https://github.com/takumiecd/kairo-ai
· scanned 2026-06-16 00:29 UTC (2 months, 2 weeks ago)
33 raw signals (0 security + 33 graph)
Last scanned 2 months, 2 weeks ago · v1 · 20 actionable findings from 1 signal source. 13 repeated signals grouped for readability. Security checks, system graph analysis, and verified AI-agent feedback are merged into one review queue.
All 541 nodes from the latest scan, grouped by kind. Each node is a unit the engine identified (file, function, endpoint, table…). Most users won't need this view — it's primarily for debugging the engine's graph extraction or for AI agents that want to enumerate the project structure.
| Label | Layer | Status | Path |
|---|---|---|---|
test_edit_distance_exact_match |
software | healthy | tests/test_eval_metrics.py:10 |
test_edit_distance_substitution |
software | healthy | tests/test_eval_metrics.py:13 |
test_edit_distance_insert_delete |
software | healthy | tests/test_eval_metrics.py:16 |
test_cer |
software | healthy | tests/test_eval_metrics.py:19 |
test_cer_result_keeps_counts |
software | healthy | tests/test_eval_metrics.py:22 |
test_mean_cer_uses_corpus_level_denominator |
software | healthy | tests/test_eval_metrics.py:29 |
test_mean_cer_rejects_mismatched_lengths |
software | healthy | tests/test_eval_metrics.py:35 |
test_clean_wikipedia_text_removes_markup |
software | healthy | tests/test_dataset_source_wikipedia.py:36 |
test_iter_wikipedia_pages_skips_redirect_and_non_article |
software | healthy | tests/test_dataset_source_wikipedia.py:48 |
test_extract_wikipedia_units_reads_bz2_dump |
software | healthy | tests/test_dataset_source_wikipedia.py:58 |
__init__ |
software | healthy | tests/test_decode_beam.py:10 |
__call__ |
software | healthy | tests/test_decode_beam.py:15 |
test_beam_search_returns_confidence_ranked_candidates |
software | healthy | tests/test_decode_beam.py:29 |
test_preserves_ascii_command_fragments |
software | healthy | tests/test_dataset_generation.py:8 |
test_mixed_identifier_and_japanese_text |
software | healthy | tests/test_dataset_generation.py:18 |
test_generates_clean_and_noisy_examples |
software | healthy | tests/test_dataset_generation.py:27 |
test_noise_preserves_literal_code_spans |
software | healthy | tests/test_dataset_generation.py:37 |
test_noise_can_touch_mutable_english_spans |
software | healthy | tests/test_dataset_generation.py:49 |
test_can_disable_literal_noise |
software | healthy | tests/test_dataset_generation.py:59 |
test_splits_mixed_text |
software | healthy | tests/test_dataset_generation.py:69 |
__call__ |
software | healthy | tests/test_train_validation.py:12 |
eval |
software | healthy | tests/test_train_validation.py:18 |
train |
software | healthy | tests/test_train_validation.py:21 |
test_unwrap_plain_dataset |
software | healthy | tests/test_train_validation.py:26 |
test_evaluate_decode_cer_none |
software | healthy | tests/test_train_validation.py:34 |
test_evaluate_decode_cer_greedy |
software | healthy | tests/test_train_validation.py:48 |
test_clean_aozora_text_removes_ruby_and_notes |
software | healthy | tests/test_dataset_source_text.py:13 |
test_extract_text_units_splits_sentences |
software | healthy | tests/test_dataset_source_text.py:30 |
test_read_source_text_reads_zip_txt |
software | healthy | tests/test_dataset_source_text.py:39 |
test_augment_includes_clean_input |
software | healthy | tests/test_noise.py:7 |
test_romaji_variant_changes_known_pattern |
software | healthy | tests/test_noise.py:16 |
test_selects_cpu_device |
software | healthy | tests/test_train_train.py:25 |
test_split_dataset_creates_validation_subset |
software | healthy | tests/test_train_train.py:28 |
test_split_dataset_rejects_invalid_ratio |
software | healthy | tests/test_train_train.py:36 |
test_load_best_valid_loss_defaults_to_inf |
software | healthy | tests/test_train_train.py:40 |
test_checkpoint_round_trip_has_epoch |
software | healthy | tests/test_train_train.py:44 |
test_resume_model_dims_prefers_checkpoint_config |
software | healthy | tests/test_train_train.py:69 |
test_resume_model_dims_reads_split_checkpoint_config |
software | healthy | tests/test_train_train.py:86 |
test_resume_model_dims_can_infer_from_state_dict |
software | healthy | tests/test_train_train.py:109 |
test_rnnt_batch_sampler_sorts_by_lattice_size |
software | healthy | tests/test_train_train.py:131 |
test_rnnt_bucket_sampler_changes_order_between_epochs |
software | healthy | tests/test_train_train.py:143 |
test_rnnt_batch_sampler_limits_padded_lattice_cells |
