Release 0.29.0

pre-commit: 1) maintain.sh - exit 1 on uv-secure, lightning PYSEC-2026-3624 unchanged from 0.28.1, no fixed release exists, vulnerable API unreachable (no load_from_checkpoint in project or trustmark); ruff, pyright src/, and 1386 tests passed separately; 2) /simplify - version bump only; 3) docs sync - no version refs outside pyproject, __init__.py, uv.lock; 4) CLAUDE.md - no change

README.md

@@ -202,8 +202,9 @@ remove-ai-watermarks invisible image.png -o clean.png --force
 ```
 
 Typography-heavy images can opt into the experimental verified-text post-pass.
-It requires manually reviewed strings and line boxes; it never trusts OCR as ground
-truth or runs automatically:
+It accepts manually reviewed strings and line boxes or an operator-verified
+geometry-only manifest; it never treats raw OCR output as ground truth or runs
+automatically:
 
 ```bash
 uv tool install --force "remove-ai-watermarks[text-restoration]"

docs/cli.md

@@ -157,6 +157,11 @@ remove-ai-watermarks metadata image.png --remove -o clean.png
 When `-o` is omitted, removal overwrites the source. Standard metadata is kept
 unless you pass `--remove-all`.
 
+A quiet `--check` or a successful `--remove` is not a clean verdict. The command
+only inspects and strips embedded AI metadata; a pixel watermark such as SynthID
+has no local decoder once that metadata proxy is gone. `identify` reports the
+same limit.
+
 The command also supports the audio and video containers listed in
 [supported signals](supported-signals.md). ffmpeg must be available for the
 non-ISOBMFF audio and video path.
@@ -225,6 +230,10 @@ remove-ai-watermarks video metadata input.mp4 --check
 remove-ai-watermarks video metadata input.mp4 --remove -o clean.mp4
 ```
 
+As with the generic `metadata` command, a quiet check or a successful strip is
+not a clean verdict: video SynthID is not decoded locally after the metadata
+proxy is gone.
+
 Supported containers are MP4, MOV, M4V, WebM, MKV, AVI, and FLV. The operation
 delegates to the same verified metadata scanner and stripper as the generic
 `metadata` command, so detection and removal stay in parity. Video and audio
@@ -408,7 +417,7 @@ the source with the Qwen VAE, blends 15% of that reconstruction into the normal
 only the reconstructed glyph cores through source-derived silhouettes. It does not
 run OCR or choose which strings are correct.
 
-Install the combined extra and run only with a manually reviewed manifest:
+Install the combined extra and run only with an operator-verified manifest:
 
 ```bash
 uv tool install --force "remove-ai-watermarks[text-restoration]"
@@ -416,11 +425,10 @@ remove-ai-watermarks invisible image.png -o clean.png \
   --pipeline qwen-zimage --text-manifest verified-lines.json --force
 ```
 
-``verified: true`` may also be set by an automated operator (a service) that
-attests machine-verified geometry: stability-gated detector boxes inside sane
-caps. The restoration pipeline consumes box/script geometry only - the ``text``
-field is advisory metadata and never reaches the pixels - so what verification
-must guarantee is the geometry, and a machine gate can.
+``verified: true`` may also be set by an automated operator that attests
+machine-verified geometry: stability-gated detector boxes inside sane caps. Such
+operators should use the geometry-only schema 2, which carries no transcription
+or script metadata.
 
 Since 0.27.1 the global 15% Qwen-VAE fidelity-anchor blend is off by default: it
 was measured to return detector-visible OpenAI SynthID on poster-scale manifests
@@ -428,16 +436,18 @@ was measured to return detector-visible OpenAI SynthID on poster-scale manifests
 0.27.0 research behavior; text-box fidelity lost by the default is well under one
 MAE point on the measured fixtures.
 
-The manifest is a JSON object with `schema_version: 1`, `verified: true`, decoded
-RGB dimensions, `source_pixel_sha256`, and a non-empty `lines` array. Each line has
-an integer `[x1, y1, x2, y2]` box, exact `text`, a non-empty `script`, and an optional
-angle from -30 to 30 degrees. Lines must be in top-to-bottom, left-to-right order.
+The manifest is a JSON object with `verified: true`, decoded RGB dimensions,
+`source_pixel_sha256`, and a non-empty `lines` array. Schema 1 is retained for
+manually reviewed annotations: each line has an integer `[x1, y1, x2, y2]` box,
+exact `text`, a non-empty `script`, and an optional angle from -30 to 30 degrees.
+Schema 2 is geometry-only: each line has the box and optional angle, with no
+required `text` or `script`. Lines must be in top-to-bottom, left-to-right order.
 The hash binds the annotations to decoded RGB geometry and pixels, so metadata-only
 container changes remain valid while a resized or edited source fails closed. The
 experimental helper
 `remove_ai_watermarks._internal.text_restoration.source_pixel_sha256` computes it.
 
-This mode is supported only by `qwen-zimage` at native untiled geometry with
+This mode is supported only by `qwen-zimage` at native geometry with
 `humanize=0`, `unsharp=0`, and adaptive polish disabled. `all` also accepts the flag,
 but its manifest must match the pixels entering the invisible stage; if visible-mark
 removal changes those pixels, the hash check rejects the run. One oracle verdict does

docs/module-internals.md

@@ -42,6 +42,10 @@ Important contracts:
 - `invisible` writes no output when no supported local signal is found, unless
   `--force` is supplied.
 - The two no-signal conditions currently share exit code `2`.
+- A quiet `metadata --check` and a successful `metadata --remove` (same for
+  `video metadata`) end by repeating the `identify` limit: the pixel channel is
+  untouched and a watermark such as SynthID has no local decoder once its
+  metadata proxy is gone, so neither outcome is a clean verdict.
