guidance-ai/guidance

19.3

Adjusted Score

19.3

Raw Score

100%

Time Factor

2026-05-21

Last Push

21.7K

Stars

Jupyter Notebook

Language

35.9K

Lines of Code

228

Files

578

Pattern Hits

2026-07-14

Scan Date

0.04

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 9MEDIUM 32LOW 537

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 578 distinct pattern matches across 16 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 Identifiers217 hits · 214 pts

Severity	File	Line	Snippet	Context
LOW	README.md	175	def zero_shot_multiple_choice(	STRING
LOW	guidance/_ast.py	217	def is_allowed_in_lark_terminal(self) -> bool:	CODE
LOW	guidance/_ast.py	225	def is_allowed_in_lark_rule_with_attrs(self) -> bool:	CODE
LOW	guidance/_ast.py	348	def is_allowed_in_lark_terminal(self) -> bool:	CODE
LOW	guidance/_ast.py	352	def is_allowed_in_lark_rule_with_attrs(self) -> bool:	CODE
LOW	guidance/_ast.py	450	def is_allowed_in_lark_terminal(self) -> bool:	CODE
LOW	guidance/_ast.py	489	def is_allowed_in_lark_terminal(self) -> bool:	CODE
LOW	guidance/_ast.py	519	def is_allowed_in_lark_terminal(self) -> bool:	CODE
LOW	guidance/_ast.py	533	def is_allowed_in_lark_terminal(self) -> bool:	CODE
LOW	guidance/_ast.py	537	def is_allowed_in_lark_rule_with_attrs(self) -> bool:	CODE
LOW	guidance/_uri_validation.py	66	def _validate_hostname_not_private(hostname: str, original_url: str) -> None:	CODE
LOW⚡	guidance/chat.py	83	def _template_class_from_string(template_str):	CODE
LOW	guidance/_utils.py	128	def strip_multiline_string_indents(f):	CODE
LOW	guidance/_utils.py	423	def apply_top_k_and_top_p_filter(logits: np.ndarray, sampling_params: Optional["SamplingParams"]) -> np.ndarray:	CODE
LOW	guidance/models/_mock.py	94	def get_next_token_with_top_k(	CODE
LOW	guidance/models/_openai_base.py	217	def streaming_chat_completions(	CODE
LOW	guidance/models/_openai_base.py	232	def streaming_chat_completions(	CODE
LOW	guidance/models/_azureai.py	76	def create_azure_openai_model(	CODE
LOW	guidance/models/_azureai.py	162	def streaming_chat_completions(	CODE
LOW	guidance/models/_azureai.py	220	def create_azure_aifoundry_model(	CODE
LOW	guidance/models/_transformers.py	179	def _byte_tokens_from_byte_decoder(	CODE
LOW	guidance/models/_transformers.py	194	def _byte_tokens_from_sp_model(	CODE
LOW	guidance/models/_transformers.py	217	def _byte_tokens_by_encoding_token_strings(	CODE
LOW	guidance/models/_transformers.py	278	def check_byte_decoder_has_all_bytes() -> None:	CODE
LOW	guidance/models/_transformers.py	287	def check_byte_decoder_complex_round_trip() -> None:	CODE
LOW	guidance/models/experimental/_litellm.py	46	def streaming_chat_completions(	CODE
LOW	guidance/models/_engine/_engine.py	542	def get_next_token_with_top_k(	CODE
LOW	guidance/models/_engine/_engine.py	659	def chat_completion_streaming(	CODE
LOW	guidance/models/_engine/_engine.py	732	def apply_temp_and_sampling_params(	CODE
LOW	guidance/models/_base/_model.py	192	def _update_open_block_captures(self) -> Self:	CODE
LOW	guidance/_bg/__init__.py	20	def _asyncio_background_thread() -> tuple[threading.Thread, AbstractEventLoop]:	CODE
LOW	tests/tokenizer_common.py	37	def base_eos_bos_token_round_trip(self, model_name: str):	CODE
LOW	tests/utils.py	114	def check_match_success_with_guards(grammar, test_string: str):	CODE
LOW	tests/utils.py	194	def check_run_with_temperature(lm: models.Model, desired_temperature: float):	CODE
LOW	tests/unit/test_parser.py	17	def test_zero_or_more_and_one_or_more():	CODE
LOW	tests/unit/test_parser.py	34	def test_zero_or_more_and_one_or_more_mixed():	CODE
LOW	tests/unit/test_parser.py	117	def test_char_set_one_or_more():	CODE
LOW	tests/unit/test_model.py	58	def test_step_every_k_injection():	CODE
LOW	tests/unit/test_model.py	84	def test_step_stop_token_trigger_injection():	CODE
LOW	tests/unit/test_grammar.py	24	def test_select_ambiguous_lexeme_boundary():	CODE
LOW	tests/unit/test_grammar.py	33	def test_select_ambiguous_lexeme_boundary_manual_fix():	STRING
LOW	tests/unit/test_grammar.py	50	def test_grammar_plus_fstring():	CODE
LOW	tests/unit/test_grammar.py	82	def test_multiple_mutual_recursion(self):	CODE
LOW	tests/unit/test_grammar.py	99	def test_branching_mutual_recursion(self):	CODE
LOW	tests/unit/test_grammar.py	140	def test_raises_on_incomplete_input(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	11	def test_https_allowed_by_default(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	16	def test_http_blocked_by_default(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	20	def test_ftp_blocked_by_default(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	24	def test_http_allowed_when_configured(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	28	def test_custom_scheme_allowed(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	32	def test_scheme_check_is_case_insensitive(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	38	def test_file_uri_blocked_when_allow_local_false(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	42	def test_file_uri_allowed_when_allow_local_true(self):	CODE
LOW⚡	tests/unit/test_uri_validation.py	46	def test_file_uri_not_subject_to_scheme_allowlist(self):	CODE
LOW	tests/unit/test_uri_validation.py	105	def test_ip_literal_in_url_blocked(self):	CODE
LOW	tests/unit/test_uri_validation.py	118	def test_dns_resolution_failure_raises(self):	CODE
LOW	tests/unit/test_uri_validation.py	129	def test_all_resolved_addresses_checked(self):	CODE
LOW⚡	tests/unit/test_bytes_from.py	12	def test_blocks_http_by_default(self):	CODE
LOW⚡	tests/unit/test_bytes_from.py	16	def test_allows_http_when_configured(self):	CODE
LOW⚡	tests/unit/test_bytes_from.py	24	def test_blocks_private_ip_by_default(self):	CODE
157 more matches not shown…

