RyanCodrai/turbovec

18.7

Adjusted Score

18.7

Raw Score

100%

Time Factor

2026-06-10

Last Push

12.7K

Stars

Python

Language

22.1K

Lines of Code

114

Files

436

Pattern Hits

2026-07-14

Scan Date

0.00

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 0MEDIUM 8LOW 428

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 436 distinct pattern matches across 11 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 Identifiers349 hits · 311 pts

Severity	File	Line	Snippet	Context
LOW	turbovec-python/python/turbovec/haystack.py	306	def count_documents_by_filter(self, filters: Dict[str, Any]) -> int:	CODE
LOW	turbovec-python/python/turbovec/haystack.py	316	def count_unique_metadata_by_filter(	CODE
LOW	turbovec-python/python/turbovec/haystack.py	352	def get_metadata_field_min_max(self, metadata_field: str) -> Dict[str, Any]:	CODE
LOW	turbovec-python/python/turbovec/haystack.py	372	def get_metadata_field_unique_values(	CODE
LOW	turbovec-python/python/turbovec/haystack.py	505	async def delete_all_documents_async(self) -> None:	CODE
LOW⚡	turbovec-python/python/turbovec/haystack.py	517	async def count_documents_by_filter_async(self, filters: Dict[str, Any]) -> int:	CODE
LOW⚡	turbovec-python/python/turbovec/haystack.py	522	async def count_unique_metadata_by_filter_async(	CODE
LOW⚡	turbovec-python/python/turbovec/haystack.py	532	async def get_metadata_fields_info_async(self) -> Dict[str, Dict[str, str]]:	CODE
LOW⚡	turbovec-python/python/turbovec/haystack.py	537	async def get_metadata_field_min_max_async(	CODE
LOW⚡	turbovec-python/python/turbovec/haystack.py	545	async def get_metadata_field_unique_values_async(	CODE
LOW⚡	turbovec-python/python/turbovec/haystack.py	555	async def embedding_retrieval_async(	CODE
LOW	turbovec-python/python/turbovec/agno.py	425	def _handles_for_content_hash(self, content_hash: str) -> List[int]:	CODE
LOW	turbovec-python/python/turbovec/agno.py	478	def _resolve_filter_to_handles(	CODE
LOW	turbovec-python/python/turbovec/agno.py	621	def get_supported_search_types(self) -> List[SearchType]:	CODE
LOW	turbovec-python/python/turbovec/langchain.py	92	def _select_relevance_score_fn(self) -> Callable[[float], float]:	CODE
LOW	turbovec-python/python/turbovec/langchain.py	260	def similarity_search_with_score(	CODE
LOW	turbovec-python/python/turbovec/langchain.py	270	async def asimilarity_search_with_score(	CODE
LOW	turbovec-python/python/turbovec/langchain.py	282	def similarity_search_by_vector(	CODE
LOW	turbovec-python/python/turbovec/langchain.py	292	async def asimilarity_search_by_vector(	CODE
LOW	turbovec-python/python/turbovec/langchain.py	377	def max_marginal_relevance_search(	CODE
LOW	turbovec-python/python/turbovec/langchain.py	387	def max_marginal_relevance_search_by_vector(	CODE
LOW	turbovec-python/python/turbovec/langchain.py	399	async def amax_marginal_relevance_search(	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	59	def test_lazy_dim_locked_on_first_add():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	66	def test_from_params_without_dim_is_lazy():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	71	def test_persist_and_load_lazy_uncommitted_store(tmp_path):	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	114	def test_ref_doc_id_preserved_through_query():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	124	def test_empty_query_returns_empty():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	