- brand.toml is the only place the product name lives; every crate and app
reads it through dac-brand (env vars with legacy fallback, settings folder
with copy-only migration, default library, preset extensions, i18n {app})
- cargo xtask brand check|test, run|install|package, docs, rename-crates,
deny and immich; packaging files are templates rendered from brand.toml
- icons regenerated from brand/icon.svg; upstream branding removed
- MIT licence (segment and fetch stay Apache-2.0); cargo deny licence policy
- clean-room rules in AGENTS.md/CLAUDE.md; docs/upstream-merge.md
- plan/ rebuilt and committed; parity tracker reopens Classic rows, adds
PLAN.md section 2 rows and the IMM Immich section; plan/immich.md and a
pinned Immich test server
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Keep current upstream UI behavior and the later detector contract checks. Validate candidate models before replacing working weights and records, with a synthetic regression that failed before the fix. Remove the unrelated GPU equivalence-test relaxation. All seven quality gates pass with the pinned hosted fonts.
Generated with Codex.
New L1 crate `lightcraft-faces`, the base of a one-drop model install:
- `ModelManifest` (role, licence with a commercial yes / no / unknown flag and a
plain-language notice, provenance, input preprocessing, output, thresholds)
with strict validation: sizes capped, numbers finite and sane, https-only
addresses, bounded text. A manifest from any source is untrusted.
- `known`: models recognised by SHA-256 (YuNet, SFace, AuraFace v1), with what is
and is not known about their licences and training data.
- `onnx`: a streaming, hostile-input-safe protobuf probe of a model's graph
inputs and outputs. It reads a 248 MB model in about 4 ms and never loads the
weights; truncation at every length and random corruption never panic.
- `suggest`: known hash -> its manifest; an unknown model shaped like a face
embedder -> a draft manifest with the assumptions spelled out; anything else
-> a plain reason it is not supported yet.
- `hash`: streamed SHA-256.
Checked against the real model files with an opt-in test (LC_FACE_MODELS). It
found an old exporter listing every weight as an input, now handled. Running a
model (tract) is not part of this crate yet.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
