This skill loads a compact rule set inspired by Michael T. Nygard’s Release It!. It tells a coding agent that a passing happy path is not production readiness, and to design failure semantics, demand limits, isolation and diagnosis up front.
It applies to services, APIs, jobs, queues, deployment paths and control tooling that must survive latency, overload, bad data and operational mistakes.
What You Get
- A mini rule set with decision rules, trigger rules and a closing checklist for production reliability
- A nano version for tight context budgets and a full version for reference
- A skill description that makes the agent load the rules for failure handling, capacity and deployment tasks
Common Use Cases
- Adding an outbound call with a timeout, retry limits and a degraded mode
- Bounding a queue, buffer, cache or result set and defining behavior when it is full
- Making migrations and one-time jobs restartable, auditable and reversible
- Reviewing an incident for the failure chain and the missing defenses
What Makes It Unique
It is the production-failure rule set in the repository. Each trigger, such as a new queue, cache, health check or API contract, lists what must be defined before the change goes ahead, including timeout, retry eligibility, capacity, full-queue behavior and rollback. It includes hostile traffic, control tooling and chaos testing with stop conditions, which the design-focused rule sets do not.
Under The Hood
The skill is a SKILL.md with a name and description, plus Markdown files holding the mini, nano and full versions of the rules. The SKILL.md tells the agent to read the mini file before design or code decisions and the full file only for harder tradeoffs. It contains no scripts. It does not fetch web pages, call APIs, run commands or write files; the agent only reads the rule text.