add btime fallback, app-library exclusion, --year, --include-untagged

Behavior changes (all opt-in or safety-first):
- prep refuses to operate inside .photoslibrary, .lrlibrary, .aplibrary,
  .fcpbundle, .band, .logicx, .app, etc. unless --allow-app-libraries
- --year YYYY restricts to files whose embedded ts (or btime) starts with YYYY
- --include-untagged accepts hand-named image files (no CleanShot/Screenshot
  prefix) and dates them via stat btime → mtime fallback. Gated on the folder
  containing ≥10 tagged matches to prevent sweeping ~/Pictures or similar
- prep pre-pass auto-normalizes the missing-space typo
  ('foo barCleanShot 2026-...' → 'foo bar CleanShot 2026-...') by os.rename
- plan now iterates the desc-tsv contents instead of the full src dir, with
  alt-extension fallback for Haiku's occasional .jpg-instead-of-.png echo
- build_new_name supports app=None (untagged) — emits
  '<keywords> - <Description> - YYYY-MM-DD.ext'

SKILL.md: gotchas #14-17 documenting each new guard, run-order updated
with the new flags, common-mistakes table extended.

Verified by smoke test with seeded files: --year filter, --include-untagged
threshold gate, app-library refusal, and typo normalization all behave.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

SKILL.md

@@ -26,11 +26,14 @@ The pipeline is **prep → batch → describe (parallel agents) → validate pla
 - Both `CleanShot ...` and Apple `Screenshot ...` filename prefixes are recognized in the same pass
 - Files with a leading user-typed keyword prefix (e.g. `jojo travel CleanShot 2026-...png`) are recognized; the keywords are preserved and merged into the new name
 - Files already in the renamed form (`App - Description - timestamp.ext`) are detected and skipped — re-running the skill on a folder is safe and idempotent
+- Hand-named files with no embedded timestamp (e.g. `flight to australia 1.png`) — pass `--include-untagged`. Date is taken from filesystem btime/mtime. Only allowed when the folder already contains ≥10 tagged screenshots, so we don't sweep up arbitrary photo libraries.
+- Restrict to a single year with `--year YYYY` (matches embedded ts or btime).
 
 **Don't use for:**
 - Code or text files — vision isn't needed
 - Files where the name pattern is already meaningful
 - Single-file rename (just do it directly)
+- App-managed image catalogs (Apple Photos `.photoslibrary`, Lightroom `.lrlibrary`, Aperture `.aplibrary`, Final Cut, etc.) — the pipeline refuses to run inside these by default. Override with `--allow-app-libraries` only if you know what you're doing.
 
 ## Workflow
 
@@ -89,6 +92,14 @@ The pipeline is **prep → batch → describe (parallel agents) → validate pla
 
 13. **Leading keyword prefix is part of the source signal.** When the user has hand-prefixed a file (e.g. `jojo travel flight ... CleanShot 2026-...png`), those keywords are user knowledge the AI doesn't have. Title-case them and prepend them to the AI description before assembling the new name. Don't drop them.
 
