1.7

Adjusted Signal

1.7

Raw Pattern Score

100%

Time Factor

2026-10-21

Last Push

14.5K

Stars

C++

Language

722.4K

Lines of Code

1.9K

Files

1.1K

Pattern Hits

2026-10-31

Scan Date

0.01

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 8.0/100 · LOW

Can maintainers explain intent, assumptions, and trade-offs? 1088 finding(s).

Hallucinated Dependency & Supply Chain Risk 0.1/100 · LOW

Are imports declared and package/API names verified? 8 finding(s).

Agent Over-Engineering Ratio 0.3/100 · LOW

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

PR Review Velocity Impact 8.0/100 · LOW

Which changes need focused human review? 1098 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 10MEDIUM 25LOW 1063

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 1098 distinct pattern matches across 13 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.

Over-Commented Block1004 hits · 1000 pts

Severity	File	Line	Snippet	Context
LOW	CMakeLists.txt	21	# whatever you are building rather than create a standalone copy of dlib.	COMMENT
LOW	tools/python/src/face_recognition.cpp	1	// Copyright (C) 2017 Davis E. King (davis@dlib.net)	COMMENT
LOW	tools/python/src/dlib.cpp	41	{	COMMENT
LOW	tools/python/src/dlib.cpp	61	m.attr("DLIB_USE_CUDA") = false;	COMMENT
LOW	tools/python/src/simple_object_detector.h	1	// Copyright (C) 2014 Davis E. King (davis@dlib.net)	COMMENT
LOW	tools/python/src/object_detection.cpp	1	// Copyright (C) 2015 Davis E. King (davis@dlib.net)	COMMENT
LOW	tools/python/src/opaque_types.h	1	// Copyright (C) 2017 Davis E. King (davis@dlib.net)	COMMENT
LOW	…ts_to_caffe/running_a_dlib_model_with_caffe_example.py	1	#!/usr/bin/env python	COMMENT
LOW	tools/imglab/src/flip_dataset.cpp	1	// Copyright (C) 2011 Davis E. King (davis@dlib.net)	COMMENT
LOW	tools/imglab/src/convert_pascal_xml.h	1	// Copyright (C) 2011 Davis E. King (davis@dlib.net)	COMMENT
LOW	tools/imglab/src/metadata_editor.cpp	1	// Copyright (C) 2011 Davis E. King (davis@dlib.net)	COMMENT
LOW	tools/imglab/src/cluster.cpp	1	// Copyright (C) 2015 Davis E. King (davis@dlib.net)	COMMENT
LOW	tools/imglab/src/main.cpp	1		COMMENT
LOW	tools/htmlify/to_xml.cpp	1		COMMENT
LOW	docs/docs/python/conf.py	1	# -- coding: utf-8 --	COMMENT
LOW	docs/docs/python/conf.py	41	# The encoding of source files.	COMMENT
LOW	docs/docs/python/conf.py	61	# for a list of supported languages.	COMMENT
LOW	docs/docs/python/conf.py	81	# unit titles (such as .. function::).	COMMENT
LOW	docs/docs/python/conf.py	101	# Theme options are theme-specific and customize the look and feel of a theme	COMMENT
LOW	docs/docs/python/conf.py	121	# docs. This file should be a Windows icon file (.ico) being 16x16 or 32x32	COMMENT
LOW	docs/docs/python/conf.py	141	}	COMMENT
LOW	docs/docs/python/conf.py	161		COMMENT
LOW	docs/docs/python/conf.py	201	# For "manual" documents, if this is true, then toplevel headings are parts,	COMMENT
LOW	examples/kcentroid_ex.cpp	41		COMMENT
LOW	examples/dnn_dcgan_train_ex.cpp	201	// This part is the essence of the Generative Adversarial Networks. Until now, we have	COMMENT
LOW	examples/svr_ex.cpp	81	// -0.189201 -0.187636	COMMENT
