3.9

Adjusted Signal

3.9

Raw Pattern Score

100%

Time Factor

2026-09-21

Last Push

11.8K

Stars

Jupyter Notebook

Language

46.8K

Lines of Code

154

Files

128

Pattern Hits

2026-10-31

Scan Date

0.00

HC Hit Rate

Risk Review · Payload v1

Use each vector to guide manual verification of the underlying findings.

Payload schema: 1.0.0 · Engine: 2.0.0 · Scoring: heuristic-density-v1 · Scan date: 2026-10-31.

Cognitive Ownership Risk 13.9/100 · LOW

Can maintainers explain intent, assumptions, and trade-offs? 90 finding(s).

Hallucinated Dependency & Supply Chain Risk 1.8/100 · LOW

Are imports declared and package/API names verified? 17 finding(s).

Agent Over-Engineering Ratio 3.8/100 · LOW

Are wrappers, guards, and repeated structures necessary? 32 finding(s).

PR Review Velocity Impact 21.1/100 · LOW

Which changes need focused human review? 128 finding(s).

What These Metrics Mean

Adjusted Signal: Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score: Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor: A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits: Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate: High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files: Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.

Severity Breakdown

Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.

CRITICAL 0HIGH 0MEDIUM 47LOW 81

Directory Score Breakdown

This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.

Pattern Findings

The scanner identified 128 distinct pattern matches across 11 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.

Slop Phrases32 hits · 61 pts

Severity	File	Line	Snippet	Context
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	67	"content": "Certainly, measuring the results of a marketing strategy and making adjustments based on those metri	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	394	"content": "Using Unreal's Gameplay Ability System (GAS) in conjunction with interpolation and prediction can pr	CODE
LOW⚡	…ma_factory_example/llama_factory_data/kto_en_demo.json	903	"content": "I apologize for the confusion. The `unique_id` parameter in the `momepy.Tessellation` function expec	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	2012	"content": "Two-factor authentication (2FA) adds an extra layer of security by requiring users to provide two fo	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	2114	"content": "To adapt the program to handle in-place modification or the creation of new objects based on a boole	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	2449	"content": "I'm not sure where you've heard that information, but as an AI language model, I can share some fact	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	2790	"content": "To derive the original matrix from the rotated one, you would need to perform a 90 degrees counter-c	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	3013	"content": "The genetic factors that contribute to the intricate design of the peacock's tail feathers are quite	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	3585	"content": "As an AI language model,I cannot determine the sentiment of the product review without further conte	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	3611	"content": "I'm unable to provide you with a real-time percentage of population living in poverty in Marlborough	CODE
LOW	…ma_factory_example/llama_factory_data/kto_en_demo.json	3690	"content": "Firstly, as a fitness trainer, I highly recommend consulting with a healthcare professional before e	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	3850	"content": "Since each medium-sized box measures 2 feet on all sides, its volume will be: 222 = 8 cubic feet.	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	4774	"content": "In Feng Shui, the five elements—wood, fire, earth, metal, and water—are believed to interact in cert	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	4897	"content": "Sure, I understand the problem. However, I'm a large language model, I cannot provide a correct answ	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	4923	"content": "As an AI language model, I am not able to provide program codes or algorithms. However, I can provid	CODE
MEDIUM	…ma_factory_example/llama_factory_data/sft_zh_demo.json	4470	"output": "As an AI language model I am capable of analyzing text and identifying emotions, however, the sentence pr	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	67	"content": "Certainly, measuring the results of a marketing strategy and making adjustments based on those metri	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	394	"content": "Using Unreal's Gameplay Ability System (GAS) in conjunction with interpolation and prediction can pr	CODE
LOW⚡	…tune/llama_factory/llama_factory_data/kto_en_demo.json	903	"content": "I apologize for the confusion. The `unique_id` parameter in the `momepy.Tessellation` function expec	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	2012	"content": "Two-factor authentication (2FA) adds an extra layer of security by requiring users to provide two fo	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	2114	"content": "To adapt the program to handle in-place modification or the creation of new objects based on a boole	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	2449	"content": "I'm not sure where you've heard that information, but as an AI language model, I can share some fact	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	2790	"content": "To derive the original matrix from the rotated one, you would need to perform a 90 degrees counter-c	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	3013	"content": "The genetic factors that contribute to the intricate design of the peacock's tail feathers are quite	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	3585	"content": "As an AI language model,I cannot determine the sentiment of the product review without further conte	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	3611	"content": "I'm unable to provide you with a real-time percentage of population living in poverty in Marlborough	CODE
LOW	…tune/llama_factory/llama_factory_data/kto_en_demo.json	3690	"content": "Firstly, as a fitness trainer, I highly recommend consulting with a healthcare professional before e	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	3850	"content": "Since each medium-sized box measures 2 feet on all sides, its volume will be: 222 = 8 cubic feet.	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	4774	"content": "In Feng Shui, the five elements—wood, fire, earth, metal, and water—are believed to interact in cert	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	4897	"content": "Sure, I understand the problem. However, I'm a large language model, I cannot provide a correct answ	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	4923	"content": "As an AI language model, I am not able to provide program codes or algorithms. However, I can provid	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/sft_zh_demo.json	4470	"output": "As an AI language model I am capable of analyzing text and identifying emotions, however, the sentence pr	CODE

Self-Referential Comments10 hits · 26 pts

Severity	File	Line	Snippet	Context
MEDIUM	demo/minicpm3/code_interpreter/code_interpreter.py	105	# Create a SamplingParams object	STRING
MEDIUM	demo/minicpm3/code_interpreter/code_interpreter.py	135	# Define a regular expression pattern to match Python code blocks	COMMENT
MEDIUM⚡	…ma_factory_example/llama_factory_data/kto_en_demo.json	887	"content": "To smoothen the output polygons without creating gaps or overlaps, you can use a combination of libr	CODE
MEDIUM⚡	…ma_factory_example/llama_factory_data/kto_en_demo.json	895	"content": "I apologize for the confusion. Let's try a different approach using the `TopologicalPreserveSimplifi	CODE
MEDIUM⚡	…ma_factory_example/llama_factory_data/kto_en_demo.json	907	"content": "I have the following code:\n#%% Generate geoTiffs with index values based on one large geotiff with	CODE
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	919	"content": "To smoothen the output polygons without creating gaps or overlays between the polygons, you can use	CODE
MEDIUM⚡	…tune/llama_factory/llama_factory_data/kto_en_demo.json	887	"content": "To smoothen the output polygons without creating gaps or overlaps, you can use a combination of libr	CODE
MEDIUM⚡	…tune/llama_factory/llama_factory_data/kto_en_demo.json	895	"content": "I apologize for the confusion. Let's try a different approach using the `TopologicalPreserveSimplifi	CODE
MEDIUM⚡	…tune/llama_factory/llama_factory_data/kto_en_demo.json	907	"content": "I have the following code:\n#%% Generate geoTiffs with index values based on one large geotiff with	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	919	"content": "To smoothen the output polygons without creating gaps or overlays between the polygons, you can use	CODE

Excessive Try-Catch Wrapping21 hits · 26 pts

Severity	File	Line	Snippet	Context
LOW	demo/minicpm4/MCP/generate_example.py	63	except Exception as e:	CODE
LOW	demo/minicpm4/MCP/eval_scripts.py	83	except Exception as e:	CODE
LOW	…o/minicpm4/SurveyGeneration/src/retriever/retriever.py	163	except Exception as e:	CODE
MEDIUM	…o/minicpm4/SurveyGeneration/src/retriever/retriever.py	164	print(f"Error in call_search_engine: {e}")	CODE
MEDIUM	…o/minicpm4/SurveyGeneration/src/retriever/retriever.py	133	def call_search_engine(tool_call, topk=10):	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/run.py	51	except Exception as e:	CODE
MEDIUM	demo/minicpm4/SurveyGeneration/src/generation/run.py	52	print(f"Error sending to WebSocket: {e}")	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/run.py	210	except Exception as e:	CODE
MEDIUM	demo/minicpm4/SurveyGeneration/src/generation/run.py	211	print(f"Error posting to frontend: {e}")	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/run.py	310	except Exception as e:	CODE
MEDIUM	demo/minicpm4/SurveyGeneration/src/generation/run.py	311	print(f"Error generating survey: {e}")	CODE
LOW	demo/minicpm3/function_call/minicpm_tool_parser.py	205	except Exception as e:	CODE
LOW	demo/minicpm3/function_call/minicpm_tool_parser.py	277	except Exception as e:	CODE
LOW	demo/minicpm3/code_interpreter/code_interpreter.py	76	except Exception as e:	STRING
LOW	tool_parsers/minicpm5xml_tool_parser.py	40	except Exception: # pragma: no cover	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	108	except Exception:	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	226	except Exception:	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	260	except Exception:	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	402	except Exception as e:	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	645	except Exception:	CODE
LOW	finetune/mlx_finetune.py	441	except Exception as e:	CODE

Deep Nesting25 hits · 25 pts

Severity	File	Line	Context
LOW	demo/minicpm4/MCP/generate_example.py	25	CODE
LOW	demo/minicpm4/MCP/generate_example.py	71	CODE
LOW	demo/minicpm4/MCP/eval_scripts.py	21	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/run.py	93	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/run.py	227	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/run.py	255	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	36	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	75	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	149	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	210	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	408	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	462	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	510	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	556	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	612	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	639	CODE
LOW	demo/minicpm3/function_call/minicpm_tool_parser.py	150	CODE
LOW	demo/minicpm3/function_call/minicpm_tool_parser.py	239	CODE
LOW	demo/minicpm3/function_call/minicpm_tool_parser.py	75	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	161	CODE
LOW	finetune/finetune.py	78	CODE
LOW	finetune/sft_dpo_trainer/finetune_dpo_trainer.py	85	CODE
LOW	finetune/sft_dpo_trainer/finetune_dpo_trainer.py	184	CODE
LOW	finetune/sft_dpo_trainer/finetune_dpo_trainer.py	228	CODE
LOW	minicpm_sala/finetune/trainer/finetune.py	78	CODE

Unused Imports17 hits · 17 pts

Severity	File	Line	Context
LOW	demo/minicpm/langchain_demo.py	35	CODE
LOW	demo/minicpm/langchain_demo.py	40	CODE
LOW	demo/minicpm4/MCP/eval_scripts.py	1	CODE
LOW	…inicpm4/SurveyGeneration/src/preprocess/build_index.py	6	CODE
LOW	…inicpm4/SurveyGeneration/src/preprocess/build_index.py	8	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	276	CODE
LOW	demo/minicpm3/code_interpreter/code_interpreter.py	3	CODE
LOW	demo/minicpm3/code_interpreter/code_interpreter.py	4	CODE
LOW	demo/minicpm3/code_interpreter/code_interpreter.py	6	CODE
LOW	demo/minicpm3/code_interpreter/code_interpreter.py	7	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	6	CODE
LOW	quantize/awq_quantize.py	4	CODE
LOW	quantize/quantize_eval.py	5	CODE
LOW	finetune/mlx_finetune.py	46	CODE
LOW	finetune/sft_dpo_trainer/finetune_dpo_trainer.py	5	CODE
LOW	finetune/sft_dpo_trainer/finetune_dpo_trainer.py	5	CODE
LOW	finetune/sft_dpo_trainer/finetune_dpo_trainer.py	19	CODE

Hyper-Verbose Identifiers10 hits · 10 pts

Severity	File	Line	Snippet	Context
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	75	def convert_survey_dict_to_str(current_survey):	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	149	def convert_survey_dict_to_abbr_str(current_survey):	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	408	def _build_user_prompt_force_correct(query, current_survey, trajs):	CODE
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	510	def build_prompt_for_generator(self):	CODE
LOW	demo/minicpm3/function_call/minicpm_tool_parser.py	75	def extract_tool_calls_streaming(	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	464	def _process_complete_block_streaming(	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	526	def _process_partial_block_streaming(	CODE
LOW	tool_parsers/minicpm5xml_tool_parser.py	565	def extract_tool_calls_streaming(	CODE
LOW	quantize/quantize_data/alpaca_data_cleaned.json	730	"output": "Here's a Python implementation of the function:\n\n```\ndef uppercase_unless_rejected(string):\n if st	CODE
LOW	finetune/sft_dpo_trainer/finetune_dpo_trainer.py	228	def encode_conversation_with_labels(self, messages, response):	CODE

Redundant / Tautological Comments4 hits · 5 pts

Severity	File	Line	Snippet	Context
LOW	demo/minicpm3/code_interpreter/code_interpreter.py	165	# Check if the response contains the termination keyword	COMMENT
LOW	finetune/mlx_finetune.py	487	# Check if any sequence is longer than 2048 tokens	COMMENT
LOW	…ma_factory_example/llama_factory_data/sft_zh_demo.json	275	"output": "def calculate_sum(numbers):\n if not isinstance(numbers, list): # Check if input is a list\n r	CODE
LOW	…tune/llama_factory/llama_factory_data/sft_zh_demo.json	275	"output": "def calculate_sum(numbers):\n if not isinstance(numbers, list): # Check if input is a list\n r	CODE

Modern AI Meta-Vocabulary2 hits · 4 pts

Severity	File	Line	Snippet	Context
MEDIUM	README.md	183	During post-training, we proceed in three steps: SFT, RL, and OPD. We first use **400B tokens of deep-th	CODE
MEDIUM	README.md	189	RL + OPD is a key part of MiniCPM5-2B post-training. During the RL stage, we adopted the critic-based algorithm	CODE

Over-Commented Block4 hits · 4 pts

Severity	File	Line	Snippet	Context
LOW	…/SurveyGeneration/frontend/minicpm4-survey/src/App.jsx	181	};	COMMENT
LOW	demo/minicpm4/SurveyGeneration/src/generation/buffer.py	1		COMMENT
LOW	demo/minicpm3/function_call/function_calling.py	41	# "tool_calls": [	COMMENT
LOW	demo/minicpm3/function_call/function_calling.py	61	# {	COMMENT

AI Slop Vocabulary2 hits · 4 pts

Severity	File	Line	Snippet	Context
MEDIUM	…ma_factory_example/llama_factory_data/kto_en_demo.json	2769	"content": " As an Azure Cloud Engineer working with Microsoft Azure, you can indeed utilize Azure Active Direct	CODE
MEDIUM	…tune/llama_factory/llama_factory_data/kto_en_demo.json	2769	"content": " As an Azure Cloud Engineer working with Microsoft Azure, you can indeed utilize Azure Active Direct	CODE

AI Structural Patterns1 hit · 1 pts

Severity	File	Line	Snippet	Context
LOW	…o/minicpm4/SurveyGeneration/src/retriever/retriever.py	131		CODE

OpenBMB/MiniCPM

Open-Source Code Quality, Security and Technical Debt Review