What problem does it solve? When adding log lines, request IDs, failure parsing, or alert logic, developers make inconsistent choices about log levels, secret scrubbing, and error exposure. This Skill loads a constitutional rule that defines exactly how the application must log, record failures, and notify the owner, so every edit follows the same observability law. ## Core Features & Use Cases - Logging rules: Defines request ID propagation, log levels, secret scrubbing by field name, and startup digest requirements. - Failure handling: Specifies single-point error parsing, separation of input/rights failures from breakages, and bounded-queue failure recording that never delays responses. - Alerting semantics: Covers spike detection over closed buckets, schedule-tick analysis, and mail notification fuses. - Use Case: When adding a new log line to the message-bus receiver, load this rule to decide the correct level, ensure fields are scrubbed, and check whether the line belongs in the failure store. ## Quick Start Load the observability rule before editing the logger, request context, or failures domain and follow its articles when adding a new log line.
