feat(scrape): disk preflight on host runner for cron jobs

run-discord-scrape-host.sh runs verify --disk-only before preflight/scrape
so setup-cron monthly jobs fail fast when archive roots are low on space.
Harden bootstrap smoke to surface failures when dry-run fails.

docs/plans/2026-05-29-027-feat-host-runner-disk-preflight-plan.md

@@ -0,0 +1,26 @@
+---
+title: feat: Disk preflight on host runner (cron path)
+type: feat
+status: complete
+date: 2026-05-29
+origin: /lfg — monthly cron calls run-discord-scrape-host.sh directly, bypassing run-documents-scrape disk check
+---
+
+# feat: Disk preflight on host runner (cron path)
+
+## Summary
+
+`setup-cron.sh` invokes `run-discord-scrape-host.sh scrape`, not `run-documents-scrape.sh`. Run the same `--disk-only` check in the host wrapper before `preflight` and `scrape` so scheduled jobs fail fast when `/home` is full.
+
+## Requirements
+
+| ID | Requirement |
+|----|-------------|
+| R1 | `run-discord-scrape-host.sh` resolves host config from `--config` (maps `/config/...` to repo config) |
+| R2 | Calls `verify-operator-ready.sh --disk-only` when the script exists (skips in minimal fake-repo smokes) |
+| R3 | `DCE_SKIP_DISK_CHECK=1` bypasses check for tests that need it |
+| R4 | `run-all-smokes.sh` still passes with `DCE_MIN_FREE_MB=0` |
+
+## Verification
+
+- `DCE_MIN_FREE_MB=0 ./scripts/run-all-smokes.sh`

docs/recurring-scrape-merge-readiness.md

@@ -57,6 +57,7 @@ df -h ~/Documents /home/brunner56/Downloads/DiscordChatExporter
 
 Override threshold: `DCE_MIN_FREE_MB=2048 ./scripts/verify-operator-ready.sh`  
 Skip check (smokes only): `DCE_MIN_FREE_MB=0`  
+Also enforced by `run-documents-scrape.sh`, `run-discord-scrape-host.sh` (cron), and `run-operator-validation.sh`.
 
 ## CI note (fork PRs)
 

scripts/run-discord-scrape-host.sh

@@ -12,6 +12,7 @@ DOCKER_BIN_OVERRIDDEN=0
 REAUTH_COMMAND=""
 COMPOSE_ENV_FILE=""
 COMPOSE_ENV_TEMP=""
+VERIFY_READY="$REPO_ROOT/scripts/verify-operator-ready.sh"
 
 if [[ -n "${DCE_DOCKER_BIN:-}" ]]; then
   DOCKER_BIN_OVERRIDDEN=1
@@ -259,6 +260,48 @@ resolve_reauth_command() {
   printf '%s\n' "$resolved_path"
 }
 
+resolve_host_config_path() {
+  local -a args=("$@")
+  local i=0 cfg="$REPO_ROOT/config/scrape-targets.json"
+
+  while (( i < ${#args[@]} )); do
+    if [[ "${args[i]}" == "--config" ]]; then
+      cfg="${args[i + 1]:-}"
+      case "$cfg" in
+        /config/*)
+          cfg="$REPO_ROOT/config/scrape-targets.json"
+          ;;
+        ./*)
+          cfg="$REPO_ROOT/${cfg#./}"
+          ;;
+        /*) ;;
+        *)
+          cfg="$REPO_ROOT/$cfg"
+          ;;
+      esac
+      break
+    fi
+    i=$((i + 1))
+  done
+
+  printf '%s\n' "$cfg"
+}
+
+run_disk_preflight_if_enabled() {
+  local -a args=("$@")
+  local cfg
+
+  if [[ "${DCE_SKIP_DISK_CHECK:-0}" == 1 ]]; then
+    return 0
+  fi
+  if [[ ! -x "$VERIFY_READY" ]]; then
+    return 0
+  fi
+
+  cfg=$(resolve_host_config_path "${args[@]}")
+  "$VERIFY_READY" --disk-only --config "$cfg"
+}
+
 is_discord_auth_failure() {
   local output_file=$1
   grep -Eqi \
@@ -381,6 +424,7 @@ main() {
   [[ -f "$COMPOSE_FILE" ]] || die "Missing compose file: $COMPOSE_FILE"
   prepare_compose_env
   REAUTH_COMMAND="${DCE_REAUTH_COMMAND:-}"
+  run_disk_preflight_if_enabled "${passthrough_args[@]}"
 
   case "$subcommand" in
     preflight|scrape)

scripts/tests/bootstrap-recurring-scrape-smoke.sh

@@ -41,9 +41,15 @@ cat >"$TMP_DIR/config.json" <<JSON
 }
 JSON
 
-"$BOOTSTRAP" --dry-run --config "$TMP_DIR/config.json" | grep -q 'Dry run complete' || {
-  printf 'bootstrap --dry-run did not complete\n' >&2
+set +e
+bootstrap_output=$("$BOOTSTRAP" --dry-run --config "$TMP_DIR/config.json" 2>&1)
+bootstrap_status=$?
+set -e
+
+if [[ "$bootstrap_status" -ne 0 ]] || ! grep -q 'Dry run complete' <<<"$bootstrap_output"; then
+  printf 'bootstrap --dry-run failed (status=%s)\n' "$bootstrap_status" >&2
+  printf '%s\n' "$bootstrap_output" >&2
   exit 1
-}
+fi
 
 printf 'bootstrap-recurring-scrape-smoke: ok\n'