132	def test_k_larger_than_ntotal_is_clamped():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	139	def test_query_without_embedding_raises():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	146	def test_mismatched_dim_raises():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	152	def test_persist_and_from_persist_path_roundtrip(tmp_path):	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	825	def test_delete_nodes_no_args_is_noop():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	832	def test_to_dict_from_dict_roundtrip():	CODE
LOW⚡	turbovec-python/tests/test_llama_index.py	842	def test_to_dict_from_dict_lazy_store():	CODE
LOW	turbovec-python/tests/test_llama_index.py	36	def test_from_params_creates_index():	CODE
LOW	turbovec-python/tests/test_llama_index.py	46	def test_constructor_no_index_is_lazy():	CODE
LOW	turbovec-python/tests/test_llama_index.py	84	def test_add_and_query_returns_nodes():	CODE
LOW	turbovec-python/tests/test_llama_index.py	99	def test_metadata_and_text_roundtrip():	CODE
LOW	turbovec-python/tests/test_llama_index.py	168	def test_from_persist_dir_loads_default_namespace(tmp_path):	CODE
LOW	turbovec-python/tests/test_llama_index.py	184	def test_from_persist_dir_with_custom_namespace(tmp_path):	CODE
LOW	turbovec-python/tests/test_llama_index.py	196	def test_delete_by_ref_doc_id_removes_every_matching_node():	CODE
LOW	turbovec-python/tests/test_llama_index.py	212	def test_delete_by_missing_ref_doc_id_is_noop():	CODE
LOW	turbovec-python/tests/test_llama_index.py	219	def test_delete_nodes_by_node_id():	CODE
LOW	turbovec-python/tests/test_llama_index.py	233	def test_add_upsert_replaces_same_node_id():	CODE
LOW	turbovec-python/tests/test_llama_index.py	245	def test_add_upsert_dim_mismatch_preserves_existing_node():	CODE
LOW	turbovec-python/tests/test_llama_index.py	275	def test_query_with_eq_filter():	CODE
LOW	turbovec-python/tests/test_llama_index.py	288	def test_query_with_in_filter():	CODE
LOW	turbovec-python/tests/test_llama_index.py	305	def test_query_with_and_filter():	CODE
LOW	turbovec-python/tests/test_llama_index.py	323	def test_query_with_or_filter():	CODE
LOW	turbovec-python/tests/test_llama_index.py	342	def test_query_filter_no_matches_returns_empty():	CODE
LOW	turbovec-python/tests/test_llama_index.py	356	def test_query_filter_selective_returns_top_k_from_matches():	CODE
LOW	turbovec-python/tests/test_llama_index.py	376	def test_query_with_node_ids_filter():	CODE
LOW	turbovec-python/tests/test_llama_index.py	391	def test_query_with_doc_ids_filter_matches_ref_doc_id_only():	CODE
LOW	turbovec-python/tests/test_llama_index.py	417	def test_query_with_node_ids_and_filters_intersect():	CODE
LOW	turbovec-python/tests/test_llama_index.py	439	def test_query_ne_filter_treats_missing_key_as_no_match():	CODE
LOW	turbovec-python/tests/test_llama_index.py	460	def test_query_text_match_is_case_sensitive():	CODE
LOW	turbovec-python/tests/test_llama_index.py	497	def test_query_text_match_insensitive_folds_case():	CODE
LOW	turbovec-python/tests/test_llama_index.py	523	def test_query_text_match_raises_type_error_on_non_string_operands():	CODE
LOW	turbovec-python/tests/test_llama_index.py	540	def test_query_all_operator_matches_set_subset():	CODE
LOW	turbovec-python/tests/test_llama_index.py	562	def test_query_any_operator_matches_set_intersection():	CODE
289 more matches not shown…

