Guillaume Meyer (The Opinionated Man)andGitHub 1cc278342e feat: multi-scheme MarkLLM text benchmark and detection (#188) ...

- bench_synthid_text.py: --scheme/--config to run any MarkLLM scheme;
  default stays synthid (backward compatible)
- detect_text_watermark.py: exp/unigram/sir schemes; --temperature/--top-p
  including per-request overrides in the serve worker
- rewrite_text.py: --markllm-scheme accepts exp/unigram/sir for the
  adaptive rewrite loop
- tests: scheme-surface coverage + bench fixture updates

2026-08-19 08:08:44 -07:00

..

fixtures

feat: add native stdlib XLSX and PPTX container metadata, text, and embedded media scrubbing (#91) (#100)

2026-08-16 17:43:09 -07:00

test_ai_generator_hints.py

fix: detect AI markers in compressed PNG text (#127)

2026-08-18 11:56:48 -07:00

test_audit_sarif_and_concurrency.py

fix: correctness and security hardening of the cleaning scripts (#122) (#126)

2026-08-17 17:15:35 -07:00

test_audit.py

test: refuse DTD/entity bombs in sitemap parser (GHSA-pjg6-92pm-mmcf) (#146)

2026-08-18 09:24:54 -07:00

test_av_meta.py

feat: audio/video AI/C2PA metadata stripping (MP4/MOV, WAV, MP3) (#139)

2026-08-18 07:47:43 -07:00

test_avif_heic.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_bench_synthid_text.py

feat: multi-scheme MarkLLM text benchmark and detection (#188)

2026-08-19 08:08:44 -07:00

test_binary_guard.py

fix: correctness and security hardening of the cleaning scripts (#122) (#126)

2026-08-17 17:15:35 -07:00

test_c2patool_report.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_clean_image.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_clean_text.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_container_meta.py

fix: correctness and security hardening of the cleaning scripts (#122) (#126)

2026-08-17 17:15:35 -07:00

test_ctrlregen_clean.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_detect_endpoint.py

feat: add reproducible SynthID-text removal benchmark (#145)

2026-08-18 18:13:56 -07:00

test_embedded_data_uris.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_entity_encoded_watermarks.py

fix: decode XML entities before Layer A scrub in OOXML/ODF (#142)

2026-08-18 12:30:35 -07:00

test_epub.py

fix: correctness and security hardening of the cleaning scripts (#122) (#126)

2026-08-17 17:15:35 -07:00

test_format_dispatch.py

fix: correctness and security hardening of the cleaning scripts (#122) (#126)

2026-08-17 17:15:35 -07:00

test_http_server.py

fix(server): preserve image format extensions in /clean and use safe writes in av_meta (#150)

2026-08-18 12:37:26 -07:00

test_image_formats_bmp_gif_tiff.py

feat: add BMP, GIF, TIFF, and EPUB format support (#107)

2026-08-16 19:47:11 -07:00

test_inspect_file.py

feat: split skill from service, add HTTP API and Docker distribution (#60)

2026-08-14 15:42:48 -07:00

test_json_exit_code.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_lightweight_skill.py

fix: pin vendored Cursor-skill text engine to the service copy (#96)

2026-08-16 18:44:18 -07:00

test_markdiffusion_harness.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_markllm_detect.py

feat: vendor text-watermark detection (Gemini SynthID, Claude seam, MarkLLM) + SynthID image scorer sidecar (#109)

2026-08-17 18:31:22 -07:00

test_markllm_schemes.py

feat: multi-scheme MarkLLM text benchmark and detection (#188)

2026-08-19 08:08:44 -07:00

test_ooxml_xlsx_pptx.py

feat: add native stdlib XLSX and PPTX container metadata, text, and embedded media scrubbing (#91) (#100)

2026-08-16 17:43:09 -07:00

test_pdf_structural_rewrite.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_precommit_hooks.py

feat: add pre-commit hook integration for staged-file checking/cleaning (#138)

2026-08-18 07:26:58 -07:00

test_rewrite_text.py

feat: iterative detection-guided Layer B rewriting (--candidates x --max-loops) (#153)

2026-08-18 18:28:19 -07:00

test_security_hardening.py

fix: linear-time metadata stripping for SVG/ODT (GHSA-7vpp-96qp-j9wh) (#147)

2026-08-18 09:45:56 -07:00

test_single_quoted_attributes.py

fix: parse single-quoted attributes in relationship/manifest pruning (#140)

2026-08-18 12:06:16 -07:00

test_stylometry_empty_sentences.py

fix: report unmeasurable burstiness instead of scoring it as max LLM-like (#141)

2026-08-18 12:14:27 -07:00

test_stylometry.py

fix: report unmeasurable burstiness instead of scoring it as max LLM-like (#141)

2026-08-18 12:14:27 -07:00

test_synthid_score.py

fix: prefer checkout venv for SynthID scorer subprocess (#115) (#123)

2026-08-17 16:34:51 -07:00

test_synthid_stdout_purity.py

chore: add Ruff linting and formatting with CI enforcement (#103)

2026-08-16 17:32:40 -07:00

test_text_detectors.py

feat: add reproducible SynthID-text removal benchmark (#145)

2026-08-18 18:13:56 -07:00