Split the 30-Word Concurrency Sentence in the Merge-Bot Caller Stub #2294

Problem

The merge-bot caller stub's concurrency header comment is one 30-word sentence, over the fleet's 25-word sentence limit:

Concurrency keys on the PR number rather than on github.ref, which under pull_request_target is the base branch and would serialize every bot PR against it, so each PR queues independently.

It sits at line 9 of both catalog/snippets/workflows/merge-bot-pull-request.yml and .github/workflows/merge-bot-pull-request.yml. Every adopter copies the stub byte for byte, so a PR-hosted reviewer flags it again in each downstream adoption PR. Copilot raised it on ptr727/homeassistant-purpleair#371. A downstream edit would fork carried text, so the fix belongs here.

prose_lint.py --check sentence-length passes the line, so the gate doesn't catch it either. That may be the same per-line counting gap as #2292, or a different comment-sentence rule.

Suggested fix

Split it at the clause boundary, one sentence per line, in both files:

# Concurrency keys on the PR number rather than on github.ref, which under pull_request_target is the base branch and would serialize every bot PR against it.
# Each PR therefore queues independently.

Then re-cut the snippet's pin per the catalog's release ordering.

🤖 Generated with Claude Code