abetlen/llama-cpp-python

9.0

Adjusted Score

9.0

Raw Score

100%

Time Factor

2026-08-17

Last Push

10.6K

Stars

Python

Language

45.3K

Lines of Code

Files

519

Pattern Hits

2026-08-29

Scan Date

0.07

HC Hit Rate

What These Metrics Mean

Adjusted Score: Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score: The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor: The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits: Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate: High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files: Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.

Severity Breakdown

Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.

CRITICAL 0HIGH 7MEDIUM 14LOW 498

Directory Score Breakdown

This horizontal bar chart decomposes the repository's raw synthetic code score by top-level directory, allowing you to pinpoint precisely which modules or components carry the highest AI authorship density. Directories with disproportionately high scores relative to their size warrant targeted manual review: concentrated AI signatures often trace back to mass-generated configuration layers, auto-ported test suites, LLM-scaffolded boilerplate classes, or entire subsystems authored under heavy copilot assistance. Use this view to prioritise your human code-review effort.

Pattern Findings

The scanner identified 519 distinct pattern matches across 14 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. 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 diagnostic confidence level (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — comment-context matches receive a ×1.5 weight because LLMs systematically over-annotate. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier as dense clusters constitute far stronger evidence of synthetic authorship than isolated hits.

Hyper-Verbose Identifiers306 hits · 202 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_llama.py	69	def llama_cpp_transformer_model_path():	CODE
LOW⚡	tests/test_llama.py	77	def llama_cpp_embedding_model_path():	CODE
LOW⚡	tests/test_llama.py	85	def llama_cpp_recurrent_model_path():	CODE
LOW⚡	tests/test_llama.py	93	def llama_cpp_hybrid_model_path():	CODE
LOW	tests/test_llama.py	23	def test_llama_cpp_tokenization():	CODE
LOW	tests/test_llama.py	270	def test_real_llama_repeated_prompt_cache(llama_cpp_model_path):	STRING
LOW	tests/test_llama.py	301	def _assert_prompt_cache_reset_handles_history_edit(	STRING
LOW	tests/test_llama.py	344	def test_recurrent_model_prompt_cache_reset(llama_cpp_recurrent_model_path):	STRING
LOW	tests/test_llama.py	352	def test_hybrid_model_prompt_cache_reset(llama_cpp_hybrid_model_path):	STRING
LOW	tests/test_llama.py	394	def _eval_alternate_same_length_prompt(model, tokens, expected_next_token):	STRING
LOW	tests/test_llama.py	418	def _assert_exact_cached_prompt_reuse_matches_fresh(	STRING
LOW	tests/test_llama.py	458	def _assert_loaded_exact_cached_prompt_reuse_matches_fresh(	STRING
LOW	tests/test_llama.py	495	def _assert_ram_cache_exact_prompt_hit_matches_fresh(	STRING
LOW	tests/test_llama.py	550	def _assert_shorter_prompt_prefix_reuse_matches_fresh(	STRING
LOW	tests/test_llama.py	578	def test_exact_cached_prompt_reuse_matches_fresh(	STRING
LOW	tests/test_llama.py	594	def test_loaded_exact_cached_prompt_reuse_matches_fresh(	STRING
LOW	tests/test_llama.py	610	def test_ram_cache_exact_prompt_hit_matches_fresh(	STRING
LOW	tests/test_llama.py	626	def test_shorter_prompt_prefix_reuse_matches_fresh(	STRING
LOW	tests/test_llama.py	639	def test_real_llama_embeddings(llama_cpp_embedding_model_path):	STRING
LOW	tests/test_llama_grammar.py	18	def test_composed_pydantic_grammar():	STRING
LOW	tests/test_llama_speculative.py	6	def test_find_candidate_pred_tokens():	CODE
