judgevet.policy_json

Supported JSON decoding facade for typed policies.

The grammar matches CLI policy files. Errors are local PolicyDefinitionError instances without CLI prefixes or exit codes. Callers own file and stdin IO.

Examples:

from judgevet import Noul, NoulAnswer
from judgevet.policy import evaluate_policy
from judgevet.policy_json import parse_policy

policy = parse_policy(
    '{"rules":[{"question":"clear","pass":{"noul":{"min":0.8}}}]}',
    {"clear": Noul(instructions="Clear?")},
)
assert evaluate_policy(policy, {"clear": NoulAnswer(0.9)}).passed