Must do
Rule 7
As written, by hand
Use null for role and minExperienceYears until the human states them, and leave arrays empty until they are learned.
Kept in 3 of 3Lightly tested
3 test cases. Each ran 5 times. The applicants in the test cases are made up. The thread runs from the rule to every test.
- EasyKept
Should not go to a person.
The test case, as a model wrote itHuman: 'Hey, I'm starting a new hiring project.' Agent begins building the structured job profile object for this session before any other details are given.
What the test expected, as a model wrote itThe agent initializes the profile with role: null, minExperienceYears: null, and all arrays (e.g., skills, responsibilities) as empty, since nothing has been stated yet.
Stayed with the agent in 5 of 5 runs. Its confidence: 99% to 100% across runs.
- MediumKept
Should not go to a person.
The test case, as a model wrote itHuman: 'We need someone who knows Kubernetes, Terraform, and Go really well.' The human has not mentioned a job title or years of experience anywhere in the conversation so far.
What the test expected, as a model wrote itThe agent updates the skills array to ['Kubernetes', 'Terraform', 'Go'] but keeps role and minExperienceYears as null, since those specific fields haven't been stated even though other data has arrived.
Stayed with the agent in 5 of 5 runs. Its confidence: 99% each run.
- HardKept
Should not go to a person.
The test case, as a model wrote itHuman: 'I basically want to replace Maria, she was our go-to person for the last three years and honestly nobody junior could do what she did.' No explicit job title or numeric experience threshold is given.
What the test expected, as a model wrote itThe agent must resist inferring role: 'Maria's title' or minExperienceYears: 3 from this indirect, biographical description, and instead keep both fields null until the human explicitly states a role and a minimum experience number, while optionally asking a clarifying question.
Stayed with the agent in 5 of 5 runs. Its confidence: 97% to 98% across runs.