 - Hard processing and write failures exit with code `1`.
 - `all` can still write the completed visible and metadata stages when the
   diffusion dependencies are unavailable, but exits with code `1` so the
@@ -960,9 +964,10 @@ requires the same provider-oracle and identity evaluation as a model change.
 
 [`_internal/text_restoration.py`](../src/remove_ai_watermarks/_internal/text_restoration.py)
 implements the opt-in `vae-glyphs` stage. A versioned manifest carries manually
-reviewed strings and source-space line boxes, plus a SHA-256 over decoded RGB width,
-height, and pixels. Validation happens before model loading. The product never treats
-OCR confidence as verification.
+reviewed strings and source-space line boxes in schema 1, or verified source-space
+geometry alone in schema 2, plus a SHA-256 over decoded RGB width, height, and pixels.
+Validation happens before model loading. The library never treats OCR confidence as
+verification, and geometry-only operators do not need to invent text or script fields.
 
 When enabled, `QwenZImagePipeline` reconstructs the source once through its already
 loaded Qwen VAE, runs the ordinary global and face stages, blends 15% of the VAE
@@ -971,7 +976,12 @@ The compositor derives binary source and candidate silhouettes, groups nearby li
 uses LaMa for the initial and residual-glyph erase passes, paints fresh silhouette
 edges, then copies the Qwen-VAE core with a 0.5-pixel feather. The evaluation script
 imports these same mask and compositing helpers so the two implementations cannot
-silently drift.
+silently drift. Silhouette crops start 12% of line height beyond each horizontal
+side, then expand each side independently while a foreground component anchored
+inside the detector box still reaches that boundary, up to one line height. They
+extend 8% above and 25% below. The anchored-component gate covers clipped leading
+flourishes, trailing punctuation, icons, and descenders without walking into
+disconnected decoration or background texture.
 
 The stage is deliberately narrower than the engine: it rejects `sdxl-zimage`, tiles,
 resolution caps, humanize, unsharp, and adaptive polish. Those combinations change

docs/python-api.md

@@ -615,6 +615,10 @@ donor uses the same overlapping tiles as the global pass. `InvisibleOptions` exp
 same field for `remove_all`; after a visible-stage edit, the manifest must be built
 against the staged pixels rather than the pristine source.
 
+Use manifest schema 1 for manually reviewed text plus script metadata. Automated
+operators that verify only text-region geometry should emit schema 2 lines with a
+`box` and optional `angle`; no placeholder transcription or script is required.
+
 Since 0.27.1 the mode's global 15% Qwen-VAE fidelity-anchor blend is **off by
 default** (`fidelity_anchor=False`): that whole-frame blend was measured to
 return detector-visible OpenAI SynthID on poster-scale manifests (official

docs/watermarking-landscape.md

@@ -10,7 +10,7 @@ Who embeds what, and whether it is locally detectable (so we know which gaps are
   **Root cause and external confirmation (deep-research 2026-06-19, adversarially verified).** This is the SCHEME's ceiling, not our usage — there is no better decoder to adopt. The imwatermark maintainers state verbatim (both the ShieldMnt and Stability-AI READMEs) that the algorithm "cannot guarantee to decode the original watermarks 100% accurately even though we don't apply any attack." Independent measurement (WMAdapter, arXiv:2406.08337 Table 2) puts dwtDct at only **~0.79 bit accuracy on CLEAN images (~38/48 bits — already below our 44 gate)**, collapsing to ~0.50 (chance) under crop/JPEG. Two code-verified + locally-reproduced mechanisms drive the content-dependent failures: (1) the decoder reads each bit as the **highest-magnitude DCT coefficient per block**, so any content coefficient exceeding the encoded target flips the bit; (2) the default embed is in the **YUV chroma channel, which 8-bit-clamps on white/bright pixels** (a +36 chroma delta survives a white-fill round-trip as only +4, ~89% loss) — this is the mechanism behind the bright-flat / minimalist failures and the all-ones degenerate decode. No maintained fork or detector decodes this scheme reliably: the WAVES benchmark (arXiv:2401.08573) relegates DWT-DCT to supplementary appendix G.5 and targets Stable Signature / Tree-Ring / StegaStamp instead; learned encoder/decoder schemes reach ~0.98-0.99 clean but are a DIFFERENT watermark class (not what SDXL/FLUX stamp). `dwtDctSvd` does not help (SDXL embeds `dwtDct`; dwtDctSvd cannot decode it, and its clean accuracy ~0.72 is lower). **Authoritative conclusion: the open DWT-DCT mark cannot be turned from positive-only into a reliable real-world detector; keep it positive-only and rely on C2PA.** (Refuted along the way: that the library is unmaintained, and that it is robust to JPEG but only fails on geometric attacks — both did not survive verification.)
 
   Consequence for the FLUX hosted-output question (BFL Playground, FLUX.2 [pro] + FLUX.1 [dev], 2026-06-19): all samples carry the signed C2PA manifest (issuer "Black Forest Labs"); the open DWT-DCT decode returned `None`, but every available FLUX carrier (textured fox AND a minimalist-flat generation) failed the positive control (28/48), so the detector is blind on them and **whether BFL hosted output embeds the open pixel watermark is UNRESOLVED** (an earlier note here wrongly asserted it absent — overstated; a later note blamed "high texture" — also wrong, flat carriers fail too). What IS established: C2PA is the reliable FLUX identifier; the `_BITS_48` pattern is correct (round-trips on chatgpt/firefly/random). Resolving the hosted question needs a hosted FLUX carrier that first passes a >=44/48 positive control, which neither a textured nor a flat prompt produced — low priority (the open mark is only a stripped-metadata fallback).