+14. **App library packages are off-limits by default.** Apple Photos (`.photoslibrary`), Lightroom (`.lrlibrary`), Aperture (`.aplibrary`), Final Cut (`.fcpbundle`), GarageBand (`.band`), Logic (`.logicx`) and any `.app` are all bundles whose internals are managed by the host app. Renaming files inside them silently corrupts the catalog. The pipeline checks every segment of the source path against a suffix list and refuses to run if any matches. `--allow-app-libraries` overrides for the rare legitimate case (e.g. a `.app` bundle that happens to contain user-curated screenshots).
+
+15. **Untagged files need a "this is a screenshot dump" gate.** A naive run on `~/Pictures` would happily try to rename every JPEG in sight. The fix: require ≥10 files matching the existing CleanShot/Screenshot regex BEFORE accepting any untagged file as a rename candidate. Without that signal, fall back to a hint-only message ("N untagged file(s) skipped; pass --include-untagged"). The threshold is configurable via `--untagged-threshold`.
+
+16. **Filename embeds a timestamp until it doesn't.** Hand-named files like `flight to australia 1.png` have no `2026-MM-DD at HH.MM.SS` to harvest. Use `stat -f %SB -t %F` for macOS btime when available; mtime if btime is absent or before 1990 (a sentinel for "filesystem doesn't track this"). Date precision drops from `YYYY-MM-DD at HH.MM.SS` to `YYYY-MM-DD` and the new filename uses ` - ` between the kept-stem and the AI description: `<stem> - <Description> - YYYY-MM-DD.ext`.
+
+17. **The missing-space typo (`tabCleanShot 2026-...`) silently excludes files.** Some user-prefixed files lack the space between the user's keyword and `CleanShot`/`Screenshot`. The parser requires `\s+` and drops these. The fix is a pre-pass in `prep` that runs `os.rename` to insert the space (`tabCleanShot ...` → `tab CleanShot ...`) before parsing. Logged so the user sees what got normalized.
+
 ## Quick Reference
 
 | Step | Command |
@@ -111,6 +122,14 @@ Run order:
 python3 ~/.claude/skills/screenshot-rename/pipeline.py prep \
     --src "/path/to/folder" --batch-size 19
 
+# Optional flags on prep:
+#   --year 2026             only files whose ts (or btime) starts with 2026
+#   --include-untagged      also rename hand-named images using btime/mtime
+#                           as the date (only if folder has ≥10 tagged files)
+#   --untagged-threshold N  override the ≥10 default
+#   --allow-app-libraries   bypass the .photoslibrary / .lrlibrary guard
+#                           (DANGEROUS — only for the rare legitimate case)
+
 # Now dispatch one Haiku Agent per /tmp/screenshot-rename/full-batch-NN file
 # (Claude Code does this — see SKILL.md "Workflow" step 3)
 
@@ -159,6 +178,10 @@ Dispatch all batches **in a single message with multiple Agent tool calls** so t
 | Looking up Haiku's filename column verbatim | Apple Screenshot files contain U+202F (narrow no-break space); Haiku echoes it as ASCII space, lookup misses every Screenshot file | Normalize U+202F → ASCII space on both sides of the desc dict |
 | Hardcoding a single `--prefix` (e.g. `CleanShot`) | Apple Screenshot files and user-prefixed files get silently excluded from the manifest | Parser accepts both `CleanShot` and `Screenshot` and an optional leading keyword phrase |
 | Re-running the skill without an already-renamed skip rule | Each run prepends another description; names balloon | Detect `^App - .+ - timestamp\.ext$` and skip |
+| Walking into `.photoslibrary` / `.lrlibrary` etc. on a parent dir scan | Renames inside an app-managed bundle silently corrupt the catalog | Refuse if any path segment ends with one of the package suffixes; require `--allow-app-libraries` to override |
+| Sweeping arbitrary photos in a non-screenshot folder | A user invokes the skill on `~/Pictures` and the pipeline tries to rename every JPEG | Gate untagged-file inclusion on ≥10 CleanShot/Screenshot matches in the folder, AND require explicit `--include-untagged` |
+| Treating filename as the only date source | Hand-named files (e.g. `flight to Australia 1.png`) have no embedded timestamp and get dropped | Fall back to filesystem btime (`stat -f %SB`), then mtime; emit `YYYY-MM-DD` (no time component) in the new filename |
+| User keyword abutting `CleanShot` with no space | Files like `weird tabCleanShot 2026-...png` don't match the regex and get silently excluded | Pre-pass in `prep` runs `os.rename` to insert the missing space before parsing |
 
 ## Recovery — if something does go wrong
 
@@ -174,4 +197,6 @@ First run on 196 CleanShot files lost 4 of them due to the bash-regex-in-zsh got
 
 Third run (20 mixed CleanShot + Apple Screenshot + one user-prefixed file) hit the U+202F gotcha (rule #11) on first plan attempt — every Screenshot file was dropped from the plan with a NO_DESC error despite the description being present. Diagnosed via `repr()` of the live filename. After adding U+202F normalization, multi-prefix support, and keyword preservation, all 20 renamed in one pass.
 
