Ingrid - Internal Auditor AI Skill

Runs internal audit as a year-round risk discipline, not a scramble before the external auditors show up.

Ingrid is a Senior Internal Auditor persona that builds a risk-based audit universe and annual plan, tests control design versus operating effectiveness separately, chooses a defensible sampling methodology (statistical or judgmental) with a stated population and confidence logic, documents process walkthroughs against how the process actually runs today, writes findings around root cause with severity and remediation owners, tracks remediation to retest rather than to a promise, and coordinates internal audit coverage with external auditors so the two never duplicate each other's testing.

Who it's for

Internal audit managers, controllers, and finance leaders who need an ongoing, risk-based internal audit function rather than a once-a-year compliance exercise, and who want sample sizes, findings, and remediation tracking that would hold up to an audit committee's questions.

Key capabilities

• Builds a risk-scored audit universe and annual plan with hour allocation by quarter
• Tests control design effectiveness and operating effectiveness as two separate steps, never one standing in for the other
• Selects statistical or judgmental sampling methodology with a stated population, confidence level, and resulting sample size
• Writes findings with condition, criteria, quantified effect, root cause, and a named remediation owner and date
• Tracks remediation to retest evidence and coordinates coverage with external auditors to avoid duplicated testing

How to use it

Paste Ingrid's SKILL.md into your Claude Project Instructions (or any AI system prompt), then describe your audit plan, control test, sampling, walkthrough, or findings problem. Works with Claude, ChatGPT, and any AI chat. Under 2 minutes to install.