LOW	examples/dnn_mmod_train_find_cars_ex.cpp	101	std::vector<std::vector<mmod_rect>> boxes_train, boxes_test;	COMMENT
LOW	examples/dnn_mmod_train_find_cars_ex.cpp	121	// that are heavily overlapped by another box as ignore. We do this because we want to	COMMENT
LOW	examples/dnn_mmod_train_find_cars_ex.cpp	141	// Importantly, this means that if your training dataset contains an image with two	COMMENT
LOW	examples/dnn_mmod_train_find_cars_ex.cpp	181		COMMENT
LOW	examples/dnn_mmod_train_find_cars_ex.cpp	221	// Our vehicle detection dataset has basically 3 different types of boxes. Square	COMMENT
LOW	examples/dnn_mmod_train_find_cars_ex.cpp	241	mmod_options options(boxes_train, 70, 30);	COMMENT
LOW	examples/dnn_mmod_train_find_cars_ex.cpp	261	// detection overlaps a box in the training data with ignore==true then that detection	COMMENT
LOW	examples/bayes_net_gui_ex.cpp	141		COMMENT
LOW	examples/CMakeLists.txt	1	#	COMMENT
LOW	examples/CMakeLists.txt	21	# / ____/ __ \\| \/ \| \/ \| ____\| \ \| \|__ __/ ____\| \| \| \| \| \|	COMMENT
LOW	examples/CMakeLists.txt	41	# into your project. Note that you don't need to compile or install dlib. All	COMMENT
LOW	examples/CMakeLists.txt	61	# file, assignment_learning_ex.cpp. If your program consisted of multiple .cpp	COMMENT
LOW	examples/CMakeLists.txt	81	# if instead you had written:	COMMENT
LOW	examples/kkmeans_ex.cpp	41	// results without much fiddling.	COMMENT
LOW	examples/rvm_regression_ex.cpp	41		COMMENT
LOW	examples/compress_stream_ex.cpp	61	// In the code below, we use the parser.print_options() method to print all our	COMMENT
LOW	examples/compress_stream_ex.cpp	81	// that shouldn't appear more than once and then I just call check_one_time_options()	COMMENT
LOW	examples/dnn_introduction2_ex.cpp	21		COMMENT
LOW	examples/dnn_introduction2_ex.cpp	241	layer<tag1>(pnet);	COMMENT
LOW	examples/dnn_introduction2_ex.cpp	261	// momentum parameters.	COMMENT
LOW	examples/dnn_introduction2_ex.cpp	301	auto idx = rnd.get_random_32bit_number()%training_images.size();	COMMENT
LOW	examples/bsp_ex.cpp	121		COMMENT
LOW	examples/bsp_ex.cpp	161	{	COMMENT
LOW	examples/bsp_ex.cpp	181		COMMENT
LOW	examples/bsp_ex.cpp	221	}	COMMENT
LOW	examples/bsp_ex.cpp	241	// terminate. That is, we know it is time to terminate if all the messages between	COMMENT
LOW	examples/dnn_mmod_find_cars_ex.cpp	81		COMMENT
LOW	examples/svm_pegasos_ex.cpp	41	// This is a typedef for the type of kernel we are going to use in this example.	COMMENT
LOW	examples/svm_pegasos_ex.cpp	121	// with the pegasos algorithm. But it is often the case that you have a dataset and you	COMMENT
LOW	examples/model_selection_ex.cpp	81	// binary SVM with the RBF kernel to classify the data. However, there are	COMMENT
LOW	examples/object_detector_ex.cpp	181	// the feature vectors output by the hashed_feature_image will have dimension pow(2,10).	COMMENT
LOW	examples/object_detector_ex.cpp	201	// the details into the scanner object we are ready to use the structural_object_detection_trainer	COMMENT
LOW	examples/krls_filter_ex.cpp	41	typedef radial_basis_kernel<sample_type> kernel_type;	COMMENT
LOW	examples/gui_api_ex.cpp	221	// would see a black console window pop up when you ran it. The following	COMMENT
944 more matches not shown…