LOW	tests/test_llama_chat_format.py	85	def test_hf_tokenizer_config_str_to_chat_formatter():	CODE
LOW⚡	examples/server/server.py	13182	def request_needs_prompt_logits(request: CompletionRequest) -> bool:	CODE
LOW⚡	examples/server/server.py	13188	def request_needs_uncached_prompt_logprobs(request: CompletionRequest) -> bool:	CODE
LOW⚡	examples/server/server.py	13192	def clear_sequence_cache_match(request: CompletionRequest) -> None:	CODE
LOW⚡	examples/server/server.py	13196	def can_lookup_sequence_cache(self, request: CompletionRequest) -> bool:	CODE
LOW⚡	examples/server/server.py	13203	def is_sequence_cache_match_usable(	CODE
LOW	examples/server/server.py	4529	def _regex_literal_prefix_and_remainder(pattern: str) -> Tuple[str, str]:	CODE
LOW	examples/server/server.py	4584	def _consume_optional_literal_prefix(	CODE
LOW	examples/server/server.py	4599	def _split_regex_alternatives(pattern: str) -> List[str]:	CODE
LOW	examples/server/server.py	4633	def _regex_lookahead_literal_specs(cls, pattern: str) -> List[Tuple[str, bool]]:	CODE
LOW	examples/server/server.py	4671	def _regex_capture_end_literal_specs(cls, pattern: str) -> List[Tuple[str, bool]]:	CODE
LOW	examples/server/server.py	4683	def _regex_capture_end_literals(cls, pattern: str) -> List[str]:	CODE
LOW	examples/server/server.py	4728	def _literal_suffix_prefix_length(text: str, literal: str) -> int:	CODE
LOW	examples/server/server.py	4763	def _compile_iterator_pattern(pattern: str) -> Optional[Tuple[str, str]]:	CODE
LOW	examples/server/server.py	4774	def _compile_iterator_block_pattern(	CODE
LOW	examples/server/server.py	4853	def _compile_word_capture_pattern(	CODE
LOW	examples/server/server.py	4872	def _consume_until_any_literal(	CODE
LOW	examples/server/server.py	4907	def _compile_tool_call_item_plan(	CODE
LOW	examples/server/server.py	4976	def _compile_segment_message_plan(	CODE
LOW	examples/server/server.py	5045	def _compile_tagged_message_plan(	CODE
LOW	examples/server/server.py	5224	def _parameter_schema_for_tool(	CODE
LOW	examples/server/server.py	5250	def _raw_string_tool_arguments(self, tool_name: str, value: str) -> Optional[Dict[str, str]]:	CODE
LOW	examples/server/server.py	5282	def _single_string_tool_argument_name(self, tool_name: str) -> Optional[str]:	CODE
LOW	examples/server/server.py	5314	def _text_tool_argument_from_object(	CODE
LOW	examples/server/server.py	5364	def _raw_object_tool_arguments(cls, value: str) -> Optional[Dict[str, Any]]:	CODE
LOW	examples/server/server.py	5386	def _trim_partial_gemma_quote_marker(value: str) -> str:	CODE
LOW	examples/server/server.py	5585	def _advance_direct_tool_call_state(self, text: str) -> Tuple[bool, List[Dict[str, Any]]]:	CODE
LOW	examples/server/server.py	5836	def _advance_direct_stream_state(self, text: str) -> Tuple[bool, List[Dict[str, Any]]]:	CODE
LOW	examples/server/server.py	6673	def _partial_regex_key_value_item(	CODE
LOW	examples/server/server.py	6708	def _trim_partial_tool_call_prefix(	CODE
LOW	examples/server/server.py	7172	def _normalize_tool_call_item(	CODE
LOW	examples/server/server.py	7255	def _serialize_partial_json_prefix(cls, value: Any) -> str:	CODE
LOW	examples/server/server.py	7302	def _contains_partial_json_value(cls, value: Any) -> bool:	CODE
LOW	examples/server/server.py	7314	def _serialize_partial_json_state(cls, value: Any) -> Tuple[str, bool]:	CODE
LOW	examples/server/server.py	7357	def _serialize_tool_arguments(cls, arguments: Any, *, partial: bool = False) -> str:	CODE
LOW	examples/server/server.py	7897	def _tools_for_response_parser(	CODE
LOW	examples/server/server.py	7933	def _response_reasoning_effort(	CODE
LOW	examples/server/server.py	7945	def _response_text_from_content(content: Any) -> str:	CODE
LOW	examples/server/server.py	7969	def _response_chat_content_from_content(	CODE
246 more matches not shown…

