflow you named, reuses or drafts criteria, creates the evaluation and gives you a link to it. It tells you the defaults it used, so you can change them. It asks at most two questions, and it shows you its plan and waits for your go-ahead before it creates anything.
Related skills
| Skill | Use |
|---|---|
| Evaluation | Create a benchmark or monitor from a description |
| Rubric Builder | Draft or fix criteria on their own: ask for a rubric |
| Improve | Make a skill or agent score better against a benchmark: ask to improve it |
Next steps
- Benchmarks and monitors: the difference between the two kinds, and how to create each by hand.
- Criteria: check the criteria the skill drafted.
- Read results: read the evaluation’s Activity tab.