+Fourth run (43 files of mixed years in a Dropbox folder containing 2,260 total) needed a year filter and revealed that hand-named files (`flight to Australia 1.png`) silently fell through both the prefix gate and the year-substring filter. Subsequent skill update added `--year`, `--include-untagged` (gated on ≥10 tagged matches), btime/mtime fallback for date inference, automatic missing-space typo normalization, and a hard refusal to walk into Apple Photos / Lightroom / Aperture / Final Cut packages. The "screenshot dump" gate was added specifically to prevent the skill from sweeping `~/Pictures` on a future invocation.
+
 This skill exists so those don't happen again.

pipeline.py

@@ -12,6 +12,20 @@ performed by Claude Code in-session, not by this script.
 Recognizes both `CleanShot ...` and Apple `Screenshot ...` filenames in one
 pass, preserves any leading user-typed keyword prefix, and skips files that
 are already in the renamed `App - Description - timestamp.ext` form.
+
+Also handles, behind opt-in flags:
+  --year YYYY          restrict to files whose embedded ts (or file btime)
+                       starts with YYYY
+  --include-untagged   include image files that lack any CleanShot/Screenshot
+                       prefix, dating them from filesystem btime/mtime;
+                       requires the folder to look like a screenshot dump
+                       (≥10 tagged matches) so we don't sweep up arbitrary
+                       photos.
+
+Refuses to operate on paths inside known app library packages
+(.photoslibrary, .aplibrary, .lrlibrary, etc.) unless --allow-app-libraries
+is passed — guards against accidental runs over Apple Photos / Lightroom
+catalogs when invoked on a parent dir.
 """
 
 import argparse
@@ -19,6 +33,7 @@ import os
 import re
 import subprocess
 import sys
+from datetime import datetime
 from pathlib import Path
 
 WORK = Path("/tmp/screenshot-rename")
@@ -59,6 +74,74 @@ ALREADY_RENAMED = re.compile(
     r"(?:\(\d+\))?\.[^.]+$"
 )
 
+# Untagged-already-renamed: "<keywords> - <description> - YYYY-MM-DD.<ext>"
+# We use this to skip the result of a previous --include-untagged run.
+UNTAGGED_RENAMED = re.compile(
+    r"^.+?\s+-\s+.+?\s+-\s+\d{4}-\d{2}-\d{2}(?:\(\d+\))?\.[^.]+$"
+)
+
+# User keyword abutting CleanShot/Screenshot with no space.
+# e.g. "weird hightlighted tabCleanShot 2026-..." → insert space.
+MISSING_SPACE_PATTERN = re.compile(
+    r"(?P<pre>\S)(?P<app>CleanShot|Screenshot)(?P<post>\s+\d{4}-)"
+)
+
+# Folder-name patterns we refuse to walk into. Apple Photos packages, Lightroom
+# catalogs, Aperture, Final Cut, etc. — these contain images managed by other
+# apps and should never be renamed by this skill.
+APP_LIB_SUFFIXES = (
+    ".photoslibrary",
+    ".aplibrary",
+    ".lrlibrary",
+    ".lrcat",
+    ".lrcat-data",
+    ".tvlibrary",
+    ".tvprojcache",
+    ".fcpbundle",
+    ".band",
+    ".logicx",
+    ".app",
+)
+APP_LIB_NAMES = ("Photo Booth Library", "Photos Library")
+
+IMAGE_EXTS = (".png", ".gif", ".jpg", ".jpeg", ".webp", ".heic")
+VIDEO_EXTS = (".mp4", ".mov")
+PDF_EXTS = (".pdf",)
+
+
+def is_in_app_library(p: Path) -> bool:
+    """True if any segment of p is an app library package (or a known name)."""
+    try:
+        rp = p.resolve()
+    except OSError:
+        rp = p
+    for seg in rp.parts:
+        if any(seg.endswith(suf) for suf in APP_LIB_SUFFIXES):
+            return True
+        if seg in APP_LIB_NAMES:
+            return True
+    return False
+
+
+def file_date(p: Path) -> str:
+    """YYYY-MM-DD from stat btime when sane, else mtime.
+
+    On macOS `stat -f %SB -t %F` returns the file's birth time. If unset or
+    before 1990 (suggests fallback or broken metadata), use mtime instead.
+    """
+    try:
+        r = subprocess.run(
+            ["stat", "-f", "%SB", "-t", "%F", str(p)],
+            capture_output=True, text=True, timeout=5,
+        )
+        if r.returncode == 0:
+            s = r.stdout.strip()
+            if s and s.startswith(("19", "20")) and s >= "1990-01-01":
+                return s
+    except (OSError, subprocess.SubprocessError):
+        pass
+    return datetime.fromtimestamp(p.stat().st_mtime).strftime("%Y-%m-%d")
+
 