Deep Nesting65 hits · 40 pts

Severity	File	Line	Context
LOW	docker/open_llama/hug_model.py	29	CODE
LOW	examples/low_level_api/low_level_api_chat_cpp.py	344	CODE
LOW	examples/low_level_api/low_level_api_chat_cpp.py	705	CODE
LOW	examples/server/server.py	15763	CODE
LOW	examples/server/server.py	329	CODE
LOW	examples/server/server.py	364	CODE
LOW	examples/server/server.py	526	CODE
LOW	examples/server/server.py	1818	CODE
LOW	examples/server/server.py	2239	CODE
LOW	examples/server/server.py	3503	CODE
LOW	examples/server/server.py	4529	CODE
LOW	examples/server/server.py	5419	CODE
LOW	examples/server/server.py	5585	CODE
LOW	examples/server/server.py	5836	CODE
LOW	examples/server/server.py	6003	CODE
LOW	examples/server/server.py	6380	CODE
LOW	examples/server/server.py	6734	CODE
LOW	examples/server/server.py	7366	CODE
LOW	examples/server/server.py	7687	CODE
LOW	examples/server/server.py	8137	CODE
LOW	examples/server/server.py	8698	CODE
LOW	examples/server/server.py	8952	CODE
LOW	examples/server/server.py	9308	CODE
LOW	examples/server/server.py	9492	CODE
LOW	examples/server/server.py	10021	CODE
LOW	examples/server/server.py	10999	CODE
LOW	examples/server/server.py	11238	CODE
LOW	examples/server/server.py	11600	CODE
LOW	examples/server/server.py	12770	CODE
LOW	examples/server/server.py	13389	CODE
LOW	examples/server/server.py	13563	CODE
LOW	examples/server/server.py	13790	CODE
LOW	examples/server/server.py	13886	CODE
LOW	examples/server/server.py	14339	CODE
LOW	examples/server/server.py	15892	CODE
LOW	examples/server/server.py	15937	CODE
LOW	examples/server/server.py	16200	CODE
LOW	examples/server/server.py	330	CODE
LOW	examples/server/server.py	377	CODE
LOW	llama_cpp/llama_chat_format.py	1460	CODE
LOW	llama_cpp/llama_chat_format.py	1823	CODE
LOW	llama_cpp/llama_chat_format.py	4132	CODE
LOW	llama_cpp/llama_chat_format.py	1490	CODE
LOW	llama_cpp/llama_chat_format.py	1574	CODE
LOW	llama_cpp/llama_chat_format.py	1874	CODE
LOW	llama_cpp/llama_chat_format.py	2139	CODE
LOW	llama_cpp/llama_chat_format.py	2850	CODE
LOW	llama_cpp/llama_chat_format.py	3140	CODE
LOW	llama_cpp/llama_chat_format.py	3401	CODE
LOW	llama_cpp/llama_chat_format.py	3674	CODE
LOW	llama_cpp/llama_chat_format.py	1624	CODE
LOW	llama_cpp/llama.py	60	CODE
LOW	llama_cpp/llama.py	704	CODE
LOW	llama_cpp/llama.py	854	CODE
LOW	llama_cpp/llama.py	1195	CODE
LOW	llama_cpp/_ctypes_extensions.py	26	CODE
LOW	llama_cpp/llama_grammar.py	439	CODE
LOW	llama_cpp/llama_grammar.py	500	CODE
LOW	llama_cpp/llama_grammar.py	697	CODE
LOW	llama_cpp/llama_grammar.py	446	CODE
5 more matches not shown…

Unused Imports40 hits · 34 pts

Severity	File	Line	Context
LOW	tests/test_llama.py	1	CODE
LOW	tests/test_llama.py	6	CODE
LOW	examples/low_level_api/ReasonAct.py	2	CODE
LOW	examples/server/server.py	17	CODE
LOW	examples/server/server.py	39	CODE
LOW	examples/batch-processing/server.py	26	CODE
LOW	llama_cpp/llama_cache.py	14	CODE
LOW	llama_cpp/llama_tokenizer.py	1	CODE
LOW	llama_cpp/llama_chat_format.py	1	CODE
LOW	llama_cpp/__init__.py	1	CODE
LOW	llama_cpp/__init__.py	2	CODE
LOW	llama_cpp/llama_cpp.py	1	CODE
LOW	llama_cpp/llama_cpp.py	16	CODE
LOW	llama_cpp/llama_cpp.py	23	CODE
LOW	llama_cpp/llama_cpp.py	23	CODE
LOW	llama_cpp/llama_cpp.py	23	CODE
LOW	llama_cpp/_internals.py	1	CODE
LOW	llama_cpp/_internals.py	7	CODE
LOW	llama_cpp/_internals.py	22	CODE
LOW	llama_cpp/llama.py	1	CODE
LOW	llama_cpp/llama.py	33	CODE
LOW	llama_cpp/llama.py	35	CODE
LOW	llama_cpp/llama.py	35	CODE
LOW	llama_cpp/llama.py	35	CODE
LOW	llama_cpp/mtmd_cpp.py	1	CODE
LOW	llama_cpp/mtmd_cpp.py	5	CODE
LOW	llama_cpp/mtmd_cpp.py	5	CODE
LOW	llama_cpp/_ctypes_extensions.py	1	CODE
LOW	llama_cpp/llama_cpp_ext.py	7	CODE
LOW	llama_cpp/llava_cpp.py	1	CODE
LOW	llama_cpp/llava_cpp.py	4	CODE
LOW	llama_cpp/server/types.py	1	CODE
LOW	llama_cpp/server/model.py	1	CODE
LOW	llama_cpp/server/cli.py	1	CODE
LOW	llama_cpp/server/settings.py	1	CODE
LOW	llama_cpp/server/app.py	1	CODE
LOW	llama_cpp/server/errors.py	1	CODE
LOW	llama_cpp/server/errors.py	6	CODE
LOW	llama_cpp/server/errors.py	6	CODE
LOW	llama_cpp/server/__main__.py	25	CODE

