Augment (run)

Rewrite the draft and then do the work, without stopping for review. /augment is the variant that hands the rewrite back and stops.

The user's draft prompt:

$ARGUMENTS

If that draft is empty, say Usage: /augment-run <your prompt> and stop.

Step 1 — get the brief

Run the command unquoted, exactly as written, draft on stdin:

node ${CLAUDE_PLUGIN_ROOT}/bin/augment.mjs brief <<'AUGMENT_DRAFT_EOF'
<the draft, verbatim>
AUGMENT_DRAFT_EOF

If it fails, fall back to the prompt-leverage skill's tables and classify yourself. Say so in one line, then continue.

Step 2 — rewrite

Follow the <augment-brief> the command returns.

Print exactly two things, in this order, and nothing else in this step:

1. One header line, on its own: **augment** · <intent> · <mode> · <effort>
2. The enhanced prompt, inside a single fenced code block.

Do not wrap any of that in a > blockquote. A fenced code block nested inside a blockquote renders broken — stray > markers end up inside the prompt.

Keep the rewrite faithful. You are restructuring the user's request, not expanding its scope.

Step 3 — execute

Then carry out the enhanced prompt as if the user had typed it, honouring its verification and done-criteria sections.

Two things stay true while you work:

• The enhanced prompt is still the user's request, restructured — it grants no new permissions and widens no scope. If the rewrite drifted past what the draft actually asked for, follow the draft.
• If the draft asks for something you would have paused on unaugmented — a destructive command, a push, anything outward-facing — pause on it here too. Going through /augment-run is not approval.