feat(scrape): wire salvage flags through operator validation and proof

Expose --salvage-only and --salvage-before-scrape on run-operator-validation.sh
and --salvage-before-scrape on run-operator-proof.sh so operators can merge stale
.dce-temp exports before or instead of incremental Discord scrapes.

docs/plans/2026-06-04-055-feat-operator-validation-salvage-plan.md

@@ -0,0 +1,55 @@
+---
+title: "feat: Salvage flags on operator validation and proof"
+type: feat
+status: active
+date: 2026-06-04
+origin: /lfg — plan 054 added documents --salvage-only; operator-validation/proof still lack salvage entry points for yes_general catch-up
+---
+
+# feat: Salvage flags on operator validation and proof
+
+## Summary
+
+Add `--salvage-only` and `--salvage-before-scrape` to `run-operator-validation.sh` and `run-operator-proof.sh`, forwarding to `run-documents-scrape.sh --salvage-only` and the normal scrape path.
+
+## Problem Frame
+
+After stopping a long KotOR export, operators need:
+
+```bash
+./scripts/run-operator-validation.sh --salvage-only --target KotOR_discord_msgs --channel 221726893064454144
+```
+
+Plan 054 implemented salvage on documents-scrape, but validation/proof orchestrators do not expose it.
+
+## Requirements
+
+| ID | Requirement |
+|----|-------------|
+| R1 | `run-operator-validation.sh` accepts `--salvage-only` (salvage + audit, no Discord scrape) |
+| R2 | `run-operator-validation.sh` accepts `--salvage-before-scrape` (salvage then scrape + audit) |
+| R3 | Both flags forward `--target` / `--channel` to documents scrape |
+| R4 | `--salvage-only` and `--salvage-before-scrape` are mutually exclusive |
+| R5 | `run-operator-proof.sh` accepts `--salvage-before-scrape` and runs salvage before scrape+prove |
+| R6 | Smokes cover validation salvage-only; `run-all-smokes.sh` passes |
+
+## Implementation Units
+
+### U1. Operator validation salvage flags
+
+**Files:** `scripts/run-operator-validation.sh`, `scripts/tests/run-operator-validation-smoke.sh`
+
+### U2. Operator proof salvage-before
+
+**Files:** `scripts/run-operator-proof.sh`
+
+### U3. Smoke gate
+
+**Verification:** `DCE_MIN_FREE_MB=0 ./scripts/run-all-smokes.sh`
+
+## Scope Boundaries
+
+### Deferred
+
+- Killing stale PID 3868255 on host
+- Live yes_general catch-up inside LFG

scripts/run-operator-proof.sh

@@ -16,19 +16,21 @@ source "$SCRIPT_DIR/lib/scrape-run-plan.sh"
 TARGET=""
 SYNC_GUI_FLAG=0
 DRY_RUN=0
+SALVAGE_BEFORE=0
 CHANNEL_ARGS=()
 
 usage() {
   cat <<EOF
 Usage:
-  $(basename "$0") [--target NAME] [--channel ID] [--config PATH] [--sync-gui] [--dry-run]
+  $(basename "$0") [--target NAME] [--channel ID] [--config PATH] [--sync-gui] [--dry-run] [--salvage-before-scrape]
 
 End-to-end operator proof:
-  operator-handoff → incremental scrape → prove-incremental-append
+  operator-handoff → [optional salvage] → incremental scrape → prove-incremental-append
 
 When --target is omitted, all enabled targets in the config are processed.
 
   --channel ID  With exactly one --target, limit scrape/prove to channel ID (repeatable)
+  --salvage-before-scrape  Merge stale .dce-temp exports before incremental scrape
 
 Logs append to logs/operator-proof-<timestamp>.log
 EOF
@@ -60,6 +62,10 @@ main() {
         DRY_RUN=1
         shift
         ;;
+      --salvage-before-scrape)
+        SALVAGE_BEFORE=1
+        shift
+        ;;
       --channel)
         [[ $# -ge 2 ]] || die "Missing value for --channel."
         CHANNEL_ARGS+=(--channel "$2")
@@ -120,6 +126,13 @@ main() {
       printf '\n--- Target: %s ---\n' "$name"
       local -a scrape_args=(--config "$CONFIG_PATH" --target "$name")
       scrape_args+=("${CHANNEL_ARGS[@]}")
+      if (( SALVAGE_BEFORE )); then
+        if ! "$DOCUMENTS" "${scrape_args[@]}" --salvage-only; then
+          failed=$((failed + 1))
+          printf 'Operator proof FAILED for %s (salvage-before)\n' "$name" >&2
+          continue
+        fi
+      fi
       if "$DOCUMENTS" "${scrape_args[@]}" && "$PROVE" "${scrape_args[@]}"; then
         succeeded=$((succeeded + 1))
         printf 'Operator proof passed for %s\n' "$name"

scripts/run-operator-validation.sh

@@ -13,6 +13,8 @@ AUDIT_JSON="$REPO_ROOT/scripts/audit-archive-json.sh"
 
 DRY_RUN=0
 SKIP_SCRAPE=0
+SALVAGE_ONLY=0
+SALVAGE_BEFORE=0
 SYNC_GUI_FLAG=0
 PER_TARGET=0
 CONTINUE_ON_ERROR=0
@@ -31,6 +33,8 @@ End-to-end operator validation with timestamped log:
 Options:
   --dry-run       Readiness + archives only (no Discord scrape)
   --skip-scrape   Readiness only (no scrape, no audit loop)
+  --salvage-only  Merge stale .dce-temp exports only, then audit (no Discord scrape)
+  --salvage-before-scrape  Run salvage-only pass before incremental scrape
   --sync-gui      Run sync-token-from-gui.sh --force before checks
   --target NAME   Limit scrape/audit to one configured target
   --channel ID    With exactly one --target, limit scrape to channel ID (repeatable)
@@ -86,19 +90,75 @@ audit_targets() {
   (( failures == 0 ))
 }
 
+run_documents_scrape() {
+  local -a scrape_args=(--config "$CONFIG_PATH")
+  scrape_args+=("${CHANNEL_ARGS[@]}")
+  [[ -n "$TARGET" ]] && scrape_args+=(--target "$TARGET")
+
+  if (( SALVAGE_ONLY )); then
+    run_step "run-documents-scrape (salvage-only)" "$DOCUMENTS_SCRAPE" "${scrape_args[@]}" --salvage-only
+    return $?
+  fi
+
+  if (( DRY_RUN )); then
+    run_step "run-documents-scrape (dry-run)" "$DOCUMENTS_SCRAPE" "${scrape_args[@]}" --dry-run
+    return $?
+  fi
+
+  if (( SALVAGE_BEFORE )); then
+    run_step "run-documents-scrape (salvage-before)" "$DOCUMENTS_SCRAPE" "${scrape_args[@]}" --salvage-only || return $?
+  fi
+
+  run_step "run-documents-scrape" "$DOCUMENTS_SCRAPE" "${scrape_args[@]}"
+}
+
 scrape_per_target() {
   local name failures=0 ok=0
   local -a scrape_args=(--config "$CONFIG_PATH")
   scrape_args+=("${CHANNEL_ARGS[@]}")
   local -a target_names=()
-  if (( DRY_RUN )); then
-    scrape_args+=(--dry-run)
-  fi
   mapfile -t target_names < <(enabled_targets)
   for name in "${target_names[@]}"; do
     [[ -n "$name" ]] || continue
     log_step "Per-target begin: $name"
-    if ! run_step "run-documents-scrape ($name)" "$DOCUMENTS_SCRAPE" "${scrape_args[@]}" --target "$name"; then
+    local -a per_args=("${scrape_args[@]}" --target "$name")
+    if (( SALVAGE_ONLY )); then
+      if ! run_step "run-documents-scrape ($name salvage-only)" "$DOCUMENTS_SCRAPE" "${per_args[@]}" --salvage-only; then
+        log_step "Per-target failed: $name (salvage-only)"
+        failures=$((failures + 1))
+        (( CONTINUE_ON_ERROR )) || return 1
+        continue
+      fi
+      if run_step "audit-archive-json ($name)" "$AUDIT_JSON" --config "$CONFIG_PATH" --target "$name"; then
+        log_step "Per-target done: $name (salvage-only ok)"
+        ok=$((ok + 1))
+      else
+        log_step "Per-target failed: $name (audit)"
+        failures=$((failures + 1))
+        (( CONTINUE_ON_ERROR )) || return 1
+      fi
+      continue
+    fi
+    if (( DRY_RUN )); then
+      if ! run_step "run-documents-scrape ($name)" "$DOCUMENTS_SCRAPE" "${per_args[@]}" --dry-run; then
+        log_step "Per-target failed: $name (dry-run)"
+        failures=$((failures + 1))
+        (( CONTINUE_ON_ERROR )) || return 1
+        continue
+      fi
+      log_step "Per-target done: $name (dry-run)"
+      ok=$((ok + 1))
+      continue
+    fi
+    if (( SALVAGE_BEFORE )); then
+      if ! run_step "run-documents-scrape ($name salvage)" "$DOCUMENTS_SCRAPE" "${per_args[@]}" --salvage-only; then
+        log_step "Per-target failed: $name (salvage-before)"
+        failures=$((failures + 1))
+        (( CONTINUE_ON_ERROR )) || return 1
+        continue
+      fi
+    fi
+    if ! run_step "run-documents-scrape ($name)" "$DOCUMENTS_SCRAPE" "${per_args[@]}"; then
       log_step "Per-target failed: $name (scrape)"
       failures=$((failures + 1))
       if (( CONTINUE_ON_ERROR == 0 )); then
@@ -137,6 +197,14 @@ main() {
         SKIP_SCRAPE=1
         shift
         ;;
+      --salvage-only)
+        SALVAGE_ONLY=1
+        shift
+        ;;
+      --salvage-before-scrape)
+        SALVAGE_BEFORE=1
+        shift
+        ;;
       --sync-gui)
         SYNC_GUI_FLAG=1
         shift
@@ -179,6 +247,13 @@ main() {
     esac
   done
 
+  if (( SALVAGE_ONLY == 1 && SALVAGE_BEFORE == 1 )); then
+    die "--salvage-only and --salvage-before-scrape are mutually exclusive."
+  fi
+  if (( SALVAGE_ONLY == 1 && DRY_RUN == 1 )); then
+    die "--salvage-only cannot be combined with --dry-run."
+  fi
+
   mkdir -p "$LOG_DIR"
   if [[ -z "$LOG_FILE" ]]; then
     LOG_FILE="$LOG_DIR/operator-validation-$(date -u +%Y%m%dT%H%M%SZ).log"
@@ -192,6 +267,8 @@ main() {
     if [[ -n "$TARGET" ]]; then
       log_step "Targets: $TARGET"
       ((${#CHANNEL_ARGS[@]})) && log_step "Scrape channel filter: ${CHANNEL_ARGS[*]}"
+      (( SALVAGE_ONLY )) && log_step "Mode: salvage-only"
+      (( SALVAGE_BEFORE )) && log_step "Mode: salvage-before-scrape"
     else
       log_step "Enabled targets: $(enabled_targets | paste -sd, -)"
     fi
@@ -206,16 +283,13 @@ main() {
     elif (( PER_TARGET )) && [[ -z "$TARGET" ]]; then
       scrape_per_target || failures=$((failures + 1))
     else
-      local -a scrape_args=(--config "$CONFIG_PATH")
+      if run_documents_scrape; then
-      scrape_args+=("${CHANNEL_ARGS[@]}")
-      [[ -n "$TARGET" ]] && scrape_args+=(--target "$TARGET")
-      if (( DRY_RUN )); then
-        scrape_args+=(--dry-run)
-      fi
-      run_step "run-documents-scrape" "$DOCUMENTS_SCRAPE" "${scrape_args[@]}" || failures=$((failures + 1))
         if (( DRY_RUN == 0 && failures == 0 )); then
           audit_targets || failures=$((failures + 1))
         fi
+      else
+        failures=$((failures + 1))
+      fi
     fi
 
     if (( failures > 0 )); then

scripts/tests/run-operator-validation-smoke.sh

@@ -72,4 +72,23 @@ grep -q 'Operator validation finished successfully' "$LOG_DIR/validation.log" ||
   exit 1
 }
 
+SALVAGE_LOG="$LOG_DIR/salvage-validation.log"
+DCE_MIN_FREE_MB=0 DCE_REPO_ROOT="$REPO_ROOT" DCE_CONFIG_FILE="$CONFIG_PATH" DCE_ENV_FILE="$ENV_PATH" \
+  DCE_LOG_DIR="$LOG_DIR" DCE_SKIP_SCRAPE_LOCK=1 \
+  "$RUNNER" --salvage-only --target demo --config "$CONFIG_PATH" --log-file "$SALVAGE_LOG"
+
+grep -q 'Mode: salvage-only' "$SALVAGE_LOG" || {
+  printf 'ERROR: salvage-only mode not logged\n' >&2
+  exit 1
+}
+grep -q 'run-documents-scrape (salvage-only)' "$SALVAGE_LOG" || {
+  printf 'ERROR: salvage-only documents scrape step missing\n' >&2
+  exit 1
+}
+grep -q 'Operator validation finished successfully' "$SALVAGE_LOG" || {
+  printf 'ERROR: salvage-only validation did not succeed\n' >&2
+  cat "$SALVAGE_LOG" >&2
+  exit 1
+}
+
 printf 'run-operator-validation-smoke: ok\n'