Docstring Block Structure5 hits · 25 pts

Severity	File	Line	Snippet	Context
HIGH	llama_cpp/llama.py	123	Load a llama.cpp model from `model_path`. Examples: Basic usage >>> import llama_cpp	STRING
HIGH	llama_cpp/llama.py	605	Tokenize a string. Args: text: The utf-8 encoded string to tokenize. add_bos: Whether t	STRING
HIGH	llama_cpp/llama.py	875	Create a generator of tokens from a prompt. Examples: >>> llama = Llama("models/ggml-7b.bin")	STRING
HIGH	llama_cpp/llama.py	1843	Generate text from a prompt. Args: prompt: The prompt to generate text from. suffix: A	STRING
HIGH	llama_cpp/llama.py	1940	Generate text from a prompt. Args: prompt: The prompt to generate text from. suffix: A	STRING

Self-Referential Comments8 hits · 22 pts

Severity	File	Line	Snippet	Context
MEDIUM	docker/open_llama/hug_model.py	44	# Creating a symbolic link from destination to "model.bin"	COMMENT
MEDIUM	docker/open_llama/hug_model.py	74	# Create an argument parser	COMMENT
MEDIUM	docker/open_llama/hug_model.py	92	# Define the parameters	COMMENT
MEDIUM	scripts/releases-to-pep-503.sh	33	# Create an index html file	COMMENT
MEDIUM	scripts/releases-to-pep-503.sh	49	# Create an index html file in llama-cpp-python directory	COMMENT
MEDIUM	llama_cpp/llama_chat_format.py	3160	"""This method is no longer used in the new implementation."""	STRING
MEDIUM	llama_cpp/_internals.py	637	# This method is deprecated in favor of using LlamaSampler directly	COMMENT
MEDIUM	llama_cpp/_internals.py	876	# Create a new wrapper around the cloned sampler	COMMENT

Excessive Try-Catch Wrapping29 hits · 21 pts

Severity	File	Line	Snippet	Context
MEDIUM	docker/open_llama/hug_model.py	135	print("Error - model choice was None")	CODE
MEDIUM	examples/server/server.py	3273	def resolve_model_path(self) -> str:	CODE
LOW	examples/server/server.py	9349	except Exception:	CODE
LOW	examples/server/server.py	10830	except Exception:	CODE
LOW	examples/server/server.py	12487	except Exception:	CODE
LOW	examples/server/server.py	12506	except Exception: # noqa: BLE001	CODE
LOW	examples/server/server.py	13234	except Exception: # noqa: BLE001	CODE
LOW	examples/server/server.py	13308	except Exception: # noqa: BLE001	CODE
LOW	examples/server/server.py	13357	except Exception: # noqa: BLE001	CODE
LOW	examples/server/server.py	16255	except Exception:	CODE
LOW	examples/server/server.py	3194	except Exception:	STRING
LOW	examples/server/server.py	3232	except Exception as exc:	STRING
LOW	examples/server/server.py	3244	except Exception:	STRING
LOW	examples/server/server.py	3303	except Exception as exc:	STRING
LOW	llama_cpp/llama_chat_format.py	717	except Exception as e:	CODE
LOW	llama_cpp/llama_chat_format.py	1010	except Exception as e:	CODE
LOW	llama_cpp/llama_chat_format.py	1737	except Exception as e:	CODE
LOW	llama_cpp/llama_chat_format.py	2094	except Exception as e:	CODE
LOW	llama_cpp/llama_chat_format.py	3085	except Exception as e:	CODE
LOW	llama_cpp/llama_chat_format.py	3620	except Exception as e:	CODE
LOW	llama_cpp/llama_chat_format.py	4315	except Exception as e:	CODE
LOW	llama_cpp/llama_chat_format.py	4434	except Exception as e:	CODE
LOW	llama_cpp/llama.py	494	except Exception as e:	CODE
LOW	llama_cpp/_ctypes_extensions.py	93	except Exception as e:	CODE
LOW	llama_cpp/_ctypes_extensions.py	103	except Exception as e:	CODE
LOW	llama_cpp/llama_grammar.py	33	except Exception as err:	CODE
LOW	llama_cpp/server/errors.py	180	except Exception as exc:	CODE
LOW	llama_cpp/server/errors.py	198	except Exception:	CODE
LOW	llama_cpp/server/__main__.py	80	except Exception as e:	CODE

Over-Commented Block31 hits · 16 pts

Severity	File	Line	Snippet	Context
LOW	examples/server/server.py	1	#!/usr/bin/env -S uv run --script	COMMENT
LOW	llama_cpp/llama_cpp.py	61	# GGML_TYPE_Q4_0 = 2,	COMMENT
LOW	llama_cpp/llama_cpp.py	81	# GGML_TYPE_IQ2_S = 22,	COMMENT
LOW	llama_cpp/llama_cpp.py	201	llama_token_p = ctypes.POINTER(llama_token)	COMMENT
LOW	llama_cpp/llama_cpp.py	241	# LLAMA_VOCAB_PRE_TYPE_MPT = 5,	COMMENT
LOW	llama_cpp/llama_cpp.py	261	# LLAMA_VOCAB_PRE_TYPE_EXAONE = 25,	COMMENT
LOW	llama_cpp/llama_cpp.py	341	LLAMA_TOKEN_TYPE_CONTROL = 3	COMMENT
LOW	llama_cpp/llama_cpp.py	381	# // LLAMA_FTYPE_MOSTLY_Q4_3 = 6, // support has been removed	COMMENT
LOW	llama_cpp/llama_cpp.py	401	# LLAMA_FTYPE_MOSTLY_IQ3_S = 26, // except 1d tensors	COMMENT
LOW	llama_cpp/llama_cpp.py	461	# enum llama_rope_scaling_type {	COMMENT
LOW	llama_cpp/llama_cpp.py	501	# enum llama_flash_attn_type {	COMMENT
LOW	llama_cpp/llama_cpp.py	621	# // (if set to NULL, the token position will be tracked automatically by llama_encode/llama_decode)	COMMENT
LOW	llama_cpp/llama_cpp.py	681	# LLAMA_KV_OVERRIDE_TYPE_STR,	COMMENT
LOW	llama_cpp/llama_cpp.py	801	# struct llama_model_params {	COMMENT
LOW	llama_cpp/llama_cpp.py	821	# llama_progress_callback progress_callback;	COMMENT
LOW	llama_cpp/llama_cpp.py	921	# uint32_t n_rs_seq; // number of recurrent-state snapshots per seq for rollback (0 = no rollback) [EXP	COMMENT
LOW	llama_cpp/llama_cpp.py	941	# float defrag_thold; // defragment the KV cache if holes/size > thold, <= 0 disabled (default)	COMMENT
LOW	llama_cpp/llama_cpp.py	961	# bool swa_full; // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-286834	COMMENT
LOW	llama_cpp/llama_cpp.py	2261	# // the currently loaded vector.	COMMENT
LOW	llama_cpp/llama_cpp.py	2941	# for backwards-compat	COMMENT
LOW	llama_cpp/llama_cpp.py	3021	dest_seq_id: llama_seq_id,	COMMENT
LOW	llama_cpp/llama_cpp.py	3061	# // Allocates a batch of tokens on the heap that can hold a maximum of n_tokens	COMMENT
LOW	llama_cpp/llama_cpp.py	3121	# // For encode-decoder contexts, processes the batch using the decoder.	COMMENT
LOW	llama_cpp/llama_cpp.py	4021	return _llama_vocab_cls(vocab)	COMMENT
LOW	llama_cpp/llama_cpp.py	4081	Returns:	COMMENT
LOW	llama_cpp/llama_cpp.py	4181	remove_special: Allow to remove BOS and EOS tokens if model is configured to do so.	COMMENT
LOW	llama_cpp/llama_cpp.py	4281	# struct llama_sampler_data {	COMMENT
LOW	llama_cpp/llama_cpp.py	4961		COMMENT
LOW	llama_cpp/mtmd_cpp.py	221	# float temp; // sampling temperature, or noise scale for flow-matching decoders	COMMENT
LOW	llama_cpp/mtmd_cpp.py	261	("state_size", c_size_t),	COMMENT
LOW	llama_cpp/mtmd_cpp.py	361	timestamp_interval_ms: int	COMMENT

