The Agent Skills Directory

[SAFE]: The skill is a comprehensive knowledge base focused on engineering safety in transactional and distributed systems. It contains only markdown documentation, reference papers, and structured decision-making workflows.
[SAFE]: No executable code, shell commands, or automated scripts are included in the skill. The instructions guide the AI agent through manual reasoning and verification steps.
[SAFE]: External links and references target official documentation and research from well-known technology organizations, including Microsoft Research, ACM, PostgreSQL, MongoDB, Redis, and Stripe. These are used appropriately to provide authoritative context for system design.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided code and architecture for review. While this constitutes an attack surface for indirect prompt injection, the skill includes explicit 'Stop Conditions' and strict validation requirements that prioritize business invariants and technical correctness, effectively serving as a defensive framework.

transactions-consistency