Unused Imports119 hits · 114 pts

Severity	File	Line	Context
LOW	guidance/_tools.py	11	CODE
LOW	guidance/__init__.py	8	CODE
LOW	guidance/__init__.py	11	CODE
LOW	guidance/_ast.py	31	CODE
LOW	guidance/_ast.py	31	CODE
LOW	guidance/_parser.py	13	CODE
LOW	guidance/_utils.py	21	CODE
LOW	guidance/metrics/__init__.py	3	CODE
LOW	guidance/metrics/__init__.py	3	CODE
LOW	guidance/metrics/__init__.py	3	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/trace/__init__.py	9	CODE
LOW	guidance/library/__init__.py	3	CODE
LOW	guidance/library/__init__.py	3	CODE
LOW	guidance/library/__init__.py	3	CODE
LOW	guidance/library/__init__.py	3	CODE
LOW	guidance/library/__init__.py	3	CODE
LOW	guidance/library/__init__.py	4	CODE
LOW	guidance/library/__init__.py	4	CODE
LOW	guidance/library/__init__.py	7	CODE
LOW	guidance/library/__init__.py	8	CODE
LOW	guidance/library/__init__.py	9	CODE
LOW	guidance/library/__init__.py	9	CODE
LOW	guidance/library/__init__.py	10	CODE
LOW	guidance/library/__init__.py	10	CODE
LOW	guidance/library/__init__.py	11	CODE
LOW	guidance/library/__init__.py	11	CODE
LOW	guidance/library/__init__.py	12	CODE
LOW	guidance/library/__init__.py	13	CODE
LOW	guidance/library/__init__.py	14	CODE
LOW	guidance/library/__init__.py	14	CODE
LOW	guidance/library/__init__.py	14	CODE
LOW	guidance/library/__init__.py	14	CODE
LOW	guidance/library/__init__.py	17	CODE
LOW	guidance/library/__init__.py	17	CODE
LOW	guidance/library/__init__.py	17	CODE
LOW	guidance/library/__init__.py	17	CODE
LOW	guidance/library/__init__.py	17	CODE
LOW	guidance/library/__init__.py	18	CODE
LOW	guidance/library/__init__.py	19	CODE
LOW	guidance/library/__init__.py	19	CODE
LOW	guidance/library/_subgrammar.py	2	CODE
59 more matches not shown…