Cross-Language Confusion2 hits · 12 pts

Severity	File	Line	Snippet	Context
HIGH	tests/test_llama_chat_format.py	75	"pad_token": null,	CODE
HIGH	llama_cpp/llama_cache.py	138	# self.cache.push(_key, side="front") # type: ignore	COMMENT

Redundant / Tautological Comments7 hits · 10 pts

Severity	File	Line	Snippet	Context
LOW	scripts/get-releases.sh	20	# Check if the response is valid JSON	COMMENT
LOW	scripts/releases-to-pep-503.sh	24	# Check if all_releases.txt exists	COMMENT
LOW	.github/workflows/build-wheels-cuda.yaml	272	# Set tag_name to <tag>-cu<cuda_version>.	COMMENT
LOW	llama_cpp/llama_chat_format.py	2816	# Check if vision is supported	COMMENT
LOW	llama_cpp/llama.py	1475	# Check if stop sequence is in the token	COMMENT
LOW	llama_cpp/server/app.py	110	# Check if yaml file	COMMENT
LOW	llama_cpp/server/__main__.py	63	# Check if yaml file	COMMENT

AI Structural Patterns16 hits · 10 pts

Severity	File	Line	Context
LOW	examples/server/server.py	11238	CODE
LOW	llama_cpp/llama_chat_format.py	1460	CODE
LOW	llama_cpp/llama_chat_format.py	1823	CODE
LOW	llama_cpp/llama_chat_format.py	4132	CODE
LOW	llama_cpp/llama_chat_format.py	68	CODE
LOW	llama_cpp/llama_chat_format.py	612	CODE
LOW	llama_cpp/llama_chat_format.py	2850	CODE
LOW	llama_cpp/llama_chat_format.py	3401	CODE
LOW	llama_cpp/llama.py	60	CODE
LOW	llama_cpp/llama.py	704	CODE
LOW	llama_cpp/llama.py	792	CODE
LOW	llama_cpp/llama.py	854	CODE
LOW	llama_cpp/llama.py	1195	CODE
LOW	llama_cpp/llama.py	1815	CODE
LOW	llama_cpp/llama.py	1912	CODE
LOW	llama_cpp/llama.py	2004	CODE

Modern AI Meta-Vocabulary3 hits · 8 pts

Severity	File	Line	Snippet	Context
MEDIUM	README.md	310	# n_ctx=2048, # Uncomment to increase the context window	COMMENT
MEDIUM	README.md	314	max_tokens=32, # Generate up to 32 tokens, set to None to generate up to the end of the context window	CODE
MEDIUM	llama_cpp/llama.py	1353	# Truncate max_tokens if requested tokens would exceed the context window	COMMENT

Structural Annotation Overuse2 hits · 3 pts

Severity	File	Line	Snippet	Context
LOW	llama_cpp/llama_types.py	300	# NOTE: The following type names are not part of the OpenAI OpenAPI specification	COMMENT
LOW	llama_cpp/llama_cpp.py	4193	# /// NOTE: This function does not use a jinja parser. It only support a pre-defined list of template. See more: https:/	COMMENT

AI Slop Vocabulary1 hit · 3 pts

Severity	File	Line	Snippet	Context
MEDIUM	llama_cpp/llama_cpp.py	3036	# // NOTE: this is a helper function to facilitate transition to the new batch API - avoid using it	COMMENT

Modern Structural Boilerplate4 hits · 1 pts

Severity	File	Line	Snippet	Context
LOW	examples/server/server.py	12202	def set_draft_processing_enabled(self, enabled: bool) -> None:	CODE
LOW	examples/server/server.py	1201	def set_target_processing_enabled(self, enabled: bool) -> None:	STRING
LOW	examples/server/server.py	1398	def set_target_processing_enabled(self, enabled: bool) -> None:	STRING
LOW	examples/server/server.py	1461	def _set_nextn_layer_offset(self, offset: int) -> None:	STRING

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings