hexo-ai/sia

23.3

Adjusted Score

23.3

Raw Score

100%

Time Factor

2026-07-02

Last Push

2.0K

Stars

Python

Language

19.0K

Lines of Code

114

Files

270

Pattern Hits

2026-07-14

Scan Date

0.01

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 1MEDIUM 55LOW 214

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 270 distinct pattern matches across 13 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.

Decorative Section Separators38 hits · 135 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	sia/orchestrator.py	846	# ========================	COMMENT
MEDIUM⚡	sia/orchestrator.py	848	# ========================	COMMENT
MEDIUM⚡	sia/orchestrator.py	852	# ========================	COMMENT
MEDIUM⚡	sia/orchestrator.py	854	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	86	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	88	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	869	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	871	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	895	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	897	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	915	# ========================	COMMENT
MEDIUM	sia/orchestrator.py	917	# ========================	COMMENT
MEDIUM	sia/tasks/_shared/reference_target_agent.py	11	# ── Tool definitions ──────────────────────────────────────────────────────────	COMMENT
MEDIUM⚡	sia/tasks/_shared/reference_target_agent.py	50	# ── Tool implementations ──────────────────────────────────────────────────────	COMMENT
MEDIUM	sia/tasks/_shared/reference_target_agent.py	101	# ── Multi-Trajectory Logger ───────────────────────────────────────────────────	COMMENT
MEDIUM	sia/tasks/_shared/reference_target_agent.py	173	# ── Agent loop ────────────────────────────────────────────────────────────────	COMMENT
MEDIUM	sia/tasks/_shared/reference_target_agent.py	225	# ── Entry point ───────────────────────────────────────────────────────────────	COMMENT
MEDIUM	sia/tasks/_shared/reference_target_agent.py	234	# ── Multi-Trajectory Usage Example ────────────────────────────────────────────	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	11	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	13	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	27	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	29	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	37	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	39	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	44	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	46	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	58	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	60	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	109	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	111	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	181	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	sia/tasks/gpqa/reference/reference_target_agent.py	183	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	228	# -----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	230	# -----------------------------------------------------------------------------	COMMENT
MEDIUM	…/spaceship-titanic/reference/reference_target_agent.py	11	# ── Tool definitions ──────────────────────────────────────────────────────────	COMMENT
MEDIUM⚡	…/spaceship-titanic/reference/reference_target_agent.py	50	# ── Tool implementations ──────────────────────────────────────────────────────	COMMENT
MEDIUM	…/spaceship-titanic/reference/reference_target_agent.py	100	# ── Agent loop ────────────────────────────────────────────────────────────────	COMMENT
MEDIUM	…/spaceship-titanic/reference/reference_target_agent.py	152	# ── Entry point ───────────────────────────────────────────────────────────────	COMMENT

Hyper-Verbose Identifiers108 hits · 132 pts

Severity	File	Line	Snippet	Context
LOW	sia/context_manager.py	183	def _format_metrics_comparison(self, prev_metrics: dict[str, Any], current_metrics: dict[str, Any]) -> str:	STRING
LOW	sia/prompts.py	39	def _build_weights_meta_prompt(	CODE
LOW	sia/prompts.py	713	def build_target_client_setup(provider: Provider, task_model: str) -> str:	STRING
LOW	sia/prepare_mlebench_dataset.py	112	def get_similar_tasks_from_gemini(competition_id: str, task_description: str) -> str:	CODE
LOW	sia/prepare_mlebench_dataset.py	175	def create_sample_task_descriptions(competition_id: str, tasks_dir: Path, similar_tasks: str) -> bool:	CODE
LOW	sia/orchestrator.py	312	def _run_target_agent_sandboxed(	STRING
LOW	sia/profiles.py	95	def load_target_agent_profile(name_or_path: str) -> TargetAgentProfile:	CODE
LOW	…asks/longcot-chess/reference/reference_target_agent.py	32	def load_chess_hard_questions(dataset_dir: Path) -> List[Dict[str, Any]]:	CODE
LOW	tests/test_orchestrator_helpers.py	8	def test_load_single_trajectory(tmp_path):	CODE
LOW	tests/test_orchestrator_helpers.py	18	def test_load_multi_trajectory(tmp_path):	CODE
LOW	tests/test_orchestrator_helpers.py	32	def test_load_missing_execution(tmp_path):	CODE
LOW	tests/test_orchestrator_helpers.py	45	def test_load_empty_multi_trajectory_folder(tmp_path):	CODE
LOW⚡	tests/test_agent_impls.py	10	def test_registry_lists_builtin_agent_impls():	CODE
LOW⚡	tests/test_agent_impls.py	14	def test_get_agent_impl_returns_callable():	CODE
LOW⚡	tests/test_agent_impls.py	19	def test_get_agent_impl_unknown_raises():	CODE
LOW⚡	tests/test_agent_impls.py	24	def test_util_reexports_registry_run_agent():	CODE
LOW⚡	tests/test_agent_impls.py	31	def test_pydantic_ai_impl_runs_with_test_model(tmp_path):	CODE
LOW	tests/test_agent_impls.py	50	def test_pydantic_ai_model_passthrough():	CODE
LOW	tests/test_agent_impls.py	60	def test_openhands_model_gets_openai_prefix_for_compatible_provider():	CODE
LOW	tests/test_agent_impls.py	71	def test_openhands_model_passthrough_without_compatible_provider():	CODE
LOW	tests/test_agent_impls.py	81	def test_run_agent_threads_provider_to_agent_impl():	CODE
LOW	tests/test_generation_loop.py	66	def test_run_target_agent_success(mock_popen_cls, tmp_path):	CODE
LOW	tests/test_generation_loop.py	98	def test_run_target_agent_failure(mock_popen_cls, tmp_path):	CODE
LOW	tests/test_generation_loop.py	126	def test_single_generation_creates_context(mock_run_ta, mock_run_fb, tmp_path):	CODE
LOW	tests/test_generation_loop.py	165	def test_run_generation_directory_structure(mock_run_ta, mock_run_fb, tmp_path):	CODE
LOW	tests/test_generation_loop.py	194	def test_two_generations_with_feedback(mock_run_ta, mock_run_fb, mock_llm, tmp_path):	CODE
LOW⚡	tests/test_providers.py	10	def test_bundled_providers_present():	CODE
LOW⚡	tests/test_providers.py	14	def test_load_anthropic_provider():	CODE
LOW⚡	tests/test_providers.py	23	def test_load_nebius_provider():	CODE
LOW⚡	tests/test_providers.py	30	def test_unknown_provider_name_raises():	CODE
LOW⚡	tests/test_providers.py	35	def test_load_provider_from_path(tmp_path):	CODE
LOW	tests/test_providers.py	58	def test_invalid_client_kind_raises(tmp_path):	CODE
LOW	tests/test_providers.py	65	def test_user_dir_overrides_bundled(tmp_path, monkeypatch):	CODE
LOW	tests/test_feedback_context_golden.py	19	def test_success_single_with_results(tmp_path):	CODE
LOW	tests/test_feedback_context_golden.py	40	def test_failure_single_no_results(tmp_path):	CODE
LOW	tests/test_feedback_context_golden.py	60	def test_success_multi_with_results(tmp_path):	CODE
LOW	tests/test_prompts_snapshot.py	31	def test_meta_prompt_anthropic_provider_is_byte_identical():	CODE
LOW	tests/test_prompts_snapshot.py	42	def test_meta_prompt_openai_provider_golden():	CODE
LOW	tests/test_prompts_snapshot.py	53	def test_feedback_prompt_golden():	CODE
LOW	tests/test_web.py	79	def test_get_run_detail_and_domains(runs_root):	CODE
LOW⚡	tests/test_web.py	95	def test_eval_details_and_artifacts(runs_root):	CODE
LOW⚡	tests/test_web.py	103	def test_trajectory_normalization(runs_root):	CODE
LOW⚡	tests/test_web.py	112	def test_missing_lookups_return_none(runs_root):	CODE
LOW⚡	tests/test_web.py	119	def test_path_traversal_is_blocked(runs_root, evil):	CODE
LOW⚡	tests/test_cli_interface.py	11	def test_top_level_help_lists_subcommands():	CODE
LOW⚡	tests/test_cli_interface.py	19	def test_run_help_exposes_orchestrator_flags():	CODE
LOW⚡	tests/test_cli_interface.py	27	def test_web_help_exposes_server_flags():	CODE
LOW⚡	tests/test_cli_interface.py	35	def test_no_args_exits_nonzero():	CODE
LOW⚡	tests/test_cli_interface.py	40	def test_default_subcommand_is_run():	CODE
LOW⚡	tests/test_cli_interface.py	45	def test_invalid_task_exits_nonzero():	CODE
LOW	tests/test_context_manager.py	37	def test_initialize_creates_context_md(context_mgr, run_dir):	CODE
LOW	tests/test_context_manager.py	66	def test_add_generation_with_results_json(context_mgr, run_dir):	CODE
LOW	tests/test_context_manager.py	88	def test_finalize_with_metrics(context_mgr, run_dir):	CODE
LOW	tests/test_context_manager.py	112	def test_multiple_generations_track_deltas(mock_llm, context_mgr, run_dir):	CODE
LOW	tests/test_config_injection.py	16	def test_load_agent_execution_honors_injected_max_size(tmp_path):	CODE
LOW	tests/test_config_injection.py	31	def test_run_evaluation_honors_injected_timeout(mock_run, tmp_path):	CODE
LOW	tests/test_config_injection.py	45	def test_context_manager_stores_injected_config(tmp_path):	CODE
LOW	tests/test_config_injection.py	52	def test_from_env_override_reaches_instance(monkeypatch):	CODE
LOW⚡	tests/test_load_execution_formats.py	12	def test_missing_returns_exact_error(tmp_path):	CODE
LOW⚡	tests/test_load_execution_formats.py	18	def test_empty_multi_folder_returns_exact_error(tmp_path):	CODE
48 more matches not shown…

Excessive Try-Catch Wrapping27 hits · 36 pts

Severity	File	Line	Snippet	Context
MEDIUM	EVALUATION_GUIDE.md	93	print(f"Error: {submission} not found")	CODE
MEDIUM	sia/prepare_mlebench_dataset.py	39	print(f"Error running mlebench prepare: {e}", file=sys.stderr)	CODE
MEDIUM	sia/prepare_mlebench_dataset.py	61	print(f"Error: Prepared directory not found: {prepared_dir}", file=sys.stderr)	CODE
LOW	sia/prepare_mlebench_dataset.py	170	except Exception as e:	CODE
LOW	sia/orchestrator.py	412	except Exception as e:	STRING
LOW	sia/agent_impls/claude.py	93	except Exception as e:	CODE
LOW	sia/agent_impls/pydantic_ai.py	129	except Exception as e:	CODE
LOW	sia/agent_impls/openhands.py	101	except Exception as e:	CODE
LOW	…asks/longcot-chess/reference/reference_target_agent.py	139	except Exception as e:	CODE
MEDIUM⚡	…asks/longcot-chess/reference/reference_target_agent.py	283	print("Error: GEMINI_API_KEY environment variable not set")	STRING
MEDIUM⚡	…asks/longcot-chess/reference/reference_target_agent.py	290	print(f"Error: Dataset directory does not exist: {args.dataset_dir}")	STRING
LOW⚡	sia/tasks/_shared/reference_target_agent.py	57	except Exception as e:	CODE
LOW⚡	sia/tasks/_shared/reference_target_agent.py	67	except Exception as e:	CODE
LOW	sia/tasks/_shared/reference_target_agent.py	86	except Exception as e:	CODE
LOW	sia/tasks/_shared/reference_target_agent.py	155	except Exception as e:	CODE
LOW	sia/tasks/gpqa/reference/reference_target_agent.py	150	except Exception as e:	CODE
LOW	sia/tasks/gpqa/reference/reference_target_agent.py	165	except Exception as exc:	CODE
MEDIUM⚡	sia/tasks/gpqa/reference/reference_target_agent.py	223	print(f"Error on question {question_result['question_id']}: {question_result['error']}")	CODE
LOW⚡	…/spaceship-titanic/reference/reference_target_agent.py	57	except Exception as e:	CODE
LOW⚡	…/spaceship-titanic/reference/reference_target_agent.py	67	except Exception as e:	CODE
LOW	…/spaceship-titanic/reference/reference_target_agent.py	86	except Exception as e:	CODE
MEDIUM	sia/tasks/lawbench/data/public/evaluate.py	78	print(f"Error: No submission.csv or predictions.csv found in {args.gen_dir}")	CODE
LOW	sia/tasks/lawbench/data/public/evaluate.py	130	except Exception as e:	CODE
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	202	except Exception as e:	CODE
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	299	except Exception as e:	CODE
LOW	sia/web/server.py	128	except Exception as exc: # pragma: no cover - background best-effort	CODE
MEDIUM	sia/web/server.py	125	def _run() -> None:	CODE

Unused Imports36 hits · 36 pts

Severity	File	Line	Context
LOW	sia/config.py	3	CODE
LOW	sia/results.py	8	CODE
LOW	sia/agent_reference.py	21	CODE
LOW	sia/agent_reference.py	26	CODE
LOW	sia/util.py	7	CODE
LOW	sia/util.py	7	CODE
LOW	sia/util.py	7	CODE
LOW	sia/layout.py	9	CODE
LOW	sia/config_files.py	10	CODE
LOW	sia/run_setup.py	7	CODE
LOW	sia/prompts.py	7	CODE
LOW	sia/providers.py	12	CODE
LOW	sia/cli.py	20	CODE
LOW	sia/orchestrator.py	59	CODE
LOW	sia/io_utils.py	9	CODE
LOW	sia/api_keys.py	3	CODE
LOW	sia/profiles.py	17	CODE
LOW	sia/logging_setup.py	8	CODE
LOW	sia/agent_impls/claude.py	3	CODE
LOW	sia/agent_impls/__init__.py	8	CODE
LOW	sia/agent_impls/__init__.py	8	CODE
LOW	sia/agent_impls/__init__.py	8	CODE
LOW	sia/agent_impls/__init__.py	9	CODE
LOW	sia/agent_impls/__init__.py	9	CODE
LOW	sia/agent_impls/__init__.py	9	CODE
LOW	sia/agent_impls/__init__.py	9	CODE
LOW	sia/agent_impls/pydantic_ai.py	12	CODE
LOW	sia/agent_impls/openhands.py	3	CODE
LOW	sia/agent_impls/base.py	11	CODE
LOW	sia/web/server.py	8	CODE
LOW	sia/web/__init__.py	9	CODE
LOW	sia/web/__init__.py	9	CODE
LOW	sia/web/__init__.py	9	CODE
LOW	sia/web/runs.py	24	CODE
LOW	tests/golden_master.py	16	CODE
LOW	tests/test_gpqa_evaluator.py	3	CODE

Structural Annotation Overuse16 hits · 28 pts

Severity	File	Line	Snippet	Context
LOW⚡	sia/prompts.py	351	### Step 1: Verify Dataset Alignment	COMMENT
LOW⚡	sia/prompts.py	358	### Step 2: Check the Sandbox (if using code execution)	COMMENT
LOW	sia/prompts.py	400	### Step 3: Isolate the Problem	COMMENT
LOW⚡	sia/prepare_mlebench_dataset.py	237	# Step 1: Run mlebench prepare	COMMENT
LOW⚡	sia/prepare_mlebench_dataset.py	242	# Step 2: Copy dataset files	COMMENT
LOW⚡	sia/prepare_mlebench_dataset.py	247	# Step 3: Move and rename description	COMMENT
LOW	sia/prepare_mlebench_dataset.py	265	# Step 6: Copy reference agent	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	97	# Step 1: Load test data	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	109	# Step 2: Load valid classes	COMMENT
LOW⚡	sia/tasks/lawbench/reference/reference_target_agent.py	122	# Step 3: Load training data for few-shot examples (optional)	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	135	# Step 4: Generate predictions with multi-trajectory logging	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	239	# Step 5: Save submission	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	262	# Step 6: Save summary execution log	COMMENT
LOW	docs/walkthrough.md	5	## Step 1: Set up the task directory	COMMENT
LOW	docs/walkthrough.md	41	## Step 2: Run the orchestrator	COMMENT
LOW	docs/walkthrough.md	68	## Step 3: Analyze results	COMMENT

Verbosity Indicators13 hits · 24 pts

Severity	File	Line	Snippet	Context
LOW⚡	sia/prompts.py	351	### Step 1: Verify Dataset Alignment	COMMENT
LOW⚡	sia/prompts.py	358	### Step 2: Check the Sandbox (if using code execution)	COMMENT
LOW	sia/prompts.py	400	### Step 3: Isolate the Problem	COMMENT
LOW⚡	sia/prepare_mlebench_dataset.py	237	# Step 1: Run mlebench prepare	COMMENT
LOW⚡	sia/prepare_mlebench_dataset.py	242	# Step 2: Copy dataset files	COMMENT
LOW⚡	sia/prepare_mlebench_dataset.py	247	# Step 3: Move and rename description	COMMENT
LOW	sia/prepare_mlebench_dataset.py	265	# Step 6: Copy reference agent	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	97	# Step 1: Load test data	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	109	# Step 2: Load valid classes	COMMENT
LOW⚡	sia/tasks/lawbench/reference/reference_target_agent.py	122	# Step 3: Load training data for few-shot examples (optional)	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	135	# Step 4: Generate predictions with multi-trajectory logging	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	239	# Step 5: Save submission	COMMENT
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	262	# Step 6: Save summary execution log	COMMENT

Deep Nesting15 hits · 15 pts

Severity	File	Line	Context
LOW	sia/context_manager.py	267	CODE
LOW	sia/context_manager.py	338	CODE
LOW	sia/context_manager.py	385	CODE
LOW	sia/context_manager.py	454	CODE
LOW	sia/orchestrator.py	91	CODE
LOW	sia/orchestrator.py	185	CODE
LOW	sia/orchestrator.py	425	CODE
LOW	sia/agent_impls/claude.py	13	CODE
LOW	…asks/longcot-chess/reference/reference_target_agent.py	93	CODE
LOW	sia/tasks/_shared/reference_target_agent.py	175	CODE
LOW	sia/tasks/gpqa/data/public/evaluate.py	111	CODE
LOW	sia/tasks/gpqa/reference/reference_target_agent.py	112	CODE
LOW	…/spaceship-titanic/reference/reference_target_agent.py	102	CODE
LOW	sia/tasks/lawbench/data/public/evaluate.py	50	CODE
LOW	sia/tasks/lawbench/data/public/evaluate.py	119	CODE

AI Slop Vocabulary5 hits · 13 pts

Severity	File	Line	Snippet	Context
MEDIUM	sia/context_manager.py	372	# Priority 3: Parse stdout (check both harness and weights mode log files)	COMMENT
MEDIUM	sia/orchestrator.py	884	# Build meta prompt based on focus mode (weights=RL tuning, harness=code improvement)	COMMENT
MEDIUM	sia/orchestrator.py	661	# Use train.py for weights mode (RL tuning), target_agent.py for harness mode	STRING
MEDIUM	sia/web/runs.py	299	# For stdout_log, check both harness and weights mode variants	COMMENT
MEDIUM	sia/web/runs.py	365	# For stdout_log, check both harness and weights mode variants	COMMENT

Modern AI Meta-Vocabulary2 hits · 6 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	sia/tasks/lawbench/reference/reference_target_agent.py	122	# Step 3: Load training data for few-shot examples (optional)	COMMENT
MEDIUM	docs/architecture.md	33	│ ├── orchestrator.py # Main orchestration logic	CODE

Modern Structural Boilerplate5 hits · 5 pts

Severity	File	Line	Snippet	Context
LOW	sia/util.py	9	__all__ = ["available_agent_impls", "get_agent_impl", "run_agent"]	CODE
LOW	sia/orchestrator.py	68	__all__ = [	CODE
LOW	sia/agent_impls/__init__.py	11	__all__ = ["available_agent_impls", "get_agent_impl", "register", "run_agent"]	CODE
LOW	sia/tasks/lawbench/reference/reference_target_agent.py	27	logger = logging.getLogger(__name__)	CODE
LOW	sia/web/__init__.py	11	__all__ = ["create_app", "serve", "serve_in_background"]	CODE

Self-Referential Comments2 hits · 4 pts

Severity	File	Line	Snippet	Context
MEDIUM	sia/context_manager.py	142	# Create a temporary directory for LLM execution	STRING
MEDIUM	tests/test_context_manager.py	17	# Create a minimal target_agent.py	COMMENT

Magic Placeholder Names1 hit · 4 pts

Severity	File	Line	Snippet	Context
HIGH⚡	…asks/longcot-chess/reference/reference_target_agent.py	285	print(" export GEMINI_API_KEY='your-api-key-here'")	STRING

Redundant / Tautological Comments2 hits · 3 pts

Severity	File	Line	Snippet	Context
LOW	sia/orchestrator.py	235	# Check if results.json was created	COMMENT
LOW	sia/tasks/longcot-chess/data/public/evaluate.py	100	# Check if any permutation matches (for unordered answers)	COMMENT

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings