safe-by-default-architect

What problem does it solve?

This Skill prevents recurring security and reliability defects by turning repeated dangerous implementation patterns into safe architectural defaults, forbidden-to-safe mappings, common layers, and enforceable static rules.

Core Features & Use Cases

• Safe Pattern Catalog Creation: Converts incident/RCA evidence into a prioritized set of approved safe patterns with rationale.
• Forbidden Pattern List & Taxonomy: Produces a structured inventory of anti-patterns mapped to concrete danger categories (e.g., injection, silent corruption, environment divergence).
• Common Layer Design: Recommends shared abstractions (services, repositories, wrappers) that make unsafe behavior hard to implement.
• Static Rule Candidate Plan: Proposes lint/semgrep/regex/static-analysis rules with false-positive risk notes and enforcement severity/rollout guidance.
• Exception Governance & Review Addendum: Defines when deviations are allowed (review-required vs approval-required vs prohibited) and adds checklist items to align human review with the new standards.

Quick Start

Use the safe-by-default-architect skill to transform your top recurring RCA findings into a safe pattern catalog, forbidden-to-safe mapping table, and a rollout-ready set of static rule candidates for your codebase.