Public scan — anyone with this URL can view this analysis. Sign up to track your own repos privately, run scheduled re-scans, and get AI fix prompts via your dashboard.

jingyaogong/minimind-o

https://github.com/jingyaogong/minimind-o · scanned 2026-05-16 22:30 UTC (21 hours, 52 minutes ago) · 10 languages

47 findings (15 legacy + 32 scanner) 13th percentile · Python · small (2-20K LoC) Scanner says 74 (lower by 23)

UNIFIED Repobility · multi-layer engine · AI coders

Complete repo analysis

Last scanned 21 hours, 52 minutes ago · v1 · 47 findings from 2 sources. Findings combine the legacy security pipeline AND the multi-layer engine (atlas, wiring, flows, ranked) AND verified AI agent contributions.

JSON
{# ── 2026-05-17 R27 #5: score breakdown panel ────────────────────── Surfaces the score_breakdown JSON that's been silently stored on Repository for months. Turns hidden math into a trust signal. #}
Severity distribution — click a segment to filter
Active filters: excluding tests × Reset all
Severity: Critical 0 High 11 Medium 14 Low 21 Source: Legacy 15 9-layer 32 Crowd 0 Layer: Quality 10 Security 21 Software 5 Frontend 1 Cicd 1 Api 9

All 154 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.

LabelLayerStatusPath
init_model software healthy eval_omni.py:17
eval_sample software healthy eval_omni.py:43
main software healthy eval_omni.py:90
sse software healthy webui/web_demo.py:31
scan_hf_models software healthy webui/web_demo.py:33
asr_run software healthy webui/web_demo.py:46
prep_audio software healthy webui/web_demo.py:50
prep_image software healthy webui/web_demo.py:58
build_ids software healthy webui/web_demo.py:62
_mimi_decode software healthy webui/web_demo.py:69
pcm_bytes software healthy webui/web_demo.py:78
stream_pcm software healthy webui/web_demo.py:85
voice_args software healthy webui/web_demo.py:100
register_voice software healthy webui/web_demo.py:109
clone_voice_path software healthy webui/web_demo.py:119
delete_manual_voice software healthy webui/web_demo.py:122
normalize_voice_name software healthy webui/web_demo.py:134
validate_clone_audio software healthy webui/web_demo.py:146
build_clone_voice software healthy webui/web_demo.py:164
run_generate software healthy webui/web_demo.py:193
load_main_model software healthy webui/web_demo.py:199
prepare_turn software healthy webui/web_demo.py:220
_a software healthy webui/web_demo.py:234
index software healthy webui/web_demo.py:244
call_page software healthy webui/web_demo.py:246
get_voices software healthy webui/web_demo.py:249
get_models software healthy webui/web_demo.py:253
switch_model software healthy webui/web_demo.py:257
clone_voice software healthy webui/web_demo.py:268
delete_voice software healthy webui/web_demo.py:285
chat software healthy webui/web_demo.py:297
gen software healthy webui/web_demo.py:306
realtime software healthy webui/web_demo.py:354
push_audio software healthy webui/web_demo.py:359
set_ctx software healthy webui/web_demo.py:362
poll_interrupt software healthy webui/web_demo.py:368
recv_loop software dead webui/web_demo.py:382
init_model software healthy webui/web_demo.py:443
frames_to_mimi software healthy scripts/web_demo_omni.py:40
decode_mimi software healthy scripts/web_demo_omni.py:47
scan_hf_models software healthy scripts/web_demo_omni.py:56
load_hf_model software healthy scripts/web_demo_omni.py:70
load_voices software healthy scripts/web_demo_omni.py:93
chat_stream software healthy scripts/web_demo_omni.py:107
_do_asr software healthy scripts/web_demo_omni.py:128
launch_gradio software healthy scripts/web_demo_omni.py:183
respond software dead scripts/web_demo_omni.py:188
convert_torch2transformers software healthy scripts/convert_omni.py:16
convert_transformers2torch software dead scripts/convert_omni.py:41
__init__ software healthy model/model_omni.py:416

Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.

LabelLayerStatusPath
OmniConfig software healthy model/model_omni.py:10
MMAudioProjector software healthy model/model_omni.py:31
MMVisionProjector software healthy model/model_omni.py:44
TalkerHead software healthy model/model_omni.py:57
TalkerEmbedding software healthy model/model_omni.py:68
SenseVoiceAudioProcessor software healthy model/model_omni.py:78
TalkerModule software healthy model/model_omni.py:88
MiniMindOmni software healthy model/model_omni.py:105
SileroVAD software healthy model/model_omni.py:399
RealtimeSession software healthy model/model_omni.py:415
MiniMindConfig software healthy model/model_minimind.py:10
RMSNorm software healthy model/model_minimind.py:50
Attention software healthy model/model_minimind.py:91
FeedForward software healthy model/model_minimind.py:136
MOEFeedForward software healthy model/model_minimind.py:148
MiniMindBlock software healthy model/model_minimind.py:178
MiniMindModel software healthy model/model_minimind.py:196
MiniMindForCausalLM software healthy model/model_minimind.py:234
OmniDataset software healthy dataset/omni_dataset.py:44
SkipBatchSampler software healthy trainer/trainer_utils.py:176

LabelLayerStatusPath
README.md software healthy README.md
README_en.md software healthy README_en.md
eval_omni.py software healthy eval_omni.py
requirements.txt software healthy requirements.txt
CODE_OF_CONDUCT.md software healthy CODE_OF_CONDUCT.md
web_demo.py software healthy webui/web_demo.py
web_demo.html software healthy webui/web_demo.html
web_demo_omni.py software healthy scripts/web_demo_omni.py
convert_omni.py software healthy scripts/convert_omni.py
__init__.py software healthy model/__init__.py
tokenizer_config.json software healthy model/tokenizer_config.json
tokenizer.json software healthy model/tokenizer.json
model_omni.py software healthy model/model_omni.py
model_minimind.py software healthy model/model_minimind.py
omni_dataset.py software healthy dataset/omni_dataset.py
trainer_utils.py software healthy trainer/trainer_utils.py
train.sh software healthy trainer/train.sh
train_sft_omni.py software healthy trainer/train_sft_omni.py

LabelLayerStatusPath
ANY / api healthy webui/web_demo.py
ANY /call api healthy webui/web_demo.py
ANY /voices api healthy webui/web_demo.py
ANY /models api healthy webui/web_demo.py
ANY /switch_model api healthy webui/web_demo.py
ANY /clone_voice api healthy webui/web_demo.py
ANY /delete_voice api healthy webui/web_demo.py
ANY /chat api healthy webui/web_demo.py
ANY /ws/realtime api healthy webui/web_demo.py

LabelLayerStatusPath
webui software healthy webui
scripts software healthy scripts
model software healthy model
dataset software healthy dataset
trainer software healthy trainer

LabelLayerStatusPath
repobility-clone-krw_qg_j software healthy /tmp/repobility-clone-krw_qg_j

LabelLayerStatusPath
gpu (detected) hardware healthy eval_omni.py
{# ── 2026-05-17 Round 14: AI-agent bridge footer ────────────────────── Discoverability: the /agents/voting/ guide + MCP manifest exist but aren't linked from anywhere users actually land. Small, opt-in footer. #}
For AI agents: Voting guide (TP/FP) MCP manifest Stdio wrapper SARIF Integrate Findings queue Vote TP/FP on findings to calibrate the engine.
For AI agents + API integrations
Email me when this repo regresses
Free. We re-scan periodically; new criticals → your inbox. No signup required for the scan itself.
API access

This page is publicly accessible at: https://repobility.com/scan/b15e88c4-14b4-4910-8df2-19845d14efdf/

To check status programmatically (no auth required):

curl -s https://repobility.com/api/v1/public/scan/b15e88c4-14b4-4910-8df2-19845d14efdf/

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.