feat(scrape): warn on bot-token preflight and document cron install

Bootstrap surfaces forbidden-channel guidance; operator checklist
documents the installed monthly cron job.

docs/plans/2026-05-29-016-feat-operator-activation-plan.md

@@ -0,0 +1,41 @@
+---
+title: feat: Operator activation — cron install and bot-token warning
+type: feat
+status: completed
+date: 2026-05-29
+origin: LFG — original request included monthly cron; cron not yet installed; bot token blocks live fetch
+---
+
+# feat: Operator activation — cron install and bot-token warning
+
+## Summary
+
+Complete the operator slice: warn when preflight only succeeds via seeded archives (typical bot token), install the monthly cron job, and document activation in the checklist.
+
+## Requirements
+
+| ID | Requirement |
+|----|-------------|
+| R1 | `bootstrap-recurring-scrape.sh` prints clear bot-token / user-token guidance after seeded-only preflight |
+| R2 | Monthly cron block installed via `setup-cron.sh` |
+| R3 | Operator checklist notes cron installed and token type |
+| R4 | Bootstrap smoke still passes |
+
+## Implementation Units
+
+### U1. Bot-token advisory
+
+**Files:** `scripts/bootstrap-recurring-scrape.sh`
+
+### U2. Install cron (runtime)
+
+**Command:** `./scripts/setup-cron.sh --skip-preflight`
+
+### U3. Docs
+
+**Files:** `docs/recurring-scrape-operator-checklist.md`
+
+## Verification
+
+- `crontab -l` contains `BEGIN discord-scrape`
+- `bootstrap-recurring-scrape-smoke.sh`

docs/recurring-scrape-operator-checklist.md

@@ -14,11 +14,13 @@ Use this after cloning or opening the **source** repo (`DiscordChatExporter`, no
 
 ```bash
 ./scripts/setup-cron.sh --dry-run
-./scripts/setup-cron.sh
+./scripts/setup-cron.sh --skip-preflight   # after bootstrap preflight already succeeded
 ```
 
 Defaults: first day of month at 04:00. Override with `--interval weekly`, `--at HH:MM`, or `--cron '0 4 1 * *'`.
 
+Installed jobs are marked `# BEGIN discord-scrape` in `crontab -l`. Logs append to `logs/discord-scrape.log`.
+
 ## Narrow a run
 
 ```bash

scripts/bootstrap-recurring-scrape.sh

@@ -134,9 +134,26 @@ main() {
     preflight_args+=(--target "$target")
   done
 
-  "${preflight_args[@]}"
+  local preflight_log preflight_status
+  preflight_log=$(mktemp "${TMPDIR:-/tmp}/dce-bootstrap-preflight.XXXXXX")
+  "${preflight_args[@]}" 2>&1 | tee "$preflight_log"
+  preflight_status=${PIPESTATUS[0]}
+
+  if (( preflight_status != 0 )); then
+    cat "$preflight_log" >&2
+    rm -f "$preflight_log"
+    exit "$preflight_status"
+  fi
 
   printf '\nBootstrap complete.\n'
+  if grep -q 'inaccessible, but .* seeded archive' "$preflight_log" \
+    || grep -qiE 'failed: forbidden|Missing Access' "$preflight_log"; then
+    printf '\nToken note: many channels returned forbidden. That usually means a bot token without message-history access.\n'
+    printf '  For live incremental downloads, put a user token in %s (see .docs/Token-and-IDs.md).\n' "$ENV_FILE"
+    printf '  Append-only archives are still safe: existing JSON is updated in place and never fully re-downloaded.\n'
+  fi
+  rm -f "$preflight_log"
+
   printf '  Scrape now:  %s\n' "$REPO_ROOT/scripts/run-documents-scrape.sh"
   printf '  Install cron: %s --dry-run\n' "$REPO_ROOT/scripts/setup-cron.sh"
 }