fix(review-runs): page the run census, so a day's window isn't the last hour

plugins/tend-ci-runner/skills/review-runs/SKILL.md

@@ -134,16 +134,24 @@ SINCE=$(date -u -d '24 hours ago' +%Y-%m-%dT%H:%M:%SZ)
 # Add the repo's extra prefixes from its `running-tend` skill: any workflow
 # running the tend action is in scope, not just the generated `tend-*` ones.
 # Step 2 prices the same list.
+#
+# `--paginate` on both calls. Both endpoints page at 30 by default and return
+# runs newest-first, so without it the census silently covers only the most
+# recent 30 runs per workflow — on a busy repo that is the last hour, not the
+# last 24. Each `--jq` here is a per-element projection, so `--paginate`
+# applying it per page is harmless.
 PREFIXES=("tend-")
 PREFIX_RE="^($(IFS='|'; echo "${PREFIXES[*]}"))"
-for workflow in $(gh api repos/$REPO/actions/workflows --jq ".workflows[] | select(.name | test(\"$PREFIX_RE\")) | .id"); do
-  gh api "repos/$REPO/actions/workflows/$workflow/runs?created=>=$SINCE&status=completed" \
+for workflow in $(gh api --paginate repos/$REPO/actions/workflows --jq ".workflows[] | select(.name | test(\"$PREFIX_RE\")) | .id"); do
+  gh api --paginate "repos/$REPO/actions/workflows/$workflow/runs?created=>=$SINCE&status=completed&per_page=100" \
     --jq '.workflow_runs[] | {databaseId: .id, conclusion, createdAt: .created_at, name: .name}'
 done
 ```
 
 If no runs found, report "no runs to review" and exit.
 
+Report the run census as the count this returns. Cross-check any workflow whose count lands on a round page boundary (30, 100) against `.total_count` before trusting it — a count that equals the page size is the signature of a page that was never followed.
+
 Then, for each run ID from above, pull its jobs and classify them:
 
 - **Long-running** (>30 min): Tend runs typically finish in single-digit minutes. Anything over 30 is worth a look — download session logs in Step 3 and diagnose where the time went (long background waits, push-wait-fix cycles, a stuck tool call).