git.99rst.org Git - sgasser-llm-shield.git/commit

]> git.99rst.org Git - sgasser-llm-shield.git/commit

git.99rst.org / sgasser-llm-shield.git / commit

author	Stefan Gasser <redacted>
	Mon, 22 Jun 2026 10:29:05 +0000 (12:29 +0200)
committer	GitHub <redacted>
	Mon, 22 Jun 2026 10:29:05 +0000 (12:29 +0200)
commit	08ddb1d94911e59415d86460c0fd9e5baee12a4b
tree	96956369166a2068fe6a63a292ccaa434717b88a	tree
parent	747caa2fa1c6ce50a0cb4264f636fb6f2c796cc2	commit \| diff

Replace Presidio with an open-source PII detector (#100)

Replace the Microsoft Presidio analyzer with a self-hosted open-source PII detector (Python/FastAPI, detector/) shipped in the all-in-one image: multilingual GLiNER NER plus a deterministic regex/checksum layer for structured identifiers.

- Add VAT_CODE (EU VAT, checksum-validated); member-state prefixes only, case-insensitive and overlap-safe so a label or word can't hide a valid number.
- /api/mask: detect secrets before PII so a connection string isn't partly masked as an email (matches the provider routes).
- Rename presidio_url to detector_url; language-agnostic detection; CPU-only torch; detector CI (ruff/pyright/pytest).
- Bump version to 0.5.0.

53 files changed:

.conductor/settings.toml		diff \| blob \| history
.dockerignore	[new file with mode: 0644]	blob
.github/workflows/ci.yml		diff \| blob \| history
.github/workflows/release.yml		diff \| blob \| history
AGENTS.md		diff \| blob \| history
CONTRIBUTING.md		diff \| blob \| history
README.md		diff \| blob \| history
benchmarks/pii-accuracy/taxonomy.ts		diff \| blob \| history
benchmarks/pii-accuracy/test-data/core.yaml		diff \| blob \| history
benchmarks/pii-accuracy/test-data/precision.yaml		diff \| blob \| history
config.example.yaml		diff \| blob \| history
detector/.gitignore	[new file with mode: 0644]	blob
detector/detector/__init__.py	[new file with mode: 0644]	blob
detector/detector/app.py	[new file with mode: 0644]	blob
detector/detector/deterministic.py	[new file with mode: 0644]	blob
detector/detector/entities.py	[new file with mode: 0644]	blob
detector/detector/gliner_layer.py	[new file with mode: 0644]	blob
detector/detector/merge.py	[new file with mode: 0644]	blob
detector/pyproject.toml	[new file with mode: 0644]	blob
detector/tests/test_analyze.py	[new file with mode: 0644]	blob
detector/tests/test_deterministic.py	[new file with mode: 0644]	blob
detector/tests/test_gliner_layer.py	[new file with mode: 0644]	blob
detector/tests/test_merge.py	[new file with mode: 0644]	blob
docker-compose.yml		diff \| blob \| history
docker/Dockerfile		diff \| blob \| history
docker/presidio/generate-configs.py	[deleted file]	blob \| history
docker/presidio/languages.yaml	[deleted file]	blob \| history
docker/supervisord.conf		diff \| blob \| history
docs/api-reference/mask.mdx		diff \| blob \| history
docs/api-reference/status.mdx		diff \| blob \| history
docs/concepts/pii-detection.mdx		diff \| blob \| history
docs/configuration/overview.mdx		diff \| blob \| history
docs/configuration/pii-detection.mdx		diff \| blob \| history
docs/installation.mdx		diff \| blob \| history
docs/introduction.mdx		diff \| blob \| history
docs/quickstart.mdx		diff \| blob \| history
docs/use-cases/chat.mdx		diff \| blob \| history
docs/use-cases/coding-tools.mdx		diff \| blob \| history
package.json		diff \| blob \| history
src/config.test.ts		diff \| blob \| history
src/config.ts		diff \| blob \| history
src/constants/languages.ts		diff \| blob \| history
src/index.ts		diff \| blob \| history
src/pii/detect.test.ts		diff \| blob \| history
src/pii/detect.ts		diff \| blob \| history
src/routes/api.test.ts		diff \| blob \| history
src/routes/api.ts		diff \| blob \| history
src/routes/codex.test.ts		diff \| blob \| history
src/routes/health.test.ts		diff \| blob \| history
src/routes/health.ts		diff \| blob \| history
src/routes/info.ts		diff \| blob \| history
src/services/language-detector.ts		diff \| blob \| history
src/services/pii.ts		diff \| blob \| history

RSS Atom

git clone https://git.99rst.org/PROJECT