Guardrail Designer
Goal
Define the minimum constraints needed to keep a plan inside safe operating limits.
This skill does not stop work by default. It translates known risks into explicit boundaries, tripwires, and do-not-cross conditions.
When To Use
- after failure modes are known
- before risky execution
- when a plan needs safe operating limits rather than broad caution
Scope Boundaries
In scope:
- define red lines and thresholds
- name stop conditions and fallback conditions
- convert vague caution into explicit boundaries
Out of scope by default:
- broad policy writing
- full compliance frameworks
Workflow
- State the action or system.
- Identify the risk areas that need control.
- Define operating limits.
- Define stop triggers.
- Define fallback or escalation response.
Output Contract
protected_actionrisk_areasguardrailsstop_triggersfallback_conditionsrecommended_followupnext_step
Guardrails
- Do not write abstract "be careful" rules.
- Keep guardrails observable and actionable.
- Prefer a small set of hard constraints over a long soft checklist.