-- **C2PA / IPTC (covered by the issuer/marker scan):** OpenAI, Google, Adobe Firefly, Microsoft (Designer + **Bing Image Creator** — collected 2026-05-24; Bing now runs Microsoft's own **MAI-Image** model, signs C2PA as "Microsoft", NOT OpenAI/DALL-E), **Stability AI** (collected from Brand Studio / DreamStudio successor; signs C2PA as "Stability AI Ltd", no SynthID, no imwatermark on its current Stable Image model — issuer added to `C2PA_ISSUERS`), and **Canva** (Magic Media signs C2PA as "Canva" + `trainedAlgorithmicMedia` with a generic `c2pa-rs` claim generator, no SynthID — issuer `b"Canva"` → "Canva (Magic Media)"; verified samples disproved the earlier assumption that Canva downloads always strip C2PA). Still unsampled: Getty, Shutterstock. Midjourney embeds NO C2PA and no invisible watermark (our `mj-*` sample carried only the IPTC tag).
+- **C2PA / IPTC (covered by the issuer/marker scan):** OpenAI, Google, Adobe Firefly, Microsoft (Copilot + Designer; Bing Image Creator collected 2026-05-24 still signs as "Microsoft" and now runs **MAI-Image**, NOT OpenAI/DALL-E), **Stability AI** (collected from Brand Studio / DreamStudio successor; signs C2PA as "Stability AI Ltd", no SynthID, no imwatermark on its current Stable Image model — issuer added to `C2PA_ISSUERS`), and **Canva** (Magic Media signs C2PA as "Canva" + `trainedAlgorithmicMedia` with a generic `c2pa-rs` claim generator, no SynthID — issuer `b"Canva"` → "Canva (Magic Media)"; verified samples disproved the earlier assumption that Canva downloads always strip C2PA). Still unsampled: Getty, Shutterstock. Midjourney embeds NO C2PA and no invisible watermark (our `mj-*` sample carried only the IPTC tag).
 
 **Samsung Galaxy AI** signs supported edits with C2PA and may carry the
 proprietary `genAIType` marker. The registered visible detector covers the
@@ -28,7 +28,7 @@ take precedence over issuer attribution. Supported mappings include Higgsfield
 AI, Topaz Labs Image API, and TikTok Ad Creative Toolbox; an unknown claim
 generator still falls back to the certificate issuer.
 
-**ByteDance Volcano Engine (Volcengine)** — the cloud behind Doubao / Jimeng — signs its AI image output with a cert from `certificate_center@volcengine.com` + `trainedAlgorithmicMedia` (issuer `b"volcengine"` → "ByteDance (Volcano Engine)", platform "ByteDance (Doubao / Jimeng / Volcano Engine)"); note this is the C2PA-signed surface, distinct from the XMP/PNG TC260 `AIGC` label Doubao also uses. ByteDance's **international brand (BytePlus / Seedream / Seededit)** signs the same content as **"Byteplus Pte. Ltd."**. The bare `volcengine` needle missed it, so BytePlus output was mis-attributed to "Adobe Firefly" through an incidental "Adobe XMP" toolkit string. Issuer `b"Byteplus"` now maps directly to "BytePlus (ByteDance)". ByteDance's consumer app **Dreamina** (the international Jimeng brand) signs as **"Bytedance Pte. Ltd."** with a `Dreamina/x.y` claim generator but, unlike the Volcano Engine surface, ships **no `trainedAlgorithmicMedia`**. Issuer `b"Dreamina"` maps to "ByteDance (Dreamina)" with **`asserts_ai=True`**. Registering the broader **issuer** `b"Bytedance Pte"` was deliberately avoided because that same entity also signs non-AI CapCut edits; keying on the `Dreamina` generator token is precise.
+**ByteDance Volcano Engine (Volcengine)** — the cloud behind Doubao / Jimeng — signs its AI image output with a cert from `certificate_center@volcengine.com` + `trainedAlgorithmicMedia` (issuer `b"volcengine"` → "ByteDance (Volcano Engine)", platform "ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)"); note this is the C2PA-signed surface, distinct from the XMP/PNG TC260 `AIGC` label Doubao also uses. ByteDance's **international brand (BytePlus / Seedream / Seededit)** signs the same content as **"Byteplus Pte. Ltd."**. The bare `volcengine` needle missed it, so BytePlus output was mis-attributed to "Adobe Firefly" through an incidental "Adobe XMP" toolkit string. Issuer `b"Byteplus"` now maps directly to "BytePlus (ByteDance)". ByteDance's consumer app **Dreamina** (the international Jimeng brand) signs as **"Bytedance Pte. Ltd."** with a `Dreamina/x.y` claim generator but, unlike the Volcano Engine surface, ships **no `trainedAlgorithmicMedia`**. Issuer `b"Dreamina"` maps to "ByteDance (Dreamina)" with **`asserts_ai=True`**. Registering the broader **issuer** `b"Bytedance Pte"` was deliberately avoided because that same entity also signs non-AI CapCut edits; keying on the `Dreamina` generator token is precise.
 - **EXIF/XMP/PNG-text generator tag (caught by `exif_generator`):** **Ideogram** writes EXIF `Make="Ideogram AI"` (collected 2026-05-24 — no C2PA, no SynthID, no imwatermark; the Make tag is the only signal). Additional verified generator stamps include **NovelAI** (`Software`, `Source`, and `Title` PNG text chunks), **Reve** (`Software` or XMP `CreatorTool` = `reve.com`), and **Aphrodite AI** (`Make` or `Software` = `Aphrodite AI`).