software | healthy | tests/test_train_train.py:163 |
test_rnnt_bucket_sampler_len_matches_actual_batches_with_dy… |
software | healthy | tests/test_train_train.py:187 |
test_compute_edit_loss_returns_trainable_loss |
software | healthy | tests/test_edit_loss.py:11 |
test_input_vocab_round_trip |
software | healthy | tests/test_vocab.py:9 |
test_output_vocab_has_rnnt_special_tokens |
software | healthy | tests/test_vocab.py:18 |
test_output_bpe_vocab_encodes_repeated_subwords |
software | healthy | tests/test_vocab.py:25 |
test_forward_returns_operation_and_insert_logits |
software | healthy | tests/test_model_edit_transducer.py:10 |
test_forward_accepts_empty_previous_hypothesis |
software | healthy | tests/test_model_edit_transducer.py:35 |
test_normalizes_candidate_scores |
software | healthy | tests/test_decode_scores.py:11 |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
README.md |
software | healthy | README.md |
CLAUDE.md |
software | healthy | CLAUDE.md |
pyproject.toml |
software | healthy | pyproject.toml |
LICENSE |
software | healthy | LICENSE |
AGENTS.md |
software | healthy | AGENTS.md |
test_eval_metrics.py |
software | healthy | tests/test_eval_metrics.py |
test_dataset_source_wikipedia.py |
software | healthy | tests/test_dataset_source_wikipedia.py |
test_decode_beam.py |
software | healthy | tests/test_decode_beam.py |
test_dataset_generation.py |
software | healthy | tests/test_dataset_generation.py |
test_train_validation.py |
software | healthy | tests/test_train_validation.py |
test_dataset_source_text.py |
software | healthy | tests/test_dataset_source_text.py |
test_noise.py |
software | healthy | tests/test_noise.py |
test_train_train.py |
software | healthy | tests/test_train_train.py |
test_edit_loss.py |
software | healthy | tests/test_edit_loss.py |
test_vocab.py |
software | healthy | tests/test_vocab.py |
test_model_edit_transducer.py |
software | healthy | tests/test_model_edit_transducer.py |
test_decode_scores.py |
software | healthy | tests/test_decode_scores.py |
test_bench_rnnt.py |
software | healthy | tests/test_bench_rnnt.py |
test_decode_edit_beam.py |
software | healthy | tests/test_decode_edit_beam.py |
test_diffusion_data.py |
software | healthy | tests/test_diffusion_data.py |
test_dataset_source_tatoeba.py |
software | healthy | tests/test_dataset_source_tatoeba.py |
test_model_transducer.py |
software | healthy | tests/test_model_transducer.py |
test_dataset_split.py |
software | healthy | tests/test_dataset_split.py |
test_decode_greedy.py |
software | healthy | tests/test_decode_greedy.py |
test_model_diffusion.py |
software | healthy | tests/test_model_diffusion.py |
test_edit_data.py |
software | healthy | tests/test_edit_data.py |
test_train_data.py |
software | healthy | tests/test_train_data.py |
test_edit_validation.py |
software | healthy | tests/test_edit_validation.py |
test_eval_run_test.py |
software | healthy | tests/test_eval_run_test.py |
test_dataset_source_manifest.py |
software | healthy | tests/test_dataset_source_manifest.py |
test_train_overfit.py |
software | healthy | tests/test_train_overfit.py |
test_decode_diffusion.py |
software | healthy | tests/test_decode_diffusion.py |
DATASET.md |
software | healthy | docs/DATASET.md |
AOZORA_SOURCES.example.json |
software | healthy | docs/AOZORA_SOURCES.example.json |
DATA_POLICY.md |
software | healthy | docs/DATA_POLICY.md |
ROADMAP.md |
software | healthy | docs/ROADMAP.md |
MODEL_DESIGN.md |
software | healthy | docs/MODEL_DESIGN.md |
diffusion.py |
software | healthy | decode/diffusion.py |
__init__.py |
software | healthy | decode/__init__.py |
greedy.py |
software | healthy | decode/greedy.py |
beam.py |
software | healthy | decode/beam.py |
edit_beam.py |
software | healthy | decode/edit_beam.py |
scores.py |
software | healthy | decode/scores.py |
__init__.py |
software | healthy | bench/__init__.py |
rnnt.py |
software | healthy | bench/rnnt.py |
metrics.py |
software | healthy | eval/metrics.py |
__init__.py |
software | healthy | eval/__init__.py |
run_test.py |
software | healthy | eval/run_test.py |
__init__.py |
software | healthy | train/__init__.py |
train.py |
software | healthy | train/refiner/train.py |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
EvalMetricsTest |
software | healthy | tests/test_eval_metrics.py:9 |
DatasetSourceWikipediaTest |
software | healthy | tests/test_dataset_source_wikipedia.py:35 |
DummyBeamModel |
software | healthy | tests/test_decode_beam.py:9 |
DecodeBeamTest |
software | healthy | tests/test_decode_beam.py:28 |
DatasetGenerationTest |
software | healthy | tests/test_dataset_generation.py:7 |
EmptyDecodeModel |
software | healthy | tests/test_train_validation.py:11 |
TrainValidationTest |
software | healthy | tests/test_train_validation.py:25 |
DatasetSourceTextTest |
software | healthy | tests/test_dataset_source_text.py:12 |
NoiseTest |
software | healthy | tests/test_noise.py:6 |
DummyRnntExample |
software | healthy | tests/test_train_train.py:19 |
TrainEntrypointTest |
software | healthy | tests/test_train_train.py:24 |
Config |
software | healthy | tests/test_train_train.py:46 |
EditLossTest |
software | healthy | tests/test_edit_loss.py:10 |
VocabTest |
software | healthy | tests/test_vocab.py:8 |
KairoEditTransducerTest |
software | healthy | tests/test_model_edit_transducer.py:9 |
DecodeScoresTest |
software | healthy | tests/test_decode_scores.py:10 |
DummyExample |
software | healthy | tests/test_bench_rnnt.py:16 |
RnntBenchmarkTest |
software | healthy | tests/test_bench_rnnt.py:20 |
DummyEditModel |
software | healthy | tests/test_decode_edit_beam.py:15 |
DecodeEditBeamTest |
software | healthy | tests/test_decode_edit_beam.py:44 |
DiffusionDataTest |
software | healthy | tests/test_diffusion_data.py:11 |
DatasetSourceTatoebaTest |
software | healthy | tests/test_dataset_source_tatoeba.py:11 |
KairoTransducerTest |
software | healthy | tests/test_model_transducer.py:6 |
DatasetSplitTest |
software | healthy | tests/test_dataset_split.py:11 |
DummyGreedyModel |
software | healthy | tests/test_decode_greedy.py:15 |
DecodeGreedyTest |
software | healthy | tests/test_decode_greedy.py:32 |
DiffusionModelTest |
software | healthy | tests/test_model_diffusion.py:11 |
EditDataTest |
software | healthy | tests/test_edit_data.py:15 |
TrainDataTest |
software | healthy | tests/test_train_data.py:11 |
DummyKeepModel |
software | healthy | tests/test_edit_validation.py:13 |
EditValidationTest |
software | healthy | tests/test_edit_validation.py:29 |
EvalRunTestTest |
software | healthy | tests/test_eval_run_test.py:9 |
DatasetSourceManifestTest |
software | healthy | tests/test_dataset_source_manifest.py:10 |
TrainOverfitTest |
software | healthy | tests/test_train_overfit.py:14 |
DiffusionDecodeTest |
software | healthy | tests/test_decode_diffusion.py:10 |
BeamState |
software | healthy | decode/beam.py:19 |
EditBeamState |
software | healthy | decode/edit_beam.py:21 |
Candidate |
software | healthy | decode/scores.py:10 |
TimingSummary |
software | healthy | bench/rnnt.py:36 |
TrainBenchmark |
software | healthy | bench/rnnt.py:47 |
DecodeBenchmark |
software | healthy | bench/rnnt.py:61 |
ValueSummary |
software | healthy | bench/rnnt.py:72 |
DatasetSummary |
software | healthy | bench/rnnt.py:83 |
CerResult |
software | healthy | eval/metrics.py:9 |
RefineTrainConfig |
software | healthy | train/refiner/train.py:35 |
EncodedRefineExample |
software | healthy | train/refiner/data.py:44 |
JsonlRefineDataset |
software | healthy | train/refiner/data.py:56 |
Trainer |
software | healthy | train/common/engine.py:258 |
TrainingVocabs |
software | healthy | train/common/data.py:16 |
TrainConfig |
software | healthy | train/rnnt/train.py:36 |
Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.
| Label | Layer | Status | Path |
|---|---|---|---|
tests |
software | healthy | tests |
docs |
software | healthy | docs |
decode |
software | healthy | decode |
bench |
software | healthy | bench |
eval |
software | healthy | eval |
train |
software | healthy | train |
refiner |
software | healthy | train/refiner |
common |
software | healthy | train/common |
rnnt |
software | healthy | train/rnnt |
edit |
software | healthy | train/edit |
diffusion |
software | healthy | train/diffusion |
model |
software | healthy | model |
dataset |
software | healthy | dataset |
| Label | Layer | Status | Path |
|---|---|---|---|
repobility-clone-qpfy4px3 |
software | healthy | /tmp/repobility-clone-qpfy4px3 |
| Label | Layer | Status | Path |
|---|---|---|---|
gpu (detected) |
hardware | healthy | README.md |
This page is publicly accessible at:
https://repobility.com/scan/0a080025-e2e4-43db-bd84-588d7b7a2d48/
To check status programmatically (no auth required):
curl -s https://repobility.com/api/v1/public/scan/0a080025-e2e4-43db-bd84-588d7b7a2d48/Important — please don't re-submit the same URL repeatedly. The submission endpoint is idempotent: re-submitting the same git URL returns this same scan_token, not a new one. To re-scan this repo, sign up free and use the dashboard.