Over-Commented Block44 hits · 44 pts

Severity	File	Line	Snippet	Context
LOW	turbovec/build.rs	1	// Emit BLAS link directives so downstream binaries (tests, examples, and	COMMENT
LOW	turbovec/tests/tqplus_calibration.rs	1	//! Wave-6 regression tests for the TQ+ calibration state machine.	COMMENT
LOW	turbovec/tests/concurrent_search.rs	1	//! Integration tests for concurrent search on a shared `TurboQuantIndex`.	COMMENT
LOW	turbovec/tests/filtering.rs	1	//! Correctness tests for slot-mask filtering on `TurboQuantIndex` and the	COMMENT
LOW	turbovec/tests/input_validation.rs	1	//! Input-validation tests for `TurboQuantIndex` and `IdMapIndex`.	COMMENT
LOW	turbovec/tests/swap_remove.rs	1	//! Correctness of `TurboQuantIndex::swap_remove`.	COMMENT
LOW	turbovec/tests/distortion.rs	1	//! Statistical validation of quantizer distortion.	COMMENT
LOW	turbovec/tests/kernel_correctness.rs	1	//! Correctness harness for the SIMD scoring kernels.	COMMENT
LOW	turbovec/tests/lazy_init.rs	1	//! Tests for lazy index construction on `TurboQuantIndex` and `IdMapIndex`.	COMMENT
LOW	turbovec/tests/rotation.rs	1	//! Correctness tests for the rotation matrix generator.	COMMENT
LOW	turbovec/examples/dump_state.rs	1	//! Dump the rotation matrix and Lloyd-Max codebook for a given (dim, bits) to	COMMENT
LOW	turbovec/examples/kernel_xtest.rs	1	//! Cross-arch kernel parity smoke-test.	COMMENT
LOW	turbovec/examples/kernel_xtest.rs	61	// Dump per-query top-K as `qi <tab> rank <tab> idx`. Scores are	COMMENT
LOW	turbovec/src/io.rs	1	//! Read/write TurboVec index files.	COMMENT
LOW	turbovec/src/error.rs	1	//! Errors returned by the user-facing add and construct paths.	COMMENT
LOW	turbovec/src/error.rs	21	use std::fmt;	COMMENT
LOW	turbovec/src/error.rs	41	/// `vectors.len()` is not a whole multiple of `dim`.	COMMENT
LOW	turbovec/src/encode.rs	1	//! Encode vectors: normalize, rotate, calibrate, quantize, bit-pack, scale.	COMMENT
LOW	turbovec/src/encode.rs	41	/// floor: at ~200 samples the calibration noise eats the precision gain	COMMENT
LOW	turbovec/src/encode.rs	241	}	COMMENT
LOW	turbovec/src/lib.rs	1	//! TurboQuant implementation for vector search.	COMMENT
LOW	turbovec/src/lib.rs	21	//!	COMMENT
LOW	turbovec/src/lib.rs	61	const ROTATION_SEED: u64 = 42;	COMMENT
LOW	turbovec/src/lib.rs	81		COMMENT
LOW	turbovec/src/lib.rs	121	bit_width: usize,	COMMENT
LOW	turbovec/src/lib.rs	221	rotation: OnceLock::new(),	COMMENT
LOW	turbovec/src/lib.rs	321	// Rotation, boundaries, and centroids remain valid (they only depend	COMMENT
LOW	turbovec/src/lib.rs	381	}	COMMENT
LOW	turbovec/src/lib.rs	401	/// query coordinate is non-finite (NaN, +Inf, -Inf) or has	COMMENT
LOW	turbovec/src/lib.rs	641	// would see `existing = None` (the lazy-first-add signal),	COMMENT
LOW	turbovec/src/search.rs	1	//! SIMD-accelerated search pipeline.	COMMENT
LOW	turbovec/src/search.rs	21	/// Process-global. Tests sample before/after a single search to verify	COMMENT
LOW	turbovec/src/search.rs	381	//	COMMENT
LOW	turbovec/src/search.rs	1261	// is the binding one: `FLUSH_EVERY * max_lut <= 65535` ⇒ max_lut ≤ 255.	COMMENT
LOW	turbovec/src/search.rs	1341	// A pair-level skip short-circuits two 32-vector blocks.	COMMENT
LOW	turbovec/src/id_map.rs	1	//! Stable external IDs on top of [`TurboQuantIndex`].	COMMENT
LOW	turbovec/src/id_map.rs	21	//!	COMMENT
LOW	turbovec/src/id_map.rs	81	/// Returns the same errors as	COMMENT
LOW	turbovec/src/id_map.rs	181		COMMENT
LOW	.cargo/config.toml	1	# Target the x86-64-v3 microarchitecture level on all x86_64 builds.	COMMENT
LOW	turbovec-python/build.rs	1	fn main() {	COMMENT
LOW	turbovec-python/python/turbovec/agno.py	121	self.dimensions: int = embedder.dimensions	COMMENT
LOW	turbovec-python/tests/test_agno.py	1021	db.create()	COMMENT
LOW	turbovec-python/src/lib.rs	241	/// Remove the vector with external id `id`. Returns `True` if it was	COMMENT

Unused Imports19 hits · 16 pts

Severity	File	Line	Context
LOW	turbovec-python/python/turbovec/llama_index.py	6	CODE
LOW	turbovec-python/python/turbovec/haystack.py	16	CODE
LOW	turbovec-python/python/turbovec/_persist.py	16	CODE
LOW	turbovec-python/python/turbovec/agno.py	10	CODE
LOW	turbovec-python/python/turbovec/agno.py	15	CODE
LOW	turbovec-python/python/turbovec/__init__.py	1	CODE
LOW	turbovec-python/python/turbovec/__init__.py	1	CODE
LOW	turbovec-python/python/turbovec/langchain.py	9	CODE
LOW	turbovec-python/python/turbovec/_dedup.py	20	CODE
LOW	turbovec-python/tests/test_llama_index.py	1	CODE
LOW	turbovec-python/tests/test_dedup.py	2	CODE
LOW	turbovec-python/tests/test_haystack.py	2	CODE
LOW	turbovec-python/tests/test_haystack.py	946	CODE
LOW	turbovec-python/tests/test_security.py	9	CODE
LOW	turbovec-python/tests/test_id_map.py	2	CODE
LOW	turbovec-python/tests/test_agno.py	2	CODE
LOW	turbovec-python/tests/test_langchain.py	1	CODE
LOW	turbovec-python/tests/test_index.py	7	CODE
LOW	turbovec-python/tests/test_filtering.py	2	CODE

Decorative Section Separators5 hits · 15 pts

Severity	File	Line	Snippet	Context
MEDIUM	turbovec/examples/kernel_xtest.rs	96	// ─── Tiny deterministic PRNG (xoshiro256** seeded via splitmix64) ────────	COMMENT
MEDIUM	turbovec/src/encode.rs	185	// ─── Norm and scale (aarch64) ────────────────────────────────────────────────	COMMENT
MEDIUM	turbovec/src/encode.rs	227	// ─── Norm and scale (fallback) ───────────────────────────────────────────────	COMMENT
MEDIUM	turbovec/src/encode.rs	243	// ─── Fused quantize + scale + pack (aarch64) ────────────────────────────────	COMMENT
MEDIUM	turbovec/src/encode.rs	355	// ─── Fused quantize + scale + pack (fallback) ───────────────────────────────	COMMENT

Modern Structural Boilerplate8 hits · 8 pts

Severity	File	Line	Snippet	Context
LOW	turbovec-python/python/turbovec/llama_index.py	678	__all__ = ["TurboQuantVectorStore"]	CODE
LOW	turbovec-python/python/turbovec/haystack.py	747	__all__ = ["TurboQuantDocumentStore"]	CODE
LOW	turbovec-python/python/turbovec/_persist.py	55	__all__ = ["check_persisted_handles"]	CODE
LOW	turbovec-python/python/turbovec/agno.py	686	def update_metadata(self, content_id: str, metadata: Dict[str, Any]) -> None:	CODE
LOW	turbovec-python/python/turbovec/agno.py	788	__all__ = ["TurboQuantVectorDb"]	CODE
LOW	turbovec-python/python/turbovec/__init__.py	3	__all__ = ["IdMapIndex", "TurboQuantIndex"]	CODE
LOW	turbovec-python/python/turbovec/langchain.py	558	__all__ = ["TurboQuantVectorStore"]	CODE
LOW	turbovec-python/python/turbovec/_dedup.py	71	__all__ = ["DuplicatePolicy", "resolve_duplicates"]	CODE

Modern AI Meta-Vocabulary2 hits · 8 pts

Severity	File	Line	Snippet	Context
MEDIUM	turbovec-python/tests/test_llama_index.py	895	# store into a LlamaIndex pipeline (RAG, query engine, etc.), so it	COMMENT
MEDIUM⚡	turbovec-python/tests/test_langchain.py	475	# Tier 4: dim is inferred from the embedding model, no explicit param.	COMMENT

AI Slop Vocabulary1 hit · 3 pts

Severity	File	Line	Snippet	Context
MEDIUM	turbovec/tests/kernel_correctness.rs	23	/// which makes the self-query invariant below robust.	COMMENT

Verbosity Indicators2 hits · 3 pts

Severity	File	Line	Snippet	Context
LOW	turbovec/src/pack.rs	26	// Step 1: Extract packed nibble bytes per vector per group	COMMENT
LOW	turbovec/src/pack.rs	57	// Step 2: Pack into platform-specific layout	COMMENT

Structural Annotation Overuse2 hits · 3 pts

Severity	File	Line	Snippet	Context
LOW	turbovec/src/pack.rs	26	// Step 1: Extract packed nibble bytes per vector per group	COMMENT
LOW	turbovec/src/pack.rs	57	// Step 2: Pack into platform-specific layout	COMMENT

Deep Nesting2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	turbovec-python/python/turbovec/haystack.py	336		CODE
LOW	benchmarks/create_diagrams.py	51		CODE

AI Structural Patterns2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	turbovec-python/python/turbovec/agno.py	61		CODE
LOW	turbovec-python/tests/test_llama_index.py	968		CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings