Summary
Allow users to create custom rule profiles with configurable severity thresholds.
What to implement
1. Rule Profiles
# .aitrust.yaml
profiles:
strict:
min_score: 90
fail_on: [critical, high, medium]
standard:
min_score: 70
fail_on: [critical, high]
relaxed:
min_score: 50
fail_on: [critical]
2. CLI Usage
aitrust validate src/ --profile strict
aitrust validate src/ --profile standard # default
3. Per-Category Thresholds
thresholds:
security:
fail_below: 80 # Security must score 80+
hallucination:
fail_below: 70
logic:
fail_below: 60
best_practices:
fail_below: 50
4. Custom Severity Override
overrides:
# Promote bare-except from medium to high in this project
bare_except: high
# Demote mutable-default from medium to low
mutable_default: low
Acceptance Criteria
Difficulty
Beginner — configuration handling, the validation engine already supports scoring.
Summary
Allow users to create custom rule profiles with configurable severity thresholds.
What to implement
1. Rule Profiles
2. CLI Usage
aitrust validate src/ --profile strict aitrust validate src/ --profile standard # default3. Per-Category Thresholds
4. Custom Severity Override
Acceptance Criteria
Difficulty
Beginner — configuration handling, the validation engine already supports scoring.