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.

ZIJIAN004/UniCOP

https://github.com/ZIJIAN004/UniCOP · scanned 2026-06-16 01:12 UTC (2 months, 1 week ago)

93 raw signals (0 security + 93 graph)

File as issue Image

UNIFIED Repobility · multi-layer engine · AI coders

Complete repo analysis

Last scanned 2 months, 1 week ago · v1 · 69 actionable findings from 1 signal source. 24 repeated signals grouped for readability. Security checks, system graph analysis, and verified AI-agent feedback are merged into one review queue.

JSON

88.9% cov · 69 findings

Severity distribution — click a segment to filter

Active filters: excluding tests × Reset all

Severity: Critical 0 High 7 Medium 7 Low 31 9-Layer: Software Security Quality Integrity Frontend Hardware Data Network Cicd Source: Security checks 0 System graph 69 Crowd 0 Layer: Software 15 Network 2 Frontend 1 Security 8 Cicd 1 Quality 40 Api 2

Exclude dismissed / FP Exclude test files

All 993 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
`get_version`	software	healthy	`check_versions.py:40`
`get_cuda_info`	software	healthy	`check_versions.py:50`
`main`	software	healthy	`check_versions.py:79`
`_notify`	software	healthy	`optimal/build_optimal.py:50`
`_solve_one`	software	healthy	`optimal/build_optimal.py:59`
`build_one`	software	healthy	`optimal/build_optimal.py:67`
`main`	software	healthy	`optimal/build_optimal.py:133`
`instances_path`	software	healthy	`optimal/generate_testset.py:47`
`generate_frozen`	software	healthy	`optimal/generate_testset.py:52`
`main`	software	healthy	`optimal/generate_testset.py:78`
`_route_cost`	software	healthy	`optimal/solvers.py:42`
`_total_cost`	software	healthy	`optimal/solvers.py:49`
`solve_instance`	software	healthy	`optimal/solvers.py:57`
`_solve_tsp_lkh`	software	healthy	`optimal/solvers.py:118`
`_parse_lkh_tour`	software	healthy	`optimal/solvers.py:157`
`_reroot_at_depot`	software	healthy	`optimal/solvers.py:176`
`_scaled_coords`	software	healthy	`optimal/solvers.py:188`
`_add_edges`	software	healthy	`optimal/solvers.py:193`
`_pyvrp_solve`	software	healthy	`optimal/solvers.py:209`
`_solve_cvrp_pyvrp`	software	healthy	`optimal/solvers.py:222`
`_solve_tsptw_pyvrp`	software	healthy	`optimal/solvers.py:253`
`_solve_vrptw_pyvrp`	software	healthy	`optimal/solvers.py:291`
`_find_by_prefix`	software	healthy	`optimal/loader.py:27`
`load_instances`	software	healthy	`optimal/loader.py:42`
`cache_path`	software	healthy	`optimal/loader.py:63`
`_find_cache`	software	healthy	`optimal/loader.py:68`
`load_costs`	software	dead	`optimal/loader.py:95`
`optimality_gap`	software	dead	`optimal/loader.py:104`
`_stamp`	software	healthy	`Latent-SFT/train.py:54`
`_report_gpu_memory`	software	healthy	`Latent-SFT/train.py:61`
`make_deepspeed_config`	software	healthy	`Latent-SFT/train.py:82`
`train_hlr`	software	healthy	`Latent-SFT/train.py:140`
`_save_hlr_checkpoint`	software	healthy	`Latent-SFT/train.py:504`
`main`	software	healthy	`Latent-SFT/train.py:527`
`_loss_stamp`	software	healthy	`Latent-SFT/model.py:33`
`_t_sync`	software	healthy	`Latent-SFT/model.py:54`
`__init__`	software	healthy	`Latent-SFT/model.py:305`
`__enter__`	software	healthy	`Latent-SFT/model.py:67`
`__exit__`	software	healthy	`Latent-SFT/model.py:73`
`hlr_timing_add`	software	healthy	`Latent-SFT/model.py:80`
`hlr_timing_tick_microstep`	software	healthy	`Latent-SFT/model.py:86`
`hlr_timing_report`	software	healthy	`Latent-SFT/model.py:92`
`forward`	software	dead	`Latent-SFT/model.py:348`
`_rotate_half`	software	healthy	`Latent-SFT/model.py:140`
`apply_rope`	software	healthy	`Latent-SFT/model.py:146`
`project_for_main_layer`	software	healthy	`Latent-SFT/model.py:403`
`build_latent_reasoner_from_main`	software	healthy	`Latent-SFT/model.py:427`
`_get_input_embedding_layer`	software	healthy	`Latent-SFT/model.py:535`
`_segment_ce`	software	healthy	`Latent-SFT/model.py:542`
`compute_hlr_loss`	software	healthy	`Latent-SFT/model.py:564`

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

Label	Layer	Status	Path
`smart_submit.sh`	software	healthy	`smart_submit.sh`
`link_nvidia_libs.sh`	software	healthy	`link_nvidia_libs.sh`
`check_env.py`	software	healthy	`check_env.py`
`README.md`	software	healthy	`README.md`
`check_deepspeed.py`	software	healthy	`check_deepspeed.py`
`paths.sh`	software	healthy	`paths.sh`
`check_versions.py`	software	healthy	`check_versions.py`
`address.md`	software	healthy	`address.md`
`auto_all.sh`	software	healthy	`auto_all.sh`
`setup_env.sh`	software	healthy	`setup_env.sh`
`__init__.py`	software	healthy	`optimal/__init__.py`
`build_optimal.py`	software	healthy	`optimal/build_optimal.py`
`README.md`	software	healthy	`optimal/README.md`
`generate_testset.py`	software	healthy	`optimal/generate_testset.py`
`solvers.py`	software	healthy	`optimal/solvers.py`
`loader.py`	software	healthy	`optimal/loader.py`
`train.py`	software	healthy	`Latent-SFT/train.py`
`model.py`	software	healthy	`Latent-SFT/model.py`
`submit_train_eval_hlr.sh`	software	healthy	`Latent-SFT/submit_train_eval_hlr.sh`
`submit_eval_hlr.sh`	software	healthy	`Latent-SFT/submit_eval_hlr.sh`
`inference.py`	software	healthy	`Latent-SFT/inference.py`
`hlr_config.py`	software	healthy	`Latent-SFT/hlr_config.py`
`submit_smoke_repro.sh`	software	healthy	`Latent-SFT/submit_smoke_repro.sh`
`smoke_test_hlr.py`	software	healthy	`Latent-SFT/smoke_test_hlr.py`
`entropy_profile.py`	software	healthy	`Latent-SFT/entropy_profile.py`
`submit_train_hlr.sh`	software	healthy	`Latent-SFT/submit_train_hlr.sh`
`data_utils.py`	software	healthy	`Latent-SFT/data_utils.py`
`submit_ablate_hlr.sh`	software	healthy	`Latent-SFT/submit_ablate_hlr.sh`
`test_batched_lr.py`	software	healthy	`Latent-SFT/test_batched_lr.py`
`eval_hlr_compare.py`	software	healthy	`Latent-SFT/eval_hlr_compare.py`
`submit_smoke_mem_hlr.sh`	software	healthy	`Latent-SFT/submit_smoke_mem_hlr.sh`
`submit_smoke_hlr.sh`	software	healthy	`Latent-SFT/submit_smoke_hlr.sh`
`rewrite_strategy.py`	software	healthy	`UniCOP-Distill/rewrite_strategy.py`
`run_sft_cvrp20.sh`	software	healthy	`UniCOP-Distill/run_sft_cvrp20.sh`
`run_sft_qwen3_instruct_cvrp20.sh`	software	healthy	`UniCOP-Distill/run_sft_qwen3_instruct_cvrp20.sh`
`check_reverse_compat.py`	software	healthy	`UniCOP-Distill/check_reverse_compat.py`
`submit_sft_template.sh`	software	healthy	`UniCOP-Distill/submit_sft_template.sh`
`rationalize_solutions.py`	software	healthy	`UniCOP-Distill/rationalize_solutions.py`
`vllm_serve_ngram.py`	software	healthy	`UniCOP-Distill/vllm_serve_ngram.py`
`submit_merge_eval_rl.sh`	software	healthy	`UniCOP-Distill/submit_merge_eval_rl.sh`
`annotate_step_reasons.py`	software	healthy	`UniCOP-Distill/annotate_step_reasons.py`
`submit_sft_qwen3_full.sh`	software	healthy	`UniCOP-Distill/submit_sft_qwen3_full.sh`
`run_generate_all.sh`	software	healthy	`UniCOP-Distill/run_generate_all.sh`
`submit_sft_qwen3_test.sh`	software	healthy	`UniCOP-Distill/submit_sft_qwen3_test.sh`
`filter_chains.py`	software	healthy	`UniCOP-Distill/filter_chains.py`
`run_merge_eval_rl.sh`	software	healthy	`UniCOP-Distill/run_merge_eval_rl.sh`
`submit_sft_qwen3_reverse_cvrp20.sh`	software	healthy	`UniCOP-Distill/submit_sft_qwen3_reverse_cvrp20.sh`
`run_sft_hybrid_cvrp20.sh`	software	healthy	`UniCOP-Distill/run_sft_hybrid_cvrp20.sh`
`check_sft_think_masking.py`	software	healthy	`UniCOP-Distill/check_sft_think_masking.py`
`problems_prompt.py`	software	healthy	`UniCOP-Distill/problems_prompt.py`

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

Label	Layer	Status	Path
`hlr_timer`	software	healthy	`Latent-SFT/model.py:59`
`SimpleRMSNorm`	software	healthy	`Latent-SFT/model.py:106`
`RotaryEmbedding`	software	healthy	`Latent-SFT/model.py:124`
`GQAAttention`	software	healthy	`Latent-SFT/model.py:156`
`SwiGLUFFN`	software	healthy	`Latent-SFT/model.py:241`
`LatentReasonerBlock`	software	healthy	`Latent-SFT/model.py:254`
`LatentReasoner`	software	healthy	`Latent-SFT/model.py:283`
`_SampleState`	software	healthy	`Latent-SFT/inference.py:50`
`HLRInferenceEngine`	software	healthy	`Latent-SFT/inference.py:77`
`HLRConfig`	software	healthy	`Latent-SFT/hlr_config.py:15`
`HLRSegment`	software	healthy	`Latent-SFT/data_utils.py:30`
`HLRSample`	software	healthy	`Latent-SFT/data_utils.py:41`
`HLRDataset`	software	healthy	`Latent-SFT/data_utils.py:48`
`_NoRepeatNgramV0`	software	healthy	`UniCOP-Distill/vllm_serve_ngram.py:36`
`FOARLGRPOTrainer`	software	healthy	`FOARL/train_grpo_foarl.py:46`
`GRPOPRMTrainer`	software	healthy	`UniCOP-Reason-Mask/grpo_prm_trainer.py:163`
`Config`	software	healthy	`UniCOP-Reason-Mask/config.py:6`
`WaveConfig`	software	healthy	`UniCOP-Reason-Mask/wave_replay.py:62`
`TrajProfile`	software	healthy	`UniCOP-Reason-Mask/wave_replay.py:70`
`WaveResult`	software	healthy	`UniCOP-Reason-Mask/wave_replay.py:84`
`StepRewards`	software	healthy	`UniCOP-Reason-Mask/pomo_prm.py:60`
`ThinkStep`	software	healthy	`UniCOP-Reason-Mask/pomo_prm.py:81`
`ThinkPRMResult`	software	healthy	`UniCOP-Reason-Mask/pomo_prm.py:89`
`POMOPRM`	software	healthy	`UniCOP-Reason-Mask/pomo_prm.py:225`
`_AlignTest`	software	healthy	`UniCOP-Reason-Mask/pomo_prm.py:1106`
`NoRepeatNGramLogitsProcessor`	software	healthy	`UniCOP-Reason-Mask/evaluate.py:803`
`CVRPMaskProcessor`	software	healthy	`UniCOP-Reason-Mask/utils/vllm_cvrp_mask_processor.py:51`
`_PatchedGenerateRequest`	software	healthy	`UniCOP-Reason-Mask/utils/vllm_serve_logprobs.py:150`
`_PatchedGenerateResponse`	software	healthy	`UniCOP-Reason-Mask/utils/vllm_serve_logprobs.py:163`
`MaskConfig`	software	healthy	`UniCOP-Reason-Mask/utils/cvrp_mask_state.py:43`
`CVRPState`	software	healthy	`UniCOP-Reason-Mask/utils/cvrp_mask_state.py:56`
`MaskDecision`	software	healthy	`UniCOP-Reason-Mask/utils/cvrp_mask_state.py:70`
`PIPDWrapper`	software	healthy	`UniCOP-Reason-Mask/utils/pipd_wrapper.py:107`
`State`	software	healthy	`UniCOP-Reason-Mask/openrlhf/reward/remote_reward_server.py:…`
`_NoRepeatNgramCallable`	software	healthy	`UniCOP-Reason-Mask/openrlhf/custom/ngram_processor.py:43`
`NoRepeatNgramProcessor`	software	healthy	`UniCOP-Reason-Mask/openrlhf/custom/ngram_processor.py:77`
`TSP`	software	healthy	`UniCOP-Reason-Mask/problems/tsp.py:8`
`VRPTW`	software	healthy	`UniCOP-Reason-Mask/problems/vrptw.py:8`
`TSPTW`	software	healthy	`UniCOP-Reason-Mask/problems/tsptw.py:14`
`CVRP`	software	healthy	`UniCOP-Reason-Mask/problems/cvrp.py:14`
`ProblemBase`	software	healthy	`UniCOP-Reason-Mask/problems/base.py:15`
`TSPDL`	software	healthy	`UniCOP-Reason-Mask/problems/tspdl.py:16`
`GRPOPRMTrainer`	software	healthy	`UniCOP-Reason/grpo_prm_trainer.py:60`
`Config`	software	healthy	`UniCOP-Reason/config.py:6`
`WaveConfig`	software	healthy	`UniCOP-Reason/wave_replay.py:62`
`TrajProfile`	software	healthy	`UniCOP-Reason/wave_replay.py:70`
`WaveResult`	software	healthy	`UniCOP-Reason/wave_replay.py:84`
`StepRewards`	software	healthy	`UniCOP-Reason/pomo_prm.py:60`
`ThinkStep`	software	healthy	`UniCOP-Reason/pomo_prm.py:81`
`ThinkPRMResult`	software	healthy	`UniCOP-Reason/pomo_prm.py:89`

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

Label	Layer	Status	Path
`optimal`	software	healthy	`optimal`
`Latent-SFT`	software	healthy	`Latent-SFT`
`UniCOP-Distill`	software	healthy	`UniCOP-Distill`
`stage1_solution`	software	healthy	`UniCOP-Distill/stage1_solution`
`stage2_reasoning`	software	healthy	`UniCOP-Distill/stage2_reasoning`
`FOARL`	software	healthy	`FOARL`
`tools`	software	healthy	`tools`
`UniCOP-Reason-Mask`	software	healthy	`UniCOP-Reason-Mask`
`utils`	software	healthy	`UniCOP-Reason-Mask/utils`
`openrlhf`	software	healthy	`UniCOP-Reason-Mask/openrlhf`
`reward`	software	healthy	`UniCOP-Reason-Mask/openrlhf/reward`
`scripts`	software	healthy	`UniCOP-Reason-Mask/openrlhf/scripts`
`custom`	software	healthy	`UniCOP-Reason-Mask/openrlhf/custom`
`configs`	software	healthy	`UniCOP-Reason-Mask/openrlhf/configs`
`problems`	software	healthy	`UniCOP-Reason-Mask/problems`
`UniCOP-Reason`	software	healthy	`UniCOP-Reason`
`utils`	software	healthy	`UniCOP-Reason/utils`
`openrlhf`	software	healthy	`UniCOP-Reason/openrlhf`
`reward`	software	healthy	`UniCOP-Reason/openrlhf/reward`
`scripts`	software	healthy	`UniCOP-Reason/openrlhf/scripts`
`custom`	software	healthy	`UniCOP-Reason/openrlhf/custom`
`configs`	software	healthy	`UniCOP-Reason/openrlhf/configs`
`problems`	software	healthy	`UniCOP-Reason/problems`

Label	Layer	Status	Path
`POST /generate/`	api	healthy	`UniCOP-Reason-Mask/utils/vllm_serve_logprobs.py`
`POST /get_reward`	api	healthy	`UniCOP-Reason-Mask/openrlhf/reward/remote_reward_server.py`
`GET /health`	api	healthy	`UniCOP-Reason-Mask/openrlhf/reward/remote_reward_server.py`

Label	Layer	Status	Path
`port:30`	network	healthy	`Latent-SFT/submit_smoke_repro.sh`
`port:480`	network	healthy	`Latent-SFT/submit_train_hlr.sh`

Label	Layer	Status	Path
`repobility-clone-z03jkzl_`	software	healthy	`/tmp/repobility-clone-z03jkzl_`

Label	Layer	Status	Path
`127.0.0.1`	network	healthy	`UniCOP-Reason-Mask/auto_train.sh`

Label	Layer	Status	Path
`gpu (detected)`	hardware	healthy	`smart_submit.sh`

Label	Layer	Status	Path
`vps::gcp`	hardware	healthy	`UniCOP-Reason-Mask/evaluate.py`

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/7fb9382d-d08a-4200-bb11-80bd3ced5411/

To check status programmatically (no auth required):

curl -s https://repobility.com/api/v1/public/scan/7fb9382d-d08a-4200-bb11-80bd3ced5411/

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.

ZIJIAN004/UniCOP

Complete repo analysis

function 50+

file 50+

class 50+

directory 23

endpoint 3

port 2

repo 1

service 1

gpu 1

vps 1