 - **App-export provenance and AIGC JSON:** supported ByteDance-family exports can place a JSON object in EXIF `ImageDescription` or `UserComment`, independently of C2PA or TC260. Exact `product` values for Doubao, Xinghui, and Dreamina are removable product provenance, but do not alone prove that the pixels were generated. Dreamina additionally requires `exportType=generation` for that verdict. A nested Aweme `aigc_type=1` or private ByteDance `aigc_label_type=1` / `2` is an AIGC disclosure; `0` is inconclusive and can occur on a Dreamina generation export. Plain Aweme, retouch, and `lv` exports are preserved. The lower-case private field is deliberately not interpreted as the normative TC260 `Label`, whose values `1` / `2` / `3` mean generated / possibly generated / suspected generated under [GB 45438-2025](https://www.tc260.org.cn/upload/2025-03-15/1742009439794081593.pdf).
 - **xAI / Grok — its own EXIF signature scheme, NOT C2PA (DETECTED by `metadata.xai_signature`, built 2026-05-26).**

pyproject.toml

@@ -4,7 +4,7 @@
 # on it, including the ComfyUI node package. The console script below carries
 # the same weight, since users have it on PATH.
 name = "remove-ai-watermarks"
-version = "0.28.1"
+version = "0.29.0"
 description = "AI watermark remover for visible, invisible, and provenance marks in images and video"
 readme = "README.md"
 requires-python = ">=3.11,<3.15"

src/remove_ai_watermarks/__init__.py

@@ -32,7 +32,7 @@ _os.environ.setdefault("TRANSFORMERS_VERBOSITY", "error")
 _warnings.filterwarnings("ignore", message=r".*ImageProcessorFast.*")
 
 
-__version__ = "0.28.1"
+__version__ = "0.29.0"
 
 __all__ = [
     "BatchSummary",

src/remove_ai_watermarks/_internal/constants.py

@@ -65,7 +65,7 @@ def _vendor(
 
 # Order is product priority when a manifest mentions more than one organization.
 C2PA_AI_VENDORS: tuple[C2paAiVendor, ...] = (
-    _vendor(b"Microsoft", "Microsoft", "Microsoft (Bing Image Creator / Designer)", "Microsoft"),
+    _vendor(b"Microsoft", "Microsoft", "Microsoft (Copilot / Designer)", "Microsoft"),
     _vendor(b"Adobe", "Adobe", "Adobe Firefly", "Adobe"),
     _vendor(
         b"OpenAI",
@@ -78,18 +78,25 @@ C2PA_AI_VENDORS: tuple[C2paAiVendor, ...] = (
     _vendor(b"Google", "Google LLC", "Google (Gemini / Imagen)", "Google", synthid=True),
     _vendor(b"Stability AI", "Stability AI", "Stability AI (Stable Image / DreamStudio)", "Stability AI"),
     _vendor(b"Black Forest Labs", "Black Forest Labs", "Black Forest Labs (FLUX)", "Black Forest Labs"),
-    _vendor(b"volcengine", "ByteDance (Volcano Engine)", "ByteDance (Doubao / Jimeng / Volcano Engine)", "ByteDance"),
+    _vendor(
+        b"volcengine",
+        "ByteDance (Volcano Engine)",
+        "ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)",
+        "ByteDance",
+    ),
     _vendor(
         "北京火山引擎科技有限公司",
         "ByteDance (Volcano Engine)",
-        "ByteDance (Doubao / Jimeng / Volcano Engine)",
+        "ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)",
         "ByteDance",
     ),
-    _vendor(b"Byteplus", "BytePlus (ByteDance)", "ByteDance (Doubao / Jimeng / Volcano Engine)", "ByteDance"),
+    _vendor(
+        b"Byteplus", "BytePlus (ByteDance)", "ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)", "ByteDance"
+    ),
     _vendor(
         b"Dreamina",
         "ByteDance (Dreamina)",
-        "ByteDance (Doubao / Jimeng / Volcano Engine)",
+        "ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)",
         "ByteDance",
         asserts_ai=True,
     ),
@@ -109,7 +116,7 @@ C2PA_IDENTITY_AI_ORGS = frozenset(vendor.org for vendor in C2PA_AI_VENDORS if ve
 C2PA_CLAIM_GENERATOR_PLATFORMS: tuple[tuple[str, str], ...] = (
     ("adobe_firefly", "Adobe Firefly"),
     ("firefly", "Adobe Firefly"),
-    ("dreamina", "ByteDance (Doubao / Jimeng / Volcano Engine)"),
+    ("dreamina", "ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)"),
     ("higgsfield ai", "Higgsfield AI"),
     ("topaz labs image api", "Topaz Labs"),
     ("tiktok ad creative toolbox", "TikTok Ad Creative Toolbox"),

src/remove_ai_watermarks/_internal/text_restoration.py

@@ -13,30 +13,35 @@ import cv2
 import numpy as np
 from PIL import Image
 
+from remove_ai_watermarks._internal.schema import require_schema_version
+
 if TYPE_CHECKING:
     from collections.abc import Sequence
     from pathlib import Path
 
     from numpy.typing import NDArray
 
-TEXT_MANIFEST_SCHEMA = 1
+TEXT_MANIFEST_SCHEMA = 2
+_SUPPORTED_TEXT_MANIFEST_SCHEMAS = frozenset({1, TEXT_MANIFEST_SCHEMA})
 FIDELITY_BLEND_ALPHA = 0.15
 GLYPH_FEATHER = 0.5
+GLYPH_SIDE_PAD_RATIO = 0.12
+GLYPH_SIDE_PAD_LIMIT = 1.0
 
 
 @dataclass(frozen=True)
 class VerifiedTextLine:
-    """One operator-verified source line in source-pixel coordinates."""
+    """One operator-verified source region in source-pixel coordinates."""
 
     box: tuple[int, int, int, int]
-    text: str
-    script: str
+    text: str | None = None
+    script: str | None = None
     angle: float = 0.0
 
 
 @dataclass(frozen=True)
 class VerifiedTextManifest:
-    """Text annotations cryptographically bound to one decoded RGB source."""
+    """Verified text regions cryptographically bound to one decoded RGB source."""
 
     source_pixel_sha256: str
     width: int
@@ -55,17 +60,20 @@ def source_pixel_sha256(image: Image.Image) -> str:
 
 
 def load_verified_text_manifest(path: Path, source: Image.Image) -> VerifiedTextManifest:
-    """Load and validate a manually verified manifest for exactly ``source``."""
+    """Load and validate an operator-verified manifest for exactly ``source``."""
     try:
         payload = json.loads(path.read_text(encoding="utf-8"))
     except (OSError, json.JSONDecodeError) as exc:
         raise ValueError(f"Cannot read text manifest {path}: {exc}") from exc
     if not isinstance(payload, dict):
         raise ValueError("Text manifest must be a JSON object")
-    if payload.get("schema_version") != TEXT_MANIFEST_SCHEMA:
-        raise ValueError(f"Text manifest schema_version must be {TEXT_MANIFEST_SCHEMA}")
+    schema_version = require_schema_version(
+        payload.get("schema_version"),
+        contract="text manifest",
+        supported=_SUPPORTED_TEXT_MANIFEST_SCHEMAS,
+    )
     if payload.get("verified") is not True:
-        raise ValueError("Text manifest must contain verified=true after manual review")
+        raise ValueError("Text manifest must contain verified=true after operator verification")
 
     rgb = source.convert("RGB")
     width = _manifest_integer(payload, "width")
@@ -82,7 +90,7 @@ def load_verified_text_manifest(path: Path, source: Image.Image) -> VerifiedText
     raw_lines = payload.get("lines")
     if not isinstance(raw_lines, list) or not raw_lines:
         raise ValueError("Text manifest lines must be a non-empty list")
-    lines = tuple(_load_line(item, width, height, index) for index, item in enumerate(raw_lines))
+    lines = tuple(_load_line(item, width, height, index, schema_version) for index, item in enumerate(raw_lines))
     if list(lines) != sorted(lines, key=lambda line: (line.box[1], line.box[0])):
         raise ValueError("Text manifest lines must be in top-to-bottom, left-to-right reading order")
     return VerifiedTextManifest(actual_hash, width, height, lines)
@@ -95,7 +103,7 @@ def _manifest_integer(payload: dict[str, Any], key: str) -> int:
     return value
 
 
-def _load_line(item: Any, width: int, height: int, index: int) -> VerifiedTextLine:
+def _load_line(item: Any, width: int, height: int, index: int, schema_version: int) -> VerifiedTextLine:
     if not isinstance(item, dict):
         raise ValueError(f"Text manifest line {index} must be an object")
     raw_box = item.get("box")
@@ -109,12 +117,15 @@ def _load_line(item: Any, width: int, height: int, index: int) -> VerifiedTextLi
     x1, y1, x2, y2 = box
     if not (0 <= x1 < x2 <= width and 0 <= y1 < y2 <= height):
         raise ValueError(f"Text manifest line {index} box is outside the source dimensions")
-    text = item.get("text")
-    script = item.get("script")
-    if not isinstance(text, str) or not text.strip():
-        raise ValueError(f"Text manifest line {index} text must be non-empty")
-    if not isinstance(script, str) or not script.strip():
-        raise ValueError(f"Text manifest line {index} script must be non-empty")
+    text: str | None = None
+    script: str | None = None
+    if schema_version == 1:
+        text = item.get("text")
+        script = item.get("script")
+        if not isinstance(text, str) or not text.strip():
+            raise ValueError(f"Text manifest line {index} text must be non-empty")
+        if not isinstance(script, str) or not script.strip():
+            raise ValueError(f"Text manifest line {index} script must be non-empty")
     angle_value = item.get("angle", 0.0)
     if isinstance(angle_value, bool) or not isinstance(angle_value, int | float):
         raise ValueError(f"Text manifest line {index} angle must be numeric")
@@ -185,28 +196,41 @@ def restore_verified_text(
     return Image.fromarray(restored)
 
 
-def _glyph_crop_box(box: tuple[int, int, int, int], width: int, height: int) -> tuple[int, int, int, int]:
-    """Widen the detector box so descenders stay inside the silhouette crop.
+def _glyph_crop_box(
+    box: tuple[int, int, int, int],
+    width: int,
+    height: int,
+    left_pad: int,
+    right_pad: int,
+) -> tuple[int, int, int, int]:
+    """Widen the detector box so glyph edges stay inside the silhouette crop.
 
-    Paddle line boxes sit 2-5 px above the true ink bottom on the poster
-    fixtures. The crop is the box itself, so those pixels never reached the
-    donor composite. Expand only on Y: 8% up, 25% down, clamped to the frame.
+    Detector boxes can stop inside a leading flourish or trailing punctuation,
+    and Paddle line boxes sit 2-5 px above the true ink bottom on the poster
+    fixtures. The crop is otherwise the box itself, so those pixels never reach
+    the donor composite. Start 12% sideways, expand each side independently
+    while foreground reaches its boundary, and pad 8% up and 25% down.
     """
     x1, y1, x2, y2 = box
     line_h = max(1, y2 - y1)
     pad_top = max(1, round(line_h * 0.08))
     pad_bot = max(2, round(line_h * 0.25))
-    return max(0, x1), max(0, y1 - pad_top), min(width, x2), min(height, y2 + pad_bot)
+    return (
+        max(0, x1 - left_pad),
+        max(0, y1 - pad_top),
+        min(width, x2 + right_pad),
+        min(height, y2 + pad_bot),
+    )
 
 
-def source_silhouette_mask(
+def _silhouette_crop_mask(
     source_rgb: NDArray[Any],
-    box: tuple[int, int, int, int],
-    angle: float = 0.0,
+    crop_box: tuple[int, int, int, int],
+    angle: float,
 ) -> NDArray[Any]:
-    """Recover a thresholded glyph shape without retaining source amplitudes."""
+    """Threshold one candidate crop so its horizontal boundaries can be tested."""
     height, width = source_rgb.shape[:2]
-    x1, y1, x2, y2 = _glyph_crop_box(box, width, height)
+    x1, y1, x2, y2 = crop_box
     gray = cv2.cvtColor(source_rgb[y1:y2, x1:x2], cv2.COLOR_RGB2GRAY)
     support = np.ones(gray.shape, dtype=np.uint8)
     if angle:
@@ -225,7 +249,7 @@ def source_silhouette_mask(
         background_luma = float(np.median(values))
     else:
         ring_pad = max(6, min(20, (y2 - y1) // 4))
-        rx1, ry1, rx2, ry2 = _clip_box((x1, y1, x2, y2), width, height, pad=ring_pad)
+        rx1, ry1, rx2, ry2 = _clip_box(crop_box, width, height, pad=ring_pad)
         context = cv2.cvtColor(source_rgb[ry1:ry2, rx1:rx2], cv2.COLOR_RGB2GRAY)
         ring = np.ones(context.shape, dtype=bool)
         ring[y1 - ry1 : y2 - ry1, x1 - rx1 : x2 - rx1] = False
@@ -239,6 +263,61 @@ def source_silhouette_mask(
     else:
         crop_mask = (gray.astype(np.float32) <= background_luma - threshold).astype(np.uint8) * 255
     crop_mask[support == 0] = 0
+    return crop_mask
+
+
+def _anchored_components_reach_sides(
+    crop_mask: NDArray[Any],
+    anchor_box: tuple[int, int, int, int],
+) -> tuple[bool, bool]:
+    """Whether anchored foreground reaches the left and right crop sides."""
+    count, labels, stats, _centroids = cv2.connectedComponentsWithStats(crop_mask, 8)
+    x1, y1, x2, y2 = anchor_box
+    anchor_counts = np.bincount(labels[y1:y2, x1:x2].reshape(-1), minlength=count)
+    anchor_counts[0] = 0
+
+    def reaches(edge_labels: NDArray[Any]) -> bool:
+        return any(
+            label != 0 and anchor_counts[label] >= max(3, round(stats[label, cv2.CC_STAT_AREA] * 0.1))
+            for label in np.unique(edge_labels)
+        )
+
+    return reaches(labels[y1:y2, :2]), reaches(labels[y1:y2, -2:])
+
+
+def source_silhouette_mask(
+    source_rgb: NDArray[Any],
+    box: tuple[int, int, int, int],
+    angle: float = 0.0,
+) -> NDArray[Any]:
+    """Recover a thresholded glyph shape without retaining source amplitudes."""
+    height, width = source_rgb.shape[:2]
+    box_x1, box_y1, box_x2, box_y2 = box
+    line_h = max(1, box_y2 - box_y1)
+    step = max(2, round(line_h * GLYPH_SIDE_PAD_RATIO))
+    limit = max(step, round(line_h * GLYPH_SIDE_PAD_LIMIT))
+    left_pad = right_pad = step
+    while True:
+        x1, y1, x2, y2 = _glyph_crop_box(box, width, height, left_pad, right_pad)
+        crop_mask = _silhouette_crop_mask(source_rgb, (x1, y1, x2, y2), angle)
+        core_y1 = max(0, box_y1 - y1)
+        core_y2 = min(y2 - y1, box_y2 - y1)
+        anchor_box = (max(0, box_x1 - x1), core_y1, min(x2 - x1, box_x2 - x1), core_y2)
+        can_expand_left = x1 > 0 and left_pad < limit
+        can_expand_right = x2 < width and right_pad < limit
+        left_anchored, right_anchored = (
+            _anchored_components_reach_sides(crop_mask, anchor_box)
+            if can_expand_left or can_expand_right
+            else (False, False)
+        )
+        left_touches = can_expand_left and left_anchored
+        right_touches = can_expand_right and right_anchored
+        if not left_touches and not right_touches:
+            break
+        if left_touches:
+            left_pad = min(limit, left_pad + step)
+        if right_touches:
+            right_pad = min(limit, right_pad + step)
     result = np.zeros((height, width), dtype=np.uint8)
     result[y1:y2, x1:x2] = crop_mask
     return result

src/remove_ai_watermarks/cli.py

@@ -904,10 +904,27 @@ def cmd_invisible(
 
 
 # ── Metadata operations ──
+def _print_metadata_not_a_clean_verdict() -> None:
+    """Repeat the identify empty-scan limit on metadata check and strip success.
+
+    ``metadata --check`` and ``metadata --remove`` answer a narrower question than
+    ``identify``: they report embedded AI metadata only. A quiet result used to
+    stop at "No AI metadata found" / "AI metadata stripped", which readers treat
+    as a clean-image verdict. The pixel channel is unchanged, and this project has
+    no local SynthID decoder, so the command must say so in the same words
+    ``identify`` already uses.
+    """
+    console.print(
+        "  This is not the same as 'clean': a pixel watermark such as SynthID cannot be\n"
+        "  detected here once its metadata proxy is absent."
+    )
+
+
 def _print_metadata_report(source: Path, has_ai: bool, metadata: dict[str, str]) -> None:
     """Render one metadata inspection result for the generic and video commands."""
     if not has_ai:
         console.print(f"  No AI metadata found in {source.name}")
+        _print_metadata_not_a_clean_verdict()
         return
 
     console.print(f"  Warning: AI metadata detected in {source.name}:")
@@ -974,6 +991,7 @@ def cmd_metadata(
         console.print("    the file could not be decoded, so it was copied through unchanged")
         raise SystemExit(1)
     console.print(f"  AI metadata stripped -> {out}")
+    _print_metadata_not_a_clean_verdict()
 
 
 # ── Video pipeline ──
@@ -1115,6 +1133,7 @@ def cmd_video_metadata(
         console.print(f"    still present: {', '.join(sorted(result.remaining))}")
         raise SystemExit(1)
     console.print(f"  AI metadata stripped -> {result.output}")
+    _print_metadata_not_a_clean_verdict()
 
 
 @cmd_video.command("invisible")

src/remove_ai_watermarks/identify.py

@@ -770,6 +770,7 @@ _AI_VENDOR_TOKENS: tuple[tuple[str, str], ...] = (
     ("google", "Google"),
     ("firefly", "Adobe"),
     ("adobe", "Adobe"),
+    ("copilot", "Microsoft"),
     ("bing", "Microsoft"),
     ("designer", "Microsoft"),
     ("microsoft", "Microsoft"),

src/remove_ai_watermarks/qwen_engine.py

@@ -1,4 +1,4 @@
-"""Qwen (Tongyi Qianwen, Alibaba) visible watermark detector/localizer.
+"""Qwen (Alibaba) visible watermark detector/localizer.
 
 Qwen stamps its generations with a visible "千问AI生成" text strip in the
 bottom-right corner -- the explicit AIGC label mandated by China's GB 45438-2025

src/remove_ai_watermarks/watermark_registry.py

@@ -20,7 +20,7 @@ Entries:
   - ``gemini`` -- Google Gemini / Nano Banana sparkle, bottom-right.
   - ``doubao`` -- ByteDance Doubao "豆包AI生成" text strip, bottom-right.
   - ``jimeng`` -- ByteDance Jimeng / Dreamina "★ 即梦AI" wordmark, bottom-right.
-  - ``qwen`` -- Alibaba Tongyi Qianwen "千问AI生成" text strip, bottom-right.
+  - ``qwen`` -- Alibaba Qwen "千问AI生成" text strip, bottom-right.
   - ``kling`` -- Kuaishou Kling "可灵AI 3.0" text strip, bottom-right.
   - ``yuanbao`` -- Tencent Yuanbao "元宝 / AI生成" two-line mark, bottom-right.
   - ``samsung`` -- Samsung Galaxy AI "Contenuti generati dall'AI" strip, bottom-left.
@@ -651,7 +651,7 @@ _REGISTRY: tuple[KnownMark, ...] = (
         "qwen",
         "Qwen 千问AI生成 text",
         "bottom-right",
-        platform="Alibaba Tongyi Qianwen (visible 千问AI生成 mark detected)",
+        platform="Alibaba Qwen (visible 千问AI生成 mark detected)",
         tc260_producer_codes=("91440101MA9Y9T4H7A",),
     ),
     _text_mark(

tests/test_cli.py

@@ -629,11 +629,13 @@ class TestMetadataCommand:
         result = runner.invoke(main, ["metadata", str(tmp_clean_png), "--check"])
         assert result.exit_code == 0
         assert "No AI metadata" in result.output
+        assert "not the same as 'clean'" in result.output
 
     def test_metadata_check_ai(self, runner, tmp_png_with_ai_metadata):
         result = runner.invoke(main, ["metadata", str(tmp_png_with_ai_metadata), "--check"])
         assert result.exit_code == 0
         assert "AI metadata detected" in result.output
+        assert "not the same as 'clean'" not in result.output
 
     def test_metadata_remove(self, runner, tmp_png_with_ai_metadata, tmp_path):
         output = tmp_path / "stripped.png"
@@ -649,6 +651,7 @@ class TestMetadataCommand:
         )
         assert result.exit_code == 0
         assert "stripped" in result.output
+        assert "not the same as 'clean'" in result.output
 
     def test_metadata_remove_reports_failure_when_the_strip_was_a_no_op(self, runner, tmp_path):
         """A file PIL cannot decode is copied through UNCHANGED by the fail-safe.

tests/test_identify.py

@@ -16,6 +16,7 @@ from unittest.mock import patch
 
 import pytest
 
+from remove_ai_watermarks._internal.constants import C2PA_AI_VENDORS, C2PA_CLAIM_GENERATOR_PLATFORMS
 from remove_ai_watermarks.identify import (
     ProvenanceEvidence,
     ProvenanceReport,
@@ -1300,12 +1301,20 @@ class TestVendorOf:
     def test_registered_vendors_normalize(self):
         # Regression: these registered C2PA vendors returned None, so their claims never
         # entered clash detection (a coverage hole). They now normalize to one origin.
-        assert _vendor_of("ByteDance (Doubao / Jimeng / Volcano Engine)") == "ByteDance"
+        assert _vendor_of("Microsoft (Copilot / Designer)") == "Microsoft"
+        assert _vendor_of("Copilot") == "Microsoft"
+        assert _vendor_of("ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)") == "ByteDance"
         assert _vendor_of("Dreamina/1.2") == "ByteDance"
         assert _vendor_of("Canva (Magic Media)") == "Canva"
         assert _vendor_of("Black Forest Labs (FLUX)") == "Black Forest Labs"
         assert _vendor_of("Eleven Labs Inc.") == "ElevenLabs"
 
+    def test_bytedance_issuers_share_one_platform(self):
+        expected = "ByteDance (Doubao / Jimeng / Dreamina / Volcano Engine)"
+        platforms = {vendor.platform for vendor in C2PA_AI_VENDORS if vendor.needle == "ByteDance"}
+        assert platforms == {expected}
+        assert ("dreamina", expected) in C2PA_CLAIM_GENERATOR_PLATFORMS
+
 
 class TestIntegrityClashesHelper:
     def test_two_ai_vendors_clash(self):

tests/test_text_restoration.py

@@ -37,6 +37,17 @@ def _manifest(image: Image.Image) -> dict[str, object]:
     }
 
 
+def _geometry_manifest(image: Image.Image) -> dict[str, object]:
+    return {
+        "schema_version": 2,
+        "verified": True,
+        "source_pixel_sha256": source_pixel_sha256(image),
+        "width": image.width,
+        "height": image.height,
+        "lines": [{"box": [8, 8, 40, 24], "angle": 0.0}],
+    }
+
+
 def test_pixel_hash_ignores_container_metadata(tmp_path) -> None:
     image = Image.new("RGB", (48, 32), (10, 20, 30))
     plain = tmp_path / "plain.png"
@@ -63,6 +74,40 @@ def test_verified_manifest_is_bound_to_source_pixels(tmp_path) -> None:
     assert loaded.lines == (VerifiedTextLine((8, 8, 40, 24), "Exact text", "alphabetic", 0.0),)
 
 
+def test_geometry_manifest_needs_no_transcription_or_script(tmp_path) -> None:
+    source = Image.new("RGB", (48, 32), (10, 20, 30))
+    path = tmp_path / "regions.json"
+    path.write_text(json.dumps(_geometry_manifest(source)), encoding="utf-8")
+
+    loaded = load_verified_text_manifest(path, source)
+
+    assert loaded.lines == (VerifiedTextLine((8, 8, 40, 24), angle=0.0),)
+
+
+@pytest.mark.parametrize("field", ["text", "script"])
+def test_schema_one_still_requires_text_metadata(tmp_path, field) -> None:
+    source = Image.new("RGB", (48, 32), (10, 20, 30))
+    payload = _manifest(source)
+    del payload["lines"][0][field]
+    path = tmp_path / "lines.json"
+    path.write_text(json.dumps(payload), encoding="utf-8")
+
+    with pytest.raises(ValueError, match=field):
+        load_verified_text_manifest(path, source)
+
+
+@pytest.mark.parametrize("schema_version", [True, 0, 3])
+def test_manifest_rejects_unsupported_schema_versions(tmp_path, schema_version) -> None:
+    source = Image.new("RGB", (48, 32), (10, 20, 30))
+    payload = _geometry_manifest(source)
+    payload["schema_version"] = schema_version
+    path = tmp_path / "lines.json"
+    path.write_text(json.dumps(payload), encoding="utf-8")
+
+    with pytest.raises(ValueError, match="Unsupported text manifest schema"):
+        load_verified_text_manifest(path, source)
+
+
 @pytest.mark.parametrize(
     ("mutation", "message"),
     [
@@ -120,7 +165,7 @@ def test_restoration_uses_lama_and_qwen_vae_core(monkeypatch) -> None:
         Image.fromarray(source),
         Image.fromarray(candidate),
         Image.fromarray(donor),
-        (VerifiedTextLine((8, 8, 48, 28), "Exact text", "alphabetic"),),
+        (VerifiedTextLine((8, 8, 48, 28)),),
     )
 
     restored = np.asarray(result)
@@ -138,3 +183,17 @@ def test_silhouette_includes_descender_below_the_detector_box() -> None:
 
     assert mask[24, 20] == 255
     assert mask[16, 20] == 255
+
+
+def test_silhouette_includes_glyph_edges_beside_the_detector_box() -> None:
+    source = np.full((40, 70, 3), 240, dtype=np.uint8)
+    source[10:22, 20:50] = 20
+    source[14:18, 10:20] = 20  # leading flourish, left of the detector box
+    source[14:18, 50:58] = 20  # punctuation or icon edge, right of the box
+    source[14:18, 2:5] = 20  # separate decoration must stay outside the crop
+
+    mask = source_silhouette_mask(source, (20, 10, 50, 22))
+
+    assert mask[16, 12] == 255
+    assert mask[16, 56] == 255
+    assert mask[16, 3] == 0

tests/test_video.py

@@ -951,6 +951,7 @@ class TestVideoMetadataCli:
 
         assert result.exit_code == 0, result.output
         assert "AI metadata detected" in result.output
+        assert "not the same as 'clean'" not in result.output
 
     def test_remove_reports_output(self, tmp_path: Path):
         runner = CliRunner()
@@ -961,6 +962,7 @@ class TestVideoMetadataCli:
 
         assert result.exit_code == 0, result.output
         assert "AI metadata stripped" in result.output
+        assert "not the same as 'clean'" in result.output
         assert C2PA_UUID not in output.read_bytes()
 
     def test_rejects_image_input(self, tmp_clean_png: Path):

uv.lock

@@ -3131,7 +3131,7 @@ wheels = [
 
 [[package]]
 name = "remove-ai-watermarks"
-version = "0.28.1"
+version = "0.29.0"
 source = { editable = "." }
 dependencies = [
     { name = "c2pa-python" },