Over-Commented Block94 hits · 93 pts

Severity	File	Line	Snippet	Context
LOW	guidance/models/broken_models/_lite_llm.py	1	# import tiktoken	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	21	# "Please install the litellm package version >= 1.7 using `pip install litellm -U` in order to use guid	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	41	# )	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	61		COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	81		COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	101	# n=1,	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	121	# else:	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	141	# raise Exception(	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	161	# except Exception as e: # TODO: add retry logic	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	181	# messages = []	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	201	# pos += end_pos + len(role_end)	COMMENT
LOW	guidance/models/broken_models/_lite_llm.py	221	# except Exception as e: # TODO: add retry logic	COMMENT
LOW	guidance/models/broken_models/_cohere.py	1	# from ._lite_llm import LiteLLMEngine, LiteLLM, LiteLLMCompletion, LiteLLMInstruct	COMMENT
LOW	guidance/models/broken_models/_cohere.py	21		COMMENT
LOW	guidance/models/broken_models/_cohere.py	41	# class CohereCompletion(Cohere, LiteLLMCompletion):	COMMENT
LOW	guidance/models/broken_models/_azure_openai.py	1	# import pathlib	COMMENT
LOW	guidance/models/broken_models/_azure_openai.py	21	# """Represents an Azure OpenAI model as exposed through their remote API.	COMMENT
LOW	guidance/models/broken_models/_azure_openai.py	41	# max_streaming_tokens: int = 1000,	COMMENT
LOW	guidance/models/broken_models/_azure_openai.py	61	# if not is_openai or not hasattr(openai_package, "OpenAI"):	COMMENT
LOW	guidance/models/broken_models/_azure_openai.py	81		COMMENT
LOW	guidance/models/broken_models/_azure_openai.py	101	# engine_instance,	COMMENT
LOW	guidance/models/broken_models/_googleai.py	1	# import re	COMMENT
LOW	guidance/models/broken_models/_googleai.py	21	# model,	COMMENT
LOW	guidance/models/broken_models/_googleai.py	41	# genai.configure(api_key=api_key)	COMMENT
LOW	guidance/models/broken_models/_googleai.py	61	# timeout=0.5,	COMMENT
LOW	guidance/models/broken_models/_googleai.py	81		COMMENT
LOW	guidance/models/broken_models/_googleai.py	101	# engine=engine_map[self.__class__](	COMMENT
LOW	guidance/models/broken_models/_googleai.py	121	# assistant_start = b"<\|im_start\|>assistant\n"	COMMENT
LOW	guidance/models/broken_models/_googleai.py	141	# end_pos = prompt[pos:].find(role_end)	COMMENT
LOW	guidance/models/broken_models/_googleai.py	161	# )	COMMENT
LOW	guidance/models/broken_models/_googleai.py	181		COMMENT
LOW	guidance/models/broken_models/_googleai.py	201	# formated_messages = []	COMMENT
LOW	guidance/models/broken_models/_googleai.py	221	# formated_messages.append(Content(role=m["role"], parts=parts))	COMMENT
LOW	guidance/models/broken_models/_googleai.py	241	# class GoogleAIChat(GoogleAI, Chat):	COMMENT
LOW	guidance/models/broken_models/_Gemini.py	1	# import os	COMMENT
LOW	guidance/models/broken_models/_Gemini.py	21		COMMENT
LOW	guidance/models/broken_models/_Gemini.py	41	# is_vertexai = True	COMMENT
LOW	guidance/models/broken_models/_Gemini.py	61	# # caching=caching,	COMMENT
LOW	guidance/models/broken_models/_Gemini.py	81	# # tokenizer=tokenizer,	COMMENT
LOW	guidance/models/broken_models/_Gemini.py	101	# # the superclass does all the work	COMMENT
LOW	guidance/models/broken_models/_Gemini.py	121	# # last_user_text = messages[-1]["content"]	COMMENT
LOW	guidance/models/broken_models/_Gemini.py	141		COMMENT
LOW	guidance/models/broken_models/_anthropic.py	1	# import os	COMMENT
LOW	guidance/models/broken_models/_anthropic.py	21	# except ModuleNotFoundError:	COMMENT
LOW	guidance/models/broken_models/_anthropic.py	41	# except:	COMMENT
LOW	guidance/models/broken_models/_anthropic.py	61	# ("assistant", b"<\|im_start\|>assistant\n"),	COMMENT
LOW	guidance/models/broken_models/_anthropic.py	81		COMMENT
LOW	guidance/models/broken_models/_anthropic.py	101	# temperature=temperature,	COMMENT
LOW	guidance/models/broken_models/_anthropic.py	121	# there are some things we cannot do, like force the model to follow a pattern inside	COMMENT
LOW	guidance/models/broken_models/_anthropic.py	141	# tokenizer=tokenizer,	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	1	# import hashlib	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	21	# class AzureAIStudioChatEngine(GrammarlessEngine):	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	41	# "Detected OpenAI compatible model; please install openai package"	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	61		COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	81	# found = True	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	101	# message_content = btext.decode("utf8")	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	121	# cache_key = self._hash_prompt(prompt)	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	141		COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	161	# response_score = requests.post(	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	181	# yield encoded_chunk	COMMENT
34 more matches not shown…

Decorative Section Separators22 hits · 69 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	guidance/chat.py	91	# --------------------------------------------------	COMMENT
MEDIUM⚡	guidance/chat.py	93	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	111	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	113	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	146	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	148	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	173	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	175	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	230	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	232	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	251	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	253	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	285	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	287	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	317	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	319	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	345	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	347	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	371	# --------------------------------------------------	COMMENT
MEDIUM	guidance/chat.py	373	# --------------------------------------------------	COMMENT
MEDIUM	guidance/models/broken_models/_azure_openai.py	49	# ----------	COMMENT
MEDIUM	guidance/models/broken_models/_azureai_studio.py	214	# ----------	COMMENT

Self-Referential Comments10 hits · 34 pts

Severity	File	Line	Snippet	Context
MEDIUM	guidance/models/_openai_base.py	484	# Create an in-memory WAV file	COMMENT
MEDIUM	guidance/models/_transformers.py	405	# Create the tokenizer	STRING
MEDIUM	guidance/models/_base/_model.py	368	# Define the target function for the thread	COMMENT
MEDIUM⚡	tests/unit/test_decorator.py	407	# Create a weak reference to the object	COMMENT
MEDIUM⚡	tests/unit/test_decorator.py	420	# Create a weak reference to the object	COMMENT
MEDIUM⚡	tests/unit/test_decorator.py	433	# Create a weak reference to the object	COMMENT
MEDIUM⚡	tests/unit/test_decorator.py	435	# Create a weak reference to the cached method	COMMENT
MEDIUM	docs/conf.py	7	# This file is execfile()d with the current directory set to its	COMMENT
MEDIUM	scripts/extract_python_from_readme.py	18	# This function contains 95% CoPilot vibes by volume	COMMENT
MEDIUM	packages/python/stitch/docs/source/conf.py	6	# This file is execfile()d with the current directory set to its	COMMENT

Deep Nesting32 hits · 32 pts

Severity	File	Line	Context
LOW	guidance/_schema.py	184	CODE
LOW	guidance/_grammar.py	61	CODE
LOW	guidance/_ast.py	682	CODE
LOW	guidance/_parser.py	324	CODE
LOW	guidance/_guidance.py	129	CODE
LOW	guidance/_guidance.py	140	CODE
LOW	guidance/metrics/_metrics.py	126	CODE
LOW	guidance/trace/_trace.py	357	CODE
LOW	guidance/library/_json.py	13	CODE
LOW	guidance/library/_substring.py	11	CODE
LOW	guidance/models/_mock.py	117	CODE
LOW	guidance/models/_openai_base.py	168	CODE
LOW	guidance/models/_openai_base.py	336	CODE
LOW	guidance/models/_llama_cpp.py	81	CODE
LOW	guidance/models/_transformers.py	217	CODE
LOW	guidance/models/_transformers.py	444	CODE
LOW	guidance/models/broken_models/_vertexai.py	21	CODE
LOW	guidance/models/broken_models/_vertexai.py	174	CODE
LOW	guidance/models/experimental/_sglang.py	99	CODE
LOW	guidance/models/experimental/_litellm.py	209	CODE
LOW	guidance/models/experimental/_vllm.py	29	CODE
LOW	guidance/models/_engine/_engine.py	94	CODE
LOW	guidance/models/_engine/_interpreter.py	65	CODE
LOW	guidance/models/_base/_model.py	115	CODE
LOW	guidance/models/_base/_model.py	161	CODE
LOW	guidance/models/_base/_model.py	361	CODE
LOW	guidance/visual/_renderer.py	199	CODE
LOW	guidance/visual/_renderer.py	329	CODE
LOW	guidance/visual/_renderer.py	524	CODE
LOW	guidance/visual/_renderer.py	541	CODE
LOW	guidance/visual/_trace.py	17	CODE
LOW	tests/unit/test_ll.py	62	CODE

Cross-File Repetition6 hits · 30 pts

Severity	File	Snippet	Context
HIGH	guidance/models/_llama_cpp.py	computes the logits for the given token state. this overrides a method from the localengine class that is used to get in	STRING
HIGH	guidance/models/_transformers.py	computes the logits for the given token state. this overrides a method from the localengine class that is used to get in	STRING
HIGH	guidance/models/_onnxruntime.py	computes the logits for the given token state. this overrides a method from the localengine class that is used to get in	STRING
HIGH	tests/model_specific/test_onnxruntime_genai.py	tweak this proverb to apply to model instructions instead. {gen("verse", max_tokens=2)}	STRING
HIGH	tests/model_specific/test_transformers.py	tweak this proverb to apply to model instructions instead. {gen("verse", max_tokens=2)}	STRING
HIGH	tests/model_specific/llama_cpp_tests/test_llama_cpp.py	tweak this proverb to apply to model instructions instead. {gen("verse", max_tokens=2)}	STRING

Redundant / Tautological Comments18 hits · 29 pts

Severity	File	Line	Snippet	Context
LOW	guidance/debug.py	27	# Check if the auto renderer has a JupyterWidgetRenderer inside	COMMENT
LOW	guidance/debug.py	53	# Check if the auto renderer has a JupyterWidgetRenderer inside	COMMENT
LOW	guidance/debug.py	81	# Check if the auto renderer has a JupyterWidgetRenderer inside	COMMENT
LOW	guidance/models/_transformers.py	299	# Check if the tokenizer has a bos_token attribute, and if it does, check	COMMENT
LOW	guidance/models/broken_models/_azureai_studio.py	123	# # Check if the result is already in the cache	COMMENT
LOW	guidance/models/_engine/_engine.py	315	# Check if the accumulated text ends with any stop string	COMMENT
LOW	guidance/models/_engine/_engine.py	357	# Check if we've accumulated enough to match the stop string	COMMENT
LOW	guidance/models/_engine/_engine.py	426	# Set flag to indicate this is an injection backtrack	COMMENT
LOW	guidance/models/_engine/_engine.py	474	# Set flag to indicate this is an injection backtrack	COMMENT
LOW	guidance/models/_engine/_interpreter.py	87	# Check if this is an injection backtrack (should happen before adding text)	COMMENT
LOW	guidance/models/_base/_model.py	181	# Set start_index to the current length	COMMENT
LOW	guidance/models/_base/_model.py	392	# Check if the thread is still alive	COMMENT
LOW	guidance/visual/_renderer.py	482	# Check if message has diverged from prev messages	COMMENT
LOW	guidance/visual/_trace.py	32	# Check if any input is a role opener or closer	COMMENT
LOW	guidance/visual/_trace.py	93	# Check if any input in active role is a RoleCloserInput	COMMENT
LOW⚡	tests/unit/test_decorator.py	415	# Check if the object was garbage collected	COMMENT
LOW⚡	tests/unit/test_decorator.py	428	# Check if the object was garbage collected	COMMENT
LOW⚡	tests/unit/test_decorator.py	444	# Check if the object was garbage collected	COMMENT

Modern Structural Boilerplate29 hits · 29 pts

Severity	File	Line	Snippet	Context
LOW	guidance/_topics.py	19	__all__ = [	CODE
LOW	guidance/__init__.py	13	__all__ = [	CODE
LOW	guidance/_ast.py	512	def set_target(self, target: RuleNode) -> None:	CODE
LOW	guidance/debug.py	8	logger = logging.getLogger(__name__)	CODE
LOW	guidance/_utils.py	23	logger = logging.getLogger(__name__)	CODE
LOW	guidance/metrics/_metrics.py	16	logger = logging.getLogger(__name__)	CODE
LOW	guidance/metrics/__init__.py	5	__all__ = [	CODE
LOW	guidance/trace/__init__.py	32	__all__ = [	CODE
LOW	guidance/trace/_trace.py	12	logger = logging.getLogger(__name__)	CODE
LOW	guidance/library/__init__.py	21	__all__ = [	CODE
LOW	guidance/library/_subgrammar.py	4	__all__ = ["as_regular_grammar", "lexeme", "regex", "subgrammar"]	CODE
LOW	guidance/library/_gen.py	10	logger = logging.getLogger(__name__)	CODE
LOW	guidance/models/_mock.py	11	logger = logging.getLogger(__name__)	CODE
LOW	guidance/models/_llama_cpp.py	30	logger = logging.getLogger(__name__)	CODE
LOW	guidance/models/__init__.py	10	__all__ = [	CODE
LOW	guidance/models/_azureai.py	27	logger = logging.getLogger(__name__)	CODE
LOW	guidance/models/experimental/__init__.py	5	__all__ = ["LiteLLM", "SglangModel", "VLLMModel"]	CODE
LOW	guidance/models/_engine/__init__.py	6	__all__ = [	CODE
LOW	guidance/models/_engine/_engine.py	27	logger = logging.getLogger(__name__)	CODE
LOW	guidance/models/_base/__init__.py	5	__all__ = [	CODE
LOW	guidance/visual/_exchange.py	11	logger = logging.getLogger(__name__)	CODE
LOW	guidance/visual/_exchange.py	70	__all__ = ["TopicExchange"]	CODE
LOW	guidance/visual/__init__.py	23	__all__ = [	CODE
LOW	guidance/visual/_renderer.py	53	logger = logging.getLogger(__name__)	CODE
LOW	guidance/visual/_jupyter.py	12	logger = logging.getLogger(__name__)	CODE
LOW	guidance/registry/__init__.py	5	__all__ = [	CODE
LOW	guidance/registry/_registry.py	82	def set_renderer(renderer: Renderer) -> None:	CODE
LOW	packages/python/stitch/docs/source/conf.py	203	logger = logging.getLogger(__name__)	CODE
LOW	packages/python/stitch/stitch/__init__.py	10	__all__ = ["StitchWidget", "__version__", "_jupyter_labextension_paths", "_jupyter_nbextension_paths", "version_info"]	CODE

