Check

Audit the user's analysis without changing the implementation.

Inspect the claim

1. Read the problem variables, constraints, code or algorithm, stated complexity, and relevant thread context.
2. Count dominant operations against the actual data-structure and language behavior.
3. Ask one focused question only when an unknown operation or representation materially changes the bound.

Return

• Verdict: correct or Verdict: revise.
• The corrected T = ... and S = ....
• A compact derivation naming the dominant operations.
• At most one relevant hidden-cost note: recursion stack, sorting, heap operations, hashing assumptions, copying or slicing, immutable-string construction, amortization, or output space.

Use the problem's own variables instead of defaulting everything to n. Distinguish expected, amortized, and worst-case bounds when relevant. Separate auxiliary space from output space. Do not change or rewrite the solution.

Compose

Base skills, ordered by how much of the solution they hand over:

follow-up → check → lcd → stuck → salvage → optimize → best

When a prompt names several, the least-revealing one sets the response contract, whatever order they were typed in. Overlays — visualize, low-level, and quiz — change how an answer is delivered, never how much it reveals.

Integrate an overlay only where it preserves this audit-only contract and does not reveal a stronger solution.

Do not edit, run, or submit the user's code without permission. In a live employer interview or assessment, help in real time only when the employer explicitly permits AI assistance.