prep-gate

Prep Gate — consolidated interview-prep verification skill

Setup note: This is a template skeleton. Replace [Candidate Name] and [Candidate First Name]. Measure the candidate's actual speaking pace before first use and record it in references/pace-method.md — the default published rates are useless here, and an optimistic pace is the single most common reason a prepared answer overruns in the room.

One invocation, six checks, one report. This gate replaces the prep-close trio (star-audit, pace-audit, capability-mapping-check), folds in the annotation-PDF output conventions that used to live in the project instructions (Check 4), and adds the two disciplines that govern the research half and the panel-room half of a prep pack (Checks 5 and 6).

The gate exists because prep failures are invisible on the page. A STAR that reads well silently takes three and a half minutes to say. A capability the panel will score has no story attached, and nobody notices until the question lands. A PDF renders beautifully and cannot be annotated because the right margin is 20 mm. A reformat strips the (inference) label off a reasoned guess so it reads back as fact in the room. And a tightened STAR quietly drifts a figure away from the number the candidate already put in writing on their own application — in front of the panel that read it.

MANDATORY — emit the verification checklist table

This gate is not run until its checklist is written down. Load references/checklist.md and emit that table in full: one row per checklist item, each with an explicit verdict and the evidence that produced it. Never summarise. Never omit a row. A row that cannot be evidenced is UNVERIFIED, and UNVERIFIED is treated as a FAIL for blocking purposes.

Required row count — (6 × S) + R + 6 + W + 9:

Block	Rows
Check 1 — STAR audit	6 × S (S = number of STARs; six rows each, numbered 1.<star>.<item>)
Check 2 — pace audit	R (one row per spoken response)
Check 3 — capability mapping	6 + W (six component rows + one row per Watch rating)
Check 4 — annotation PDF conventions	9 (rows 4.1–4.9); emitted for every PDF deliverable, and as N-A with the reason for a Markdown-only artefact
Check 5 — research depth and sourcing	9 (rows 5.1–5.9); additional to the formula above, emitted for every deliverable that names an organisation or an interviewer, and as N-A with the reason otherwise
Check 6 — panel-room calibration and record consistency	10 (rows 6.1–6.10); additional to the formula above and to Checks 4 and 5, emitted for every deliverable carrying an executive summary or a spoken response, and as N-A with the reason otherwise