# Gap plan: {profile.candidate.name} for All employers

<!-- Placeholders:
     {requirement.<field>} = one item of `job.ai_requirements` (label, kind, importance, min_years, level, evidence); repeat the line per item
     {match.<field>} = this job's object in fit.json (tier, required_met, gaps, judgements, ...)
     {profile.<path>} = candidate-profile.json at that dotted path; `[]` = each item of a list
     Anything in <angle brackets> is yours to write, from those sources only. -->

<!-- From fit.json across the shortlist: unmet required items first, then preferred. Count how many jobs
     each gap blocks; the biggest unlock goes first. Never propose claiming what is not yet true. -->

| Gap (label) | Kind | Blocks (required) | Improves (preferred) | Posting's words | Plan | By |
|---|---|---|---|---|---|---|
| {requirement.label} | {requirement.kind} | <N jobs where it is the only unmet required item> | <M jobs> | "{requirement.evidence}" | <course, certificate, project, or evidence you already have> | <date> |

## One-cert-away

<certification> would move <N> jobs here from reach to good or strong: {match.title}, ...

## Claims to evidence-link first

Self-reported claims that decide a tier: {profile.claims[].label}. Add a document reference to `evidence` and
set `verification` to `evidence_linked`.