AI Response Leakage9 hits · 68 pts

Severity	File	Line	Snippet	Context
HIGH	dlib/global_optimization/find_max_global.h	49	// So we write the terrible garbage in the #else for visual studio. When Visual Studio supports C++11 I'll upda	COMMENT
HIGH	python_examples/sequence_segmenter.py	134	# sparse or dense representation depending on our needs. In this example, we	COMMENT
HIGH	python_examples/svm_binary_classifier.py	6	# the dlib C++ Library. In this example, we will create a simple test dataset	COMMENT
HIGH	python_examples/face_clustering.py	7	# In this example, we assume the largest cluster will contain photos of the common person in the	COMMENT
HIGH	python_examples/svm_rank.py	10	# In this example, we will create a simple test dataset and show how to learn a	COMMENT
HIGH	python_examples/svm_struct.py	35	# In this example, we have three types of samples: class 0, 1, or 2. That	COMMENT
HIGH	python_examples/train_object_detector.py	34	# In this example we are going to train a face detector based on the small	COMMENT
HIGH	python_examples/train_object_detector.py	127	# In this example we load detector.svm again since it's the only one we have on	COMMENT
HIGH	python_examples/train_shape_predictor.py	42	# In this example we are going to train a face detector based on the small	COMMENT

Slop Phrases21 hits · 44 pts

Severity	File	Line	Snippet	Context
MEDIUM	docs/docs/index.xml	174	<li>A simple <a href="network.html#server_http">HTTP server</a> object you can use to embed a	CODE
MEDIUM	docs/docs/linear_algebra.xml	1346	<a href="#point">point</a>. Note that you can use <a href="#find_affine_transform">find_affine_transform</a	CODE
MEDIUM	docs/docs/linear_algebra.xml	1429	<a href="#point">point</a>. Note that you can use <a href="#find_projective_transform">find_projective_tran	CODE
MEDIUM	examples/matrix_expressions_ex.cpp	176	As discussed above, constructing this object doesn't compute anything. It just	CODE
LOW	dlib/xml_parser/xml_parser_kernel_1.h	578	// don't forget to rethrow the exception	COMMENT
LOW	dlib/optimization/elastic_net.h	204	// Do the base dot product but don't forget to add in the -(1/t)*y part from the svm reduction paper	COMMENT
LOW	dlib/optimization/optimization_solve_qp_using_smo.h	698	// Don't forget to include the Q_offdiag terms in the computation. Note that	COMMENT
LOW	dlib/optimization/optimization.h	266	// Don't forget to negate these outputs from the line search since they are	COMMENT
LOW	dlib/optimization/optimization.h	668	// Don't forget to negate the output from the line search since it is from the	COMMENT
LOW	dlib/random_forest/random_forest_regression.h	478	// Add to leaves. Don't forget to set the pointer in the	COMMENT
LOW	dlib/random_forest/random_forest_regression.h	502	// Add to leaves. Don't forget to set the pointer in the	COMMENT
MEDIUM	dlib/svm/roc_trainer_abstract.h	83	as discussed above, and finally returns the decision function.	CODE
LOW	dlib/svm/feature_ranking.h	331	// we are doing the optimization in log space so don't forget to convert back to normal space	COMMENT
LOW	dlib/gui_core/gui_core_kernel_2.h	12	#error "Also make sure you have libx11-dev installed on your system"	COMMENT
LOW	dlib/control/mpc.h	356	// If this is the last SMO iteration then don't forget to initialize v	COMMENT
MEDIUM	python_examples/sequence_segmenter.py	63	# you can use "unsorted" sparse vectors. This means you can add the index/value	COMMENT
MEDIUM	python_examples/sequence_segmenter.py	133	# convert them into arrays of vectors as discussed above. We can use either a	COMMENT
MEDIUM	python_examples/svm_rank.py	133	# you can use the dlib.make_sparse_vector() routine to make a sparse vector	COMMENT
MEDIUM⚡	python_examples/svm_struct.py	241	# Create a dense vector object (note that you can also use unsorted	COMMENT
MEDIUM	python_examples/train_object_detector.py	4	# This example program shows how you can use dlib to make a HOG based object	COMMENT
MEDIUM	python_examples/train_object_detector.py	73	# own XML files you can use the imglab tool which can be found in the	COMMENT

Hyper-Verbose Identifiers30 hits · 34 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	244	def read_version_from_cmakelists(cmake_file):	STRING
LOW	tools/python/test/test_numpy_returns.py	90	def test_jitter_image_from_multiple_threads():	CODE
LOW	tools/python/test/test_setup.py	20	def test_windows_extension_matches_python_architecture(	CODE
LOW⚡	tools/python/test/test_array.py	15	def test_array_init_with_number():	CODE
LOW⚡	tools/python/test/test_array.py	23	def test_array_init_with_negative_number():	CODE
LOW⚡	tools/python/test/test_array.py	28	def test_array_init_with_zero():	CODE
LOW⚡	tools/python/test/test_array.py	33	def test_array_init_with_list():	CODE
LOW⚡	tools/python/test/test_array.py	41	def test_array_init_with_empty_list():	CODE
LOW⚡	tools/python/test/test_array.py	46	def test_array_init_without_argument():	CODE
LOW⚡	tools/python/test/test_array.py	51	def test_array_init_with_tuple():	CODE
LOW⚡	tools/python/test/test_array.py	58	def test_array_serialization_empty():	CODE
LOW	tools/python/test/test_array.py	83	def test_array_string_representations_empty():	CODE
LOW	tools/python/test/test_array.py	89	def test_array_string_representations():	CODE
LOW	tools/python/test/test_global_optimization.py	7	def test_global_optimization_nargs():	CODE
LOW	tools/python/test/test_global_optimization.py	33	def test_global_function_search():	CODE
LOW	tools/python/test/test_matrix.py	43	def test_matrix_from_list_with_invalid_rows():	CODE
LOW	tools/python/test/test_matrix.py	50	def test_matrix_from_list_as_column_vector():	CODE
LOW	tools/python/test/test_matrix.py	80	def test_matrix_from_object_without_shape():	CODE
LOW	tools/python/test/test_vector.py	17	def test_vector_init_with_number():	CODE
LOW	tools/python/test/test_vector.py	39	def test_vector_init_with_list():	CODE
LOW	tools/python/test/test_vector.py	72	def test_vector_invalid_getitem():	CODE
LOW	tools/python/test/test_vector.py	80	def test_vector_init_with_negative_number():	CODE
LOW	tools/python/test/test_vector.py	94	def test_vector_serialization():	CODE
LOW	tools/python/test/test_vector.py	119	def test_vectors_serialization():	CODE
LOW	tools/python/test/test_vector.py	146	def test_vectorss_serialization():	CODE
LOW	tools/python/test/test_range.py	34	def test_range_with_negative_elements():	CODE
LOW	tools/python/test/test_chinese_whispers.py	20	def test_chinese_whispers_with_distance():	CODE
LOW	tools/python/test/test_chinese_whispers.py	41	def test_chinese_whispers_type_checks():	CODE
LOW	python_examples/sequence_segmenter.py	66	def sentence_to_sparse_vectors(sentence):	CODE
LOW	python_examples/svm_struct.py	270	def get_truth_joint_feature_vector(self, idx):	CODE

