BAD PR 1 (DO NOT MERGE): regress dv-data SKILL bulk-create guidance - #73
Closed
saurabhrb wants to merge 6 commits into
Closed
BAD PR 1 (DO NOT MERGE): regress dv-data SKILL bulk-create guidance#73saurabhrb wants to merge 6 commits into
saurabhrb wants to merge 6 commits into
Azure Pipelines / DVSkillsPlugin-Evals-PR
failed
Jun 2, 2026 in 7m 35s
Build #20260602.12 • dry-run • plugin-main • dv_data.biceval.json had test failures
Details
- Failed: 1 (33.33%)
- Passed: 2 (66.67%)
- Other: 0 (0.00%)
- Total: 3
Annotations
Check failure on line 577 in Build log
azure-pipelines / DVSkillsPlugin-Evals-PR
Build log #L577
Test pass rate 66.7% is below 100% threshold. Failing pipeline.
Check failure on line 579 in Build log
azure-pipelines / DVSkillsPlugin-Evals-PR
Build log #L579
Script failed with exit code: 1
Check failure on line 1 in data.data_003_skill_contract
azure-pipelines / DVSkillsPlugin-Evals-PR
data.data_003_skill_contract
1 evaluator(s) failed:
[P1] CortexConfigurations:Common/SEVAL/LMChecklist.prompty: score=0 < threshold=1
Reasoning: Assertion=[NOT_CONTAINS: one HTTP call per record]: The AI's response explicitly includes the quoted phrase 'one HTTP call per record' under the 'Creating Many Records (Loop pattern)' section. Therefore, the assertion ('NOT_CONTAINS: one HTTP call per record') fails, as the phrase is clearly present in the provided response.
Loading