You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: .github/skills/azure-typespec-author/evaluate/evals/001001.eval.yaml
+78-40Lines changed: 78 additions & 40 deletions
Original file line number
Diff line number
Diff line change
@@ -16,61 +16,99 @@ config:
16
16
# Test cases
17
17
stimuli:
18
18
- name: 001001-version-spread-property
19
-
prompt: I added `...Azure.ResourceManager.ManagedServiceIdentityProperty;` which updates all my existing API versions and introduces a breaking change. I want to introduce the properties of the spread model '...ManagedServiceIdentityProperty' in API version 2025-05-04-preview only.
20
-
environment:
19
+
tags:
20
+
- 'main'
21
+
prompt: 'I added `...Azure.ResourceManager.ManagedServiceIdentityProperty;` which updates all my existing API versions and introduces a breaking change. I want to introduce the properties of the spread model ''...ManagedServiceIdentityProperty'' in API version 2025-05-04-preview only.'
prompt: Verify changes are scoped to this task only, with no unrelated edits beyond introducing preview-only identity via versioning.
64
-
model: claude-opus-4.6
65
-
scoring: scale_1_5
66
-
threshold: 1.0
65
+
prompt: 'Verify changes are scoped to this task only, with no unrelated edits beyond introducing preview-only identity via versioning.'
66
+
model: 'claude-opus-4.6'
67
+
threshold: 1
68
+
scoring: 'scale_1_5'
69
+
constraints:
70
+
max_turns: 5
71
+
max_tokens: 50000
72
+
- name: 001001-version-spread-property-trigger
73
+
tags:
74
+
- 'trigger'
75
+
prompt: 'I added `...Azure.ResourceManager.ManagedServiceIdentityProperty;` which updates all my existing API versions and introduces a breaking change. I want to introduce the properties of the spread model ''...ManagedServiceIdentityProperty'' in API version 2025-05-04-preview only.'
76
+
environment: *shared_environment
77
+
graders:
78
+
- type: 'tool-calls'
79
+
config:
80
+
required:
81
+
- 'edit'
82
+
- 'azure-sdk-mcp-azsdk_run_typespec_validation'
83
+
- type: 'skill-invocation'
84
+
config:
85
+
required:
86
+
- 'azure-typespec-author'
87
+
constraints:
88
+
max_turns: 5
89
+
max_tokens: 50000
90
+
- name: 001001-version-spread-property-forced
91
+
tags:
92
+
- 'forced'
93
+
prompt: '@azure-typespec-author I added `...Azure.ResourceManager.ManagedServiceIdentityProperty;` which updates all my existing API versions and introduces a breaking change. I want to introduce the properties of the spread model ''...ManagedServiceIdentityProperty'' in API version 2025-05-04-preview only.'
prompt: 'Verify change scope and quality only: add a default value of 21 for the `age` property in `EmployeeProperties` that only applies to the 2025-11-01 API version. The old age renamed and removed for the new version.'
74
-
model: claude-opus-4.6
75
+
constraints:
76
+
max_turns: 6
77
+
max_tokens: 100000
78
+
- name: 001002-version-default-value-trigger
79
+
tags:
80
+
- 'trigger'
81
+
prompt: |
82
+
add a default value `21` for property age in model EmployeeProperties only for the new api version 2025-11-01
83
+
84
+
environment: *shared_environment
85
+
graders:
86
+
- type: 'tool-calls'
87
+
config:
88
+
required:
89
+
- 'edit'
90
+
- 'azure-sdk-mcp-azsdk_run_typespec_validation'
91
+
- type: 'skill-invocation'
92
+
config:
93
+
required:
94
+
- 'azure-typespec-author'
95
+
constraints:
96
+
max_turns: 5
97
+
max_tokens: 50000
98
+
- name: 001002-version-default-value-forced
99
+
tags:
100
+
- 'forced'
101
+
prompt: |
102
+
@azure-typespec-author add a default value `21` for property age in model EmployeeProperties only for the new api version 2025-11-01
0 commit comments