Cross-Language Confusion3 hits · 20 pts

Severity	File	Line	Snippet	Context
HIGH	tests/unit/library/json/test_allOf.py	352	# valid: foo is null, bar is equal to 5, baz is null	COMMENT
HIGH	tests/unit/library/json/test_allOf.py	443	# valid: foo is null, bar is equal to 5, baz is null	COMMENT
HIGH	packages/python/stitch/stitch/tests/conftest.py	43	_widget_attrs["_comm_default"] = getattr(Widget, "_comm_default", undefined)	CODE

Excessive Try-Catch Wrapping10 hits · 10 pts

Severity	File	Line	Snippet	Context
LOW	guidance/metrics/_metrics.py	93	except Exception as e: # noqa BLE001	CODE
LOW	guidance/metrics/_metrics.py	141	except Exception as e: # noqa BLE001	CODE
LOW	guidance/metrics/_metrics.py	177	except Exception as e:	CODE
LOW	guidance/models/_transformers.py	310	except Exception as e:	CODE
LOW	guidance/models/broken_models/_vertexai.py	132	except Exception as e: # TODO: add retry logic	CODE
LOW	guidance/visual/_renderer.py	153	except Exception: # noqa: BLE001	CODE
LOW	guidance/visual/_renderer.py	195	except Exception as _: # noqa: BLE001	CODE
LOW	guidance/visual/_renderer.py	237	except Exception as _: # noqa: BLE001	CODE
LOW	tests/unit/test_visual.py	82	except Exception as e:	CODE
LOW	tests/model_specific/test_visual.py	51	except Exception as e:	CODE

AI Structural Patterns9 hits · 9 pts

Severity	File	Line	Context
LOW	guidance/_grammar.py	27	CODE
LOW	guidance/_utils.py	314	CODE
LOW	guidance/_utils.py	331	CODE
LOW	guidance/library/_json.py	13	CODE
LOW	guidance/library/_gen.py	13	CODE
LOW	guidance/models/_azureai.py	76	CODE
LOW	guidance/models/_azureai.py	276	CODE
LOW	guidance/models/_transformers.py	582	CODE
LOW	guidance/models/_onnxruntime.py	31	CODE

Fake / Example Data4 hits · 4 pts

Severity	File	Line	Snippet	Context
LOW	guidance/resources/graphpaper-inline.html	29	/*! @license DOMPurify 3.2.6 \| (c) Cure53 and other contributors \| Released under the Apache license 2.0 and Mozilla Pub	COMMENT
LOW	tests/model_specific/test_transformers.py	35	big_opts = ["Lorem ipsum dolor sit amet", "Duis aute irure dolor "]	CODE
LOW	tests/model_specific/test_transformers.py	35	big_opts = ["Lorem ipsum dolor sit amet", "Duis aute irure dolor "]	CODE
LOW	tests/model_integration/library/test_subgrammar.py	62	lm += "John Doe's name, age, and birthday:\n"	CODE

AI Slop Vocabulary2 hits · 3 pts

Severity	File	Line	Snippet	Context
LOW	guidance/models/_transformers.py	127	# (that will just add extra spaces during an encode-decode cycle)	COMMENT
LOW	guidance/models/_engine/_engine.py	614	# If we have top_k, we can just return the first one	COMMENT

Slop Phrases1 hit · 2 pts

Severity	File	Line	Snippet	Context
LOW	guidance/models/broken_models/_lite_llm.py	139	# # make sure you don't try and instruct the same model twice	COMMENT

Overly Generic Function Names2 hits · 2 pts

Severity	File	Line	Snippet	Context
LOW	tests/model_integration/test_model.py	41	def my_function(lm):	STRING
LOW	tests/model_integration/test_model.py	77	def my_function(lm):	CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings