Share producer identities across mark routing, attribution, and pinned signers.
Preserve nested Honor TC260 fields and report unsupported RSA signatures.
Document the Google Photos checks and public sample search.
Use an SDK-signed synthetic carrier for negative EKU checks after the original
Microsoft certificates expired. Skip boundary regex scans when no C2PA
soft-binding identifier is present.
MiniMax signs its TC260 label with SM3withSM2 per TC260-PG-202511A. A
verification-only SM3/SM2 (no dependency provides the SM2 curve) checks the
appendix A.1 label message; a signature under the pinned MiniMax key is a
high-confidence aigc_signature signal, a failing one is a caveat, and every
result is the aigc_signature metadata marker. Pinned by the GB/T 32905 SM3
vectors, OpenSSL's SM2 verify vectors and real MiniMax files. Over 411 local
files the only verdict-record change is the marker on nine MiniMax videos.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The upstream list widened com.museblossom.contentsdefence.1 from image to
audio, image and video. Verdicts over 411 local files are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MiniMax writes a TC260-PG-202511A SecurityData object with SM3withSM2
signatures. The label signature verifies on all four local MiniMax videos
against the guide's appendix A.1 message, and the Binding signature over the
compact Bindings entry; the SM3 content hash does not match a naive selection.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Bump the package to 0.43.0 and the agent skill to 1.0.10. The skill now names
the wan image and vidu video mark keys, which 0.42.0 rejects, so the probe's
MIN_CLI_VERSION rises to 0.43.0 and both keys carry a "since CLI" note that
the floor test reads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
module-internals still called FLUX.2 klein pending; the sweep still counted
three inconclusive Gemini rows where only one was an explicit tool verdict,
and the removal page stated a Google chroma leak the data only warns of.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fidelity_metrics.py at each engine's clean strengths: klein keeps Latin and
Cyrillic sign lettering better on two Google carriers, but loses every face
comparison and the verified CJK text, which the 512-pixel LPIPS had hidden.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The fourth Google carrier clears at 0.25, so the klein operating point with
the usual cross-source margin is 0.40, where the limiting carrier is at
LPIPS 0.270 against qwen-zimage's 0.183. klein is not a replacement global
stage for either vendor. Also record that Gemini's image-check allowance is
per account and the new "checks on similar images" quota wording.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three of four Google carriers clear with klein at or below 0.20, where it
matches qwen-zimage's PSNR at 0.35, beats its LPIPS, and keeps the objects
qwen-zimage redraws. The fourth is detected at 0.10 and inconclusive at 0.20,
so there is no Google operating point yet.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Stock Flux2KleinInpaintPipeline feeds the clean source back as reference
tokens, so it copies the source and OpenAI SynthID survives strength 0.30.
Without those tokens klein clears OpenAI between 0.05 and 0.20, at lower
fidelity than qwen-zimage at its floor. Google is unmeasured: the Gemini
Verify AI tool answered with a tool error or quota limit, which also casts
doubt on two of the earlier color-restoration "inconclusive" rows.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Transplanting the source's a,b channels onto qwen-zimage outputs at the vendor
floors cut CIEDE2000 on all seven tracked originals, kept every OpenAI API
verdict not_detected, but left three of four Gemini checks inconclusive and
painted old object colors onto redrawn ones. Not shipped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The manifest stores forward-slash paths, and str() of a Windows path uses
backslashes, so the index test failed on both Windows CI jobs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Main's claim_generator_platform and c2pa_vendor_platform now use the
branch's whole-word registry match, which replaces
claim_generator_word_platform. The branch's provenance veto covers every
video mark and made main's Veo-over-Sora override unreachable, so the
override is removed. Main's credential rule carries into the veto: an
invalid credential vetoes nothing. Provenance that confirms the mark's
own vendor also vetoes nothing. Over 78 local videos with a C2PA AI
claim, the only verdict changes against main are Kling/Hailuo false
positives on Veo and FLUX clips that the veto now removes. C2PA verdicts
over 1,379 files are unchanged against the pre-merge branch.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- A decoded manifest no longer takes its generator from a raw
claim_generator CBOR value or a soft binding from raw bytes. Device tokens
(NIKON, Pixel Camera, sony.sig) stay a byte scan, since the decoded strings
do not carry them, but read the manifest store: an EXIF Make beside an
unrelated AI manifest is not a verified capture. The source-kind scan keeps
the whole buffer because it also reads IPTC digitalSourceType in XMP.
- The byte fallback's c2pa.ai-disclosure kind now uses the structured rule:
human_validated beside prompt_guided is a generation, not an enhancement.
- Wan and Yuanbao share BottomRightAnchoredEngine instead of two copies of
the corner gate. The Yuanbao anchor test shifted the mark outside the locate
window, so it passed with or without the gate; it now shifts it inside.
Verdicts over 1,379 C2PA files, 468 fixture and capture files, and Wan and
Yuanbao detections over 285 images are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Higgsfield's own DoP model ("Higgsfield Lite" in the web app, job model
image2video/lite) is not exposed by the MCP connector, so the capture was made
in the web app on a JPEG frame after a first job on a PNG failed and was
refunded. Like Genjutsu, the output carries no provenance: no C2PA, no TC260
label and no visible mark, only a Lavf re-mux.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Extend the whole-word rule to every identity lookup. Once the reader has
decoded a manifest, no identity is re-derived from its raw bytes: not the
signer, not a signer-platform token, not the AI-tool fallback for the
generator, and not the SynthID vendor inference, in identify and in
metadata.synthid_source alike. The byte scans read the manifest store only
when the reader could not open it.
Claim-generator tokens match as whole words through one shared
claim_generator_word_platform (suno no longer matches Sunoco), for images
and video alike.
The verdict now reads SynthID from the decoded C2PA info when the metadata
dict lacks it; three synthetic Google Photos tests had relied on the byte
fallback for that.
Verdicts over 1,379 C2PA files and 468 fixture and capture files are
unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Registry tokens were raw substrings, so the Canva token matched the "Nova
Canvas" agent of an Amazon Bedrock image. Raw manifest bytes cannot take a
word-boundary rule: across the 1,379 C2PA files in the local corpus, 374 of
37,325 store matches had a letter or digit beside them, nearly all CBOR text
headers and DER tags. Decoded strings have no such framing.
- Match signer, claim-generator, agent and video-marker strings with
registry_word_matches.
- When the reader decoded the manifest, never re-derive the signer from raw
bytes: an unregistered signer stays unknown. Bytes decide only for
manifests the reader cannot open.
- Add fal's signer "Features & Labels"; its attribution had relied on the byte
fallback finding the "fal-ai" generator prefix.
Verdicts over 468 images and 365 videos are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Nova Canvas images published in aws-samples were reported as Canva: with no
Amazon row, the byte fallback matched the Canva token inside the "Nova
Canvas" software agent. Add an Amazon Web Services row ahead of Canva, which
also names Nova Reel video.
When the reader cannot open a manifest, scan for signers in the C2PA store
the container locates (JPEG APP11, PNG caBX) instead of the whole metadata
region. A vivo X300 camera capture was labeled "C2PA signer: Adobe" from its
XMP toolkit string. The SynthID provenance scan reads the same store.
Adds the Nova Canvas and Nova Reel samples as MIT-0 fixtures and records the
public-sample search for the other conformance-list signers.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Collect provider samples from Higgsfield, Runway and Dreamina plus Google
Photos, Gemini, Grok and YouTube re-encodes into data/captures/2026-09 with
a hash-checked manifest, and use them to fix attribution and false positives:
- Attribute aggregator outputs: Higgsfield job metadata, Dreamina's
re-signed manifests, Runway's own C2PA signer, TC260 video producers.
- Veto any visible video mark when a valid C2PA claim names another vendor
(a Gemini Omni clip read as Sora on steam), and gate Kling and Hailuo
matches on edge, anchor and white-label evidence.
- Veto the Gemini sparkle when trusted provenance names another vendor.
- Report SynthID as present for Google Photos AI edits: Google's checker
found it on all four measured edits, though manifests record no action.
- Add Vidu video and Wan image marks, TC260 audio placements, and record
the AIGC key in metadata records.
- Document the research sweep, the Dreamina badge decision, and verified
Google statements on SynthID in Photos.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Lock OpenCV 5 and cap NumPy below 2.4 through uv constraint-dependencies so
PaddleX 3.7 keeps resolving; Dependabot ignores mirror the cap.
- OpenCV 5 draws Hershey putText glyphs differently, so the Dola and Kling
video templates ship as committed OpenCV 4 renders with a recipe script, and
the OpenArt alpha recipe now refuses any OpenCV but 4.x.
- Break Sora/Veo cross-matches in favor of Veo under valid Google AI-video
provenance, and require the full Sora mascot shape for mascot-only matches.
- Let the Doubao engine detect marks tinted by saturated backgrounds.
- Share C2PA claim-generator, credential-level and vendor-platform helpers,
JPEG suffixes and the DCT basis instead of per-module copies.
- Log model download failures, document the optional GITHUB_TOKEN, mark the
photo and source classifier model cards archived, and refresh the lock.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
scripts/sync_c2pa_soft_bindings.py --check hits api.github.com to resolve
the latest upstream commit SHA, unauthenticated. That endpoint's 60/hour
rate limit is per source IP, which GitHub Actions runners exhaust fast
since many workflows share the runner IP pool -- this was failing the
lint job with "403: rate limit exceeded" on every push regardless of
whether the change touched anything related.
Send an Authorization: Bearer header built from GITHUB_TOKEN/GH_TOKEN
when set, scoped to api.github.com only (raw.githubusercontent.com is a
separate, much higher-limit CDN and doesn't need it). Wire the workflow's
own GITHUB_TOKEN secret into that step's env so CI actually has one to
send, raising the limit to 5000/hour.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The OpenArt visible-mark engine (commit 9d29d2c) was registered without
a data/fixtures/visible/openart/example.png gallery canary and without
an _ALTERNATE_SIZE entry in test_visible_fill_quality.py, breaking
TestGallery::test_every_registered_mark_has_an_example,
TestGallery::test_engine_detects_its_own_example[openart], and
test_cv2_mask_quality_smoke[alternate-openart] on every CI platform.
Generate the fixture via render_visible_examples.py (self-detects at
conf=0.968) and add the missing size entry.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Every existing text-mark engine (Doubao, Jimeng, Qwen, Baidu, Kling,
Yuanbao) is tuned to one vendor's exact wordmark silhouette, so none of
them match a bare "AI生成" strip with no brand name in front of it.
Confirmed on raiw-app production: a free `visible` run left a bottom-right
"AI生成" watermark completely unremoved because no detector fired.
Research found the wording is not tied to one dominant generator: vivo's
own support documentation confirms its Gallery stamps a bare "AI生成"
label after AI消除/AI扩图/大片模式 editing, and independent reporting
documents the same watermark on Xiaomi Gallery's AI-edit output. Since
the label is a GB 45438-2025/TC260 compliance stamp rather than one
vendor's brand mark, this adds one generic fallback (`generic_ai_label`)
instead of another per-OEM tuned config.
The detection silhouette is font-rendered (no logo element to capture)
via the shared scripts/render_vendor_silhouettes.py pipeline already
used for Qwen/Baidu/Kling/Yuanbao's own text-only marks. Calibration is
synthetic-only (no real captured generator screenshot was available);
the module docstring and DETECT_NCC_THRESHOLD comment carry that caveat
explicitly and this should be recalibrated once real captures exist.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
OpenArt.ai stamps free-plan generations with a semi-transparent white
"OpenArt" wordmark (bowtie/infinity icon + brand name) placed near the
CENTER of the frame rather than in a corner. Confirmed from a customer-
reported raiw-app production case where the mark went completely
undetected and unremoved, with no C2PA/EXIF/TC260 signal for identify
to fall back on.
- Add a new "cc" (frame-center) corner mode to the shared
TextMarkConfig/TextMarkEngine base, alongside the existing
corner/edge anchors.
- Add openart_engine.py, following the existing text-mark engine
pattern. No real capture exists (OpenArt requires an account, out of
scope to create), so the bundled alpha asset is a procedural
reconstruction of the public logo and the NCC threshold is an
explicitly-labeled provisional guess, not a measured calibration.
- Register the mark in watermark_registry.py, which auto-derives
identify's platform attribution.
- Add tests/test_openart_engine.py (synthetic composites only) plus
registry/identify test coverage, and document the mark and its
calibration caveats in README.md, docs/supported-signals.md,
docs/module-internals.md, and the skill's commands.md.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
A decorative sparkle/glint glyph in stylized typography (e.g. a book-cover
title) can clear the visible Gemini sparkle's false-positive gate at
confidence just under _SPARKLE_FP_CONF, since it shares the same bright,
near-white, sharp four-point signature the gate uses to recognize a genuine
sparkle. Confirmed against a raiw-app production incident (2026-08-27) where
this caused the removal step to permanently smudge unrelated cover art.
Investigated tightening _SPARKLE_FP_CONF and the margin/gradient/saturation
checks (plus a prototyped local edge-density signal); each was disproved by
a constructed counter-example showing a genuine weak Gemini sparkle over a
noisy/textured photo background lands in the same confidence band with the
same signature, so no safe targeted fix exists without a real labeled
corpus. No detection/removal behavior changed.
Adds DEBUG logging at the exact FP-gate bypass path for future calibration,
a regression test (TestDecorativeGlyphFalsePositive) pinning the confirmed
false positive and the blocking counter-example, and a docs record of the
finding in the Gemini sparkle section of module-internals.md.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