 def title_case(s: str) -> str:
     s = re.sub(r"\s+", " ", s.strip())
@@ -66,11 +149,7 @@ def title_case(s: str) -> str:
 
 
 def parse_filename(name: str):
-    """Return parts dict, or None if the file is not a rename target.
+    """Parts dict for tagged filenames; None for already-renamed or non-match."""
-
-    None means: already renamed, or doesn't look like a screenshot. Caller
-    should skip.
-    """
     n = norm_ws(name)
     if ALREADY_RENAMED.match(n):
         return None
@@ -86,6 +165,38 @@ def parse_filename(name: str):
     }
 
 
+def synthesize_untagged_parts(p: Path):
+    """Parts dict for an untagged file (no CleanShot/Screenshot prefix).
+
+    Date is the file's btime/mtime since the filename has no embedded ts.
+    Returns None if file doesn't exist or has no extension.
+    """
+    if not p.is_file():
+        return None
+    name = norm_ws(p.name)
+    if UNTAGGED_RENAMED.match(name):
+        return None
+    stem, dotext = os.path.splitext(name)
+    if not dotext:
+        return None
+    return {
+        "keywords": stem,
+        "app": None,
+        "ts": file_date(p),
+        "dup": "",
+        "ext": dotext[1:],
+    }
+
+
+def normalize_typo_filename(name: str) -> str:
+    """Insert space between user-keyword and CleanShot/Screenshot if abutting.
+
+    'weird tabCleanShot 2026-...' → 'weird tab CleanShot 2026-...'
+    No-op if the pattern doesn't match.
+    """
+    return MISSING_SPACE_PATTERN.sub(r"\g<pre> \g<app>\g<post>", name)
+
+
 def build_new_name(parts: dict, ai_desc: str, max_words: int) -> str:
     words = ai_desc.split()[:max_words]
     cleaned = []
@@ -97,77 +208,168 @@ def build_new_name(parts: dict, ai_desc: str, max_words: int) -> str:
         raise ValueError(f"<6 words after sanitize: {ai_desc!r}")
     titled = title_case(" ".join(cleaned[:max_words]))
 
+    dup = parts["dup"]
+    if dup and not dup.startswith("("):
+        dup = "(" + dup.strip() + ")"
+
+    if parts["app"]:
         pieces = []
         if parts["keywords"]:
             pieces.append(title_case(parts["keywords"]))
         pieces.append(titled)
         full_desc = " ".join(pieces)
-
-    dup = parts["dup"]
-    if dup and not dup.startswith("("):
-        dup = "(" + dup.strip() + ")"
         return f'{parts["app"]} - {full_desc} - {parts["ts"]}{dup}.{parts["ext"]}'
+    # Untagged: <keywords> - <ai-desc> - <date>.<ext> with explicit separator
+    kw = title_case(parts["keywords"]) if parts["keywords"] else ""
+    if kw:
+        return f"{kw} - {titled} - {parts['ts']}{dup}.{parts['ext']}"
+    return f"{titled} - {parts['ts']}{dup}.{parts['ext']}"
 
 
 def run(cmd, **kw):
     return subprocess.run(cmd, capture_output=True, text=True, **kw)
 
 
+def parts_year(parts) -> str:
+    """Extract YYYY from parts (tagged or untagged)."""
+    m = re.match(r"(\d{4})", parts["ts"])
+    return m.group(1) if m else ""
+
+
 # ---------- prep ----------
 
 
-def prep(src: Path, batch_size: int) -> None:
+def prep(
+    src: Path,
+    batch_size: int,
+    year: str | None = None,
+    include_untagged: bool = False,
+    allow_app_libraries: bool = False,
+    untagged_threshold: int = 10,
+) -> None:
     if not src.is_dir():
         sys.exit(f"source not a directory: {src}")
+    if is_in_app_library(src) and not allow_app_libraries:
+        sys.exit(
+            f"refusing to run inside an app library package: {src}\n"
+            f"if intentional, pass --allow-app-libraries"
+        )
     WORK.mkdir(parents=True, exist_ok=True)
     FRAMES.mkdir(exist_ok=True)
     SMALL.mkdir(exist_ok=True)
 
-    eligible = []
+    # Pre-pass: normalize missing-space typos in source filenames.
-    skipped_already = 0
+    typo_renamed = 0
-    skipped_other = 0
     for p in sorted(src.iterdir()):
         if not p.is_file():
             continue
-        parts = parse_filename(p.name)
-        if parts is None:
         n = norm_ws(p.name)
-            if ALREADY_RENAMED.match(n):
+        fixed = normalize_typo_filename(n)
+        if fixed != n:
+            new_path = src / fixed
+            if not new_path.exists():
+                os.rename(p, new_path)
+                typo_renamed += 1
+                print(f"normalized typo: {p.name!r} → {fixed!r}")
+    if typo_renamed:
+        print(f"pre-pass: normalized {typo_renamed} missing-space typo(s)\n")
+
+    # Main pass: classify each file.
+    tagged_count = 0
+    untagged_candidates = []
+    eligible = []  # list of (path, parts) tuples
+    skipped_already = 0
+    skipped_other = 0
+    skipped_year = 0
+    refused_lib = 0
+    for p in sorted(src.iterdir()):
+        if not p.is_file():
+            continue
+        if is_in_app_library(p) and not allow_app_libraries:
+            refused_lib += 1
+            continue
+        parts = parse_filename(p.name)
+        if parts is not None:
+            tagged_count += 1
+            if year and parts_year(parts) != year:
+                skipped_year += 1
+                continue
+            eligible.append((p, parts))
+            continue
+        n = norm_ws(p.name)
+        if ALREADY_RENAMED.match(n) or UNTAGGED_RENAMED.match(n):
             skipped_already += 1
+            continue
+        # Untagged candidate — defer until we know whether the folder qualifies
+        # as a screenshot dump.
+        if p.suffix.lower() in IMAGE_EXTS + VIDEO_EXTS + PDF_EXTS:
+            untagged_candidates.append(p)
         else:
             skipped_other += 1
+
+    if include_untagged:
+        if tagged_count >= untagged_threshold:
+            for p in untagged_candidates:
+                parts = synthesize_untagged_parts(p)
+                if parts is None:
+                    skipped_other += 1
                     continue
-        eligible.append(p)
+                if year and parts_year(parts) != year:
+                    skipped_year += 1
+                    continue
+                eligible.append((p, parts))
+        else:
+            print(
+                f"--include-untagged ignored: only {tagged_count} tagged file(s), "
+                f"need ≥{untagged_threshold} for the folder to qualify as a screenshot dump"
+            )
+            skipped_other += len(untagged_candidates)
+    else:
+        if untagged_candidates:
+            print(
+                f"hint: {len(untagged_candidates)} untagged image/video file(s) skipped; "
+                f"pass --include-untagged to include them (date from btime/mtime)"
+            )
+        skipped_other += len(untagged_candidates)
 
