QIN2DIM/hcaptcha-challenger v0.18.7

on GitHub

What's Changed

• build 🚀: RAG by @beiyuouo in #6
• sky left airplane challenger by @beiyuouo in #8
• feat ✨: pypi release by @beiyuouo in #24
• feat(add): elephant solution by @beiyuouo in #37
• feat(add): YOLOWithAugmentation by @QIN2DIM in #45
• chore: fix multi-instances error by @vgabs in #73
• chore(requirements): added scipy by @Vinyzu in #119
• Updated Lion with closed eyes by @Vinyzu in #139
• v0.4.0 by @QIN2DIM in #215
• 0.4.1 by @QIN2DIM in #217
• feat(add): split prompt message by @QIN2DIM in #220
• v0.4.2 by @QIN2DIM in #221
• chore: merge pypi repo #299 by @QIN2DIM in #300
• update: parser rules by @QIN2DIM in #322
• feat(add): parameter slowdown by @QIN2DIM in #399
• Linux: allow headful operation if X11 present by @randrej in #467
• fix: #233 by @mohdrasbi in #502
• Catch ResourceWarning by @QIN2DIM in #516
• Update(requires) sync version by @QIN2DIM in #574
• v0.5.0 by @QIN2DIM in #575
• chore(examples): remove main.py by @QIN2DIM in #578
• update(skeleton): new package structure by @QIN2DIM in #579
• build(poetry): add pyproject.toml by @QIN2DIM in #580
• Create .gitattributes by @QIN2DIM in #581
• Update .gitattributes by @QIN2DIM in #582
• Update pyproject.toml by @QIN2DIM in #583
• feat(playwright-agent): build radagon agent by @QIN2DIM in #584
• Update(agent-status): as property by @QIN2DIM in #585
• feat(prompt-handler): diagnose_task by @QIN2DIM in #586
• feat(onnx): add yolov8 motion-flow by @QIN2DIM in #590
• feat(auto-challenge): merge binary and area_select:point by @QIN2DIM in #591
• feat(agent): export rqdata from response event by @QIN2DIM in #597
• Pattern rules by @QIN2DIM in #599
• Feat(benchmarks): add memory profile by @QIN2DIM in #602
• mode fix by @supahero1 in #603
• Feat(agent): add Coroutine MQ by @QIN2DIM in #604
• Feat(components): area selector by @QIN2DIM in #605
• Update(ONNX): finetune keypoint digit9 by @QIN2DIM in #607
• Update(example): mini dataset collector by @QIN2DIM in #609
• Feat(objects): merge yolo names to objects.yaml by @QIN2DIM in #610
• Feat(agents): solution of bounding_box challenge by @QIN2DIM in #612
• Fixed(parse_label): challenge passed by @QIN2DIM in #614
• Feat(ONNX): handle challenge head of animal by @QIN2DIM in #616
• docs(benchmarks): model slot by @QIN2DIM in #618
• Update(demo): startup playwright with chromium browser by @QIN2DIM in #620
• Feat(ONNX): model slot by @QIN2DIM in #619
• Sentinel by @QIN2DIM in #627
• ff by @QIN2DIM in #636
• chore(sentinel): update progress log by @QIN2DIM in #639
• remove selenium agent by @QIN2DIM in #642
• feat(automation): collector with pigeon by @QIN2DIM in #650
• ci(istock)： build by @QIN2DIM in #684
• chore: workflows, automation, pyproject.toml by @QIN2DIM in #691
• Feat(downloader): coroutine in main thead by @QIN2DIM in #697
• Async dm by @QIN2DIM in #706
• Update pyproject.toml by @QIN2DIM in #708
• Fix(Radagon): tmp_dir by @QIN2DIM in #719
• Feat(Radagon): catch all keypoint challenge by @QIN2DIM in #721
• Feat(modelhub): new rules by @QIN2DIM in #727
• Update(automation): move to factory by @QIN2DIM in #743
• Fix(requirements): detach playwright by @QIN2DIM in #745
• Feat(components): LBC by @QIN2DIM in #758
• Feat(cv-component): 761 OpenCV toolkit by @QIN2DIM in #764
• Fix auto challenge success not returning rqdata by @tarekwiz in #769
• Fix spectral clustering by @QIN2DIM in #775
• Perf(control): acc of detection by @QIN2DIM in #776
• Feat(utils): Monitoring download progress by @QIN2DIM in #789
• Update(utils): install --flush-yolo by @QIN2DIM in #792
• Feat(control): unsupervised methods by @QIN2DIM in #809
• Feat(ONNX): ResNet Rank by @QIN2DIM in #827
• example: yolo mocker by @QIN2DIM in #830
• patch(onnx): rank model by @QIN2DIM in #831
• feat(onnx): YOLOv8Seg by @QIN2DIM in #834
• Feat(components): nested classifier by @QIN2DIM in #836
• perf(components): find unique object by @QIN2DIM in #839
• perf(control): improve performance of trident by @QIN2DIM in #841
• Feat(app): FastAPI localhost by @QIN2DIM in #798
• docs: appears_only_once by @QIN2DIM in #845
• refactor: examples, assets by @QIN2DIM in #846
• feat(install): project self upgrade by @QIN2DIM in #850
• feat(collector): ignore examples by @QIN2DIM in #856
• feat(onnx): ViT zero-shot image classification by @QIN2DIM in #863
• feat(control): unsupervised binary challenge by @QIN2DIM in #865
• fix(control): CLIP self-supervised tasks by @QIN2DIM in #866
• perf(self-supervised): remove unnecessary deceleration by @QIN2DIM in #867
• docs(flowchat): datalake-post by @QIN2DIM in #870
• fix(install): detach clip by @QIN2DIM in #871
• perf(self-supervised) : switch to RN50.openai.onnx by @QIN2DIM in #874
• feat(components): add prompt handler by @QIN2DIM in #875
• update(examples): star bricks, auto-labeling by @QIN2DIM in #876
• feat(components): self-supervised classifier by @QIN2DIM in #877
• fix(control): detach segment and detection task by @QIN2DIM in #878
• update(tests): handle prompts by @QIN2DIM in #879
• fix(classifier): self-supervised mode by @QIN2DIM in #882
• update(tests): prompt-handler by @QIN2DIM in #890
• 900 perfself supersived dramatically improve clips correctness in unprompted scenarios by @QIN2DIM in #901
• fix(rank-models): Decoupling + Efficiency by @QIN2DIM in #904
• feat(prompt-handler): add rules by @QIN2DIM in #918
• feat(control): image_label_multiple_choice by @QIN2DIM in #920
• remove(tests): downloader, normal playwright by @QIN2DIM in #922
• fix(control): challenge listener by @QIN2DIM in #924
• feat(pipline-control): new agent by @QIN2DIM in #948
• feat(tenacity): Reinforced skeleton network by @QIN2DIM in #950
• feat(agent): challenge pipline by @QIN2DIM in #954
• feat(agent): catch-all binary challenge by @QIN2DIM in #955
• update(AgentT): from_page classmethod by @QIN2DIM in #956
• fix(onnx): missing fields by @QIN2DIM in #957
• fix small CLIP issue matching correct label by @tarekwiz in #966
• InvokeTool: multimodal large language model by @QIN2DIM in #980
• AIOps adjust the project organizational structure by @QIN2DIM in #988
• refactor by @QIN2DIM in #989
• Archive obsolete files by @QIN2DIM in #990
• Adjust project structure by @QIN2DIM in #991
• Refactor image classifier by @QIN2DIM in #992
• update AgentConfig by @QIN2DIM in #993
• AgentStategy: handle challenge image_label_area_select by @QIN2DIM in #994
• AgentStrategy: create SpatialPathReasoner for challenge by @QIN2DIM in #997
• add tests by @QIN2DIM in #999
• update tests by @QIN2DIM in #1000
• Msg package by @QIN2DIM in #1001
• decoupled dependencies by @QIN2DIM in #1002
• tenacity retry by @QIN2DIM in #1003
• update gemini-2.5-pro-exp-03-25 by @QIN2DIM in #1004
• update MODEL LIST by @QIN2DIM in #1005
• update env-generator by @QIN2DIM in #1006
• Create FunCaptcha experiment by @QIN2DIM in #1007
• add captcha tasklist model by @QIN2DIM in #1008
• fix: disable response_schema by @QIN2DIM in #1009
• fix: response_schema by @QIN2DIM in #1012
• feat: dataset management tool by @QIN2DIM in #1013
• fix(cli): dataset check by @QIN2DIM in #1015
• feat: handle nested frames by @QIN2DIM in #1019
• feat(config): ignore request questions by @QIN2DIM in #1020
• add: gemini-2.5-flash-preview-04-17 by @QIN2DIM in #1021
• replace bad code by @QIN2DIM in #1022
• feat: save model answer by @QIN2DIM in #1023
• Cost calculator by @QIN2DIM in #1024
• Support image_drag_multi by @QIN2DIM in #1026
• feat: compatible with rebrowser by @QIN2DIM in #1028
• feat(challenge): support camoufox by @QIN2DIM in #1029
• fix(challenge): HSW reverse fallback by @QIN2DIM in #1030
• Use Gemini async api to avoid blocking event loop by @cheewba in #1031
• Gemini 2.5 pro preview 05 06 by @QIN2DIM in #1033
• fix(image_classifier): coordinate hallucination by @QIN2DIM in #1035
• feat(BoundingBoxCoordinate): Coordinate system mapping by @QIN2DIM in #1037
• interfaces: update IMAGE_DRAG_SINGLE_MOST_SIMILAR by @QIN2DIM in #1038
• Generic M by @QIN2DIM in #1039
• fix: prompt-inv by @QIN2DIM in #1047
• feat(prompt): visual guidance SAME_NUMBER_OF_HOLES by @QIN2DIM in #1048
• feat(tools): ChallengeRouter fallback by @QIN2DIM in #1049
• update(models): set default values by @QIN2DIM in #1051
• ci-builder by @QIN2DIM in #1052
• Update prompts.py by @QIN2DIM in #1054

New Contributors

• @beiyuouo made their first contribution in #6
• @QIN2DIM made their first contribution in #45
• @vgabs made their first contribution in #73
• @Vinyzu made their first contribution in #119
• @randrej made their first contribution in #467
• @mohdrasbi made their first contribution in #502
• @supahero1 made their first contribution in #603
• @tarekwiz made their first contribution in #769
• @cheewba made their first contribution in #1031

Full Changelog: model...v0.18.7

trigger1