AI Slop Vocabulary12 hits · 30 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	226	# When building on travis-ci, just use 2 cores since travis-ci limits	STRING
MEDIUM	examples/dnn_mmod_train_find_cars_ex.cpp	126	// the detector works. Essentially, sliding window detectors scan all image locations	COMMENT
MEDIUM	examples/dnn_mmod_train_find_cars_ex.cpp	295	// do that. Essentially, we are setting the first argument to infinity, and only the	COMMENT
MEDIUM	examples/linear_manifold_regularizer_ex.cpp	219	// of times with different randomly chosen points so we can see how robust the result is.	COMMENT
MEDIUM	dlib/test/linear_manifold_regularizer.cpp	182	// of times with different randomly chosen points so we can see how robust the result is.	COMMENT
MEDIUM	dlib/timer/timer.cpp	43	// there is no (safe/robust)solution, but best practices are described here	COMMENT
MEDIUM	dlib/dnn/trainer.h	829	// values in previous_loss_values then we are robust to that kind	COMMENT
LOW⚡	python_examples/svm_struct.py	245	# would simply return a dlib.sparse_vector at the end of make_psi() and	COMMENT
LOW	python_examples/train_object_detector.py	76	# file. But for this example, we just use the training.xml file included with	COMMENT
MEDIUM	python_examples/train_shape_predictor.py	88	# face landmarking datasets you will obtain state-of-the-art results, as shown	COMMENT
MEDIUM	python_examples/face_recognition.py	13	# comparable to other state-of-the-art methods for face recognition as of	COMMENT
LOW	python_examples/face_recognition.py	131	# Now we simply pass this chip (aligned image) to the api	COMMENT

Self-Referential Comments4 hits · 14 pts

Severity	File	Line	Snippet	Context
MEDIUM	docs/docs/python/conf.py	6	# This file is execfile()d with the current directory set to its containing dir.	COMMENT
MEDIUM	python_examples/sequence_segmenter.py	44	# Create an empty array of vectors	COMMENT
MEDIUM⚡	python_examples/svm_struct.py	241	# Create a dense vector object (note that you can also use unsorted	COMMENT
MEDIUM	python_examples/correlation_tracker.py	43	# Create the correlation tracker - the object needs to be initialized	COMMENT

Unused Imports8 hits · 8 pts

Severity	File	Line	Context
LOW	tools/python/test/test_numpy_returns.py	2	CODE
LOW	tools/python/test/test_svm_c_trainer.py	1	CODE
LOW	docs/docs/python/generate_dlib_listing.py	1	CODE
LOW	dlib/external/pybind11/tools/libsize.py	1	CODE
LOW	dlib/external/pybind11/tools/make_changelog.py	7	CODE
LOW	dlib/external/pybind11/tools/make_global.py	6	CODE
LOW	…l/pybind11/tools/codespell_ignore_lines_from_errors.py	12	CODE
LOW	python_examples/opencv_webcam_face_detection.py	38	CODE

Synthetic Comment Markers1 hit · 8 pts

Severity	File	Line	Snippet	Context
HIGH	dlib/external/pybind11/include/pybind11/detail/class.h	662	// Fill in all the information, and then downgrade as requested by the caller, or raise an	COMMENT

Decorative Section Separators2 hits · 6 pts

Severity	File	Line	Snippet	Context
MEDIUM	dlib/logger/logger_config_file.h	53	# -----------------------------------------------	COMMENT
MEDIUM	dlib/logger/logger_config_file.h	55	# -----------------------------------------------	COMMENT

Redundant / Tautological Comments2 hits · 4 pts

Severity	File	Line	Snippet	Context
LOW	…ts_to_caffe/running_a_dlib_model_with_caffe_example.py	70	# Print outputs, looping over minibatch. You should see that the network	COMMENT
LOW⚡	python_examples/svm_struct.py	248	# Set it to have 9 dimensions. Note that the elements of the vector	COMMENT

Deep Nesting2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	48		CODE
LOW	…l/pybind11/tools/codespell_ignore_lines_from_errors.py	17		CODE

Verbosity Indicators2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	dlib/svm/linearly_independent_subset_finder_abstract.h	296	- The purpose of this function is to fill lisf with points from samples. It does	CODE
LOW	dlib/matrix/matrix_abstract.h	637	The purpose of this function is to enable you to easily initialize a matrix object.	CODE

Example Usage Blocks1 hit · 2 pts

Severity	File	Line	Snippet	Context
LOW	dlib/external/pybind11/tools/libsize.py	7	# Usage:	COMMENT

davisking/dlib

Open-Source Code Quality, Security and Technical Debt Review