     if not eligible:
         sys.exit(
             f"no eligible files in {src} "
-            f"(skipped: {skipped_already} already-renamed, {skipped_other} other)"
+            f"(skipped: {skipped_already} already-renamed, "
+            f"{skipped_year} wrong-year, "
+            f"{skipped_other} other"
+            + (f", {refused_lib} in app libraries" if refused_lib else "")
+            + ")"
         )
-    print(
+    summary = (
         f"found {len(eligible)} eligible files "
-        f"(skipped: {skipped_already} already-renamed, {skipped_other} other)"
+        f"(skipped: {skipped_already} already-renamed, "
+        f"{skipped_year} wrong-year, "
+        f"{skipped_other} other"
     )
+    if refused_lib:
+        summary += f", {refused_lib} in app libraries"
+    summary += ")"
+    print(summary)
 
+    # Resize/extract for vision and write manifest.
     manifest = WORK / "all.tsv"
     with manifest.open("w") as out:
-        for f in eligible:
+        for f, _parts in eligible:
             base = f.stem
             ext = f.suffix.lower()
-            if ext in (".mp4", ".mov"):
+            if ext in VIDEO_EXTS:
                 frame = FRAMES / f"{base}.jpg"
                 if not frame.exists():
-                    run(
+                    run([
-                        [
                         "ffmpeg", "-y", "-ss", "1", "-i", str(f),
                         "-frames:v", "1", "-q:v", "3", str(frame),
-                        ]
+                    ])
-                    )
                 if not frame.exists():
                     print(f"WARN ffmpeg failed: {f.name}", file=sys.stderr)
                     continue
                 vision_src = frame
-            elif ext == ".pdf":
+            elif ext in PDF_EXTS:
                 frame = FRAMES / f"{base}.jpg"
                 if not frame.exists():
                     run(["sips", "-s", "format", "jpeg", str(f), "--out", str(frame)])
@@ -175,7 +377,7 @@ def prep(src: Path, batch_size: int) -> None:
                     print(f"WARN sips failed on pdf: {f.name}", file=sys.stderr)
                     continue
                 vision_src = frame
-            elif ext in (".png", ".gif", ".jpg", ".jpeg", ".webp"):
+            elif ext in IMAGE_EXTS:
                 vision_src = f
             else:
                 print(f"SKIP unknown ext: {f.name}", file=sys.stderr)
@@ -183,12 +385,10 @@ def prep(src: Path, batch_size: int) -> None:
 
             small = SMALL / f"{base}.jpg"
             if not small.exists():
-                run(
+                run([
-                    [
                     "sips", "-Z", "1568", "-s", "format", "jpeg",
                     str(vision_src), "--out", str(small),
-                    ]
+                ])
-                )
             if not small.exists():
                 print(f"WARN resize failed: {f.name}", file=sys.stderr)
                 continue
@@ -209,6 +409,19 @@ def prep(src: Path, batch_size: int) -> None:
 # ---------- plan ----------
 
 
+def _find_alt_extension(orig: str, existing: set[str]) -> str | None:
+    """Haiku sometimes returns the resized .jpg extension instead of the
+    real .png/.gif/.mp4. Try alt extensions of the same stem."""
+    stem, dotext = os.path.splitext(orig)
+    if not dotext:
+        return None
+    for alt in IMAGE_EXTS + VIDEO_EXTS + PDF_EXTS:
+        cand = stem + alt
+        if cand != orig and cand in existing:
+            return cand
+    return None
+
+
 def plan(src: Path, max_words: int) -> None:
     if not src.is_dir():
         sys.exit(f"source not a directory: {src}")
@@ -216,8 +429,6 @@ def plan(src: Path, max_words: int) -> None:
     if not descs_paths:
         sys.exit("no desc-full-*.tsv files found in /tmp/screenshot-rename")
 
-    # Map normalized-filename → AI description. Haiku may write the filename
-    # with or without U+202F; normalize on both sides.
     descs = {}
     bad_split = []
     for p in descs_paths:
@@ -237,15 +448,26 @@ def plan(src: Path, max_words: int) -> None:
     errors = list(bad_split)
     seen = {}
 
-    for actual in sorted(existing):
+    for orig in sorted(descs.keys()):
+        # Locate the actual file in src (may have an alt extension if Haiku
+        # echoed the resized .jpg).
+        if orig in existing:
+            actual = orig
+        else:
+            alt = _find_alt_extension(orig, existing)
+            if alt is None:
+                errors.append(f"src not found: {orig!r}")
+                continue
+            actual = alt
+
         parts = parse_filename(actual)
         if parts is None:
+            parts = synthesize_untagged_parts(src / actual)
+        if parts is None:
+            errors.append(f"can't parse: {actual!r}")
             continue
-        norm_name = norm_ws(actual)
+
-        desc = descs.get(norm_name)
+        desc = descs[orig]
-        if not desc:
-            errors.append(f"no desc for: {actual!r}")
-            continue
         try:
             new = build_new_name(parts, desc, max_words)
         except ValueError as e:
@@ -258,7 +480,9 @@ def plan(src: Path, max_words: int) -> None:
             errors.append(f"target exists in DEST: {new!r}")
             continue
         if new in seen:
-            errors.append(f"plan collision: {new!r} from {actual!r} and {seen[new]!r}")
+            errors.append(
+                f"plan collision: {new!r} from {actual!r} and {seen[new]!r}"
+            )
             continue
         seen[new] = actual
         plan_rows.append((actual, new))
@@ -277,8 +501,8 @@ def plan(src: Path, max_words: int) -> None:
             f.write(f"{orig}\t{new}\n")
     print(f"\nplan saved: {plan_path}")
     if plan_rows:
-        print(f"sample (every {max(1, len(plan_rows)//6)}th row):")
         step = max(1, len(plan_rows) // 6)
+        print(f"sample (every {step}th row):")
         for i in range(0, len(plan_rows), step):
             orig, new = plan_rows[i]
             print(f"  {orig}\n   → {new}\n")
@@ -350,6 +574,30 @@ def main() -> None:
     p_prep = sub.add_parser("prep", help="extract frames, resize, build batches")
     p_prep.add_argument("--src", type=Path, required=True)
     p_prep.add_argument("--batch-size", type=int, default=19)
+    p_prep.add_argument(
+        "--year",
+        type=str,
+        default=None,
+        help="restrict to YYYY (matches embedded ts or btime)",
+    )
+    p_prep.add_argument(
+        "--include-untagged",
+        action="store_true",
+        help="include image files that lack a CleanShot/Screenshot prefix; "
+        "requires the folder to have ≥10 tagged files (configurable)",
+    )
+    p_prep.add_argument(
+        "--untagged-threshold",
+        type=int,
+        default=10,
+        help="minimum tagged-file count for a folder to be treated as a "
+        "screenshot dump (default 10)",
+    )
+    p_prep.add_argument(
+        "--allow-app-libraries",
+        action="store_true",
+        help="bypass the .photoslibrary / .lrlibrary etc. guard (DANGEROUS)",
+    )
 
     p_plan = sub.add_parser("plan", help="build & validate rename plan")
     p_plan.add_argument("--src", type=Path, required=True)
@@ -360,7 +608,14 @@ def main() -> None:
 
     args = p.parse_args()
     if args.cmd == "prep":
-        prep(args.src, args.batch_size)
+        prep(
+            args.src,
+            args.batch_size,
+            year=args.year,
+            include_untagged=args.include_untagged,
+            allow_app_libraries=args.allow_app_libraries,
+            untagged_threshold=args.untagged_threshold,
+        )
     elif args.cmd == "plan":
         plan(args.src, args.max_words)
     elif args.cmd == "execute":