• There are 147 files with units, with 24,899 lines of code.
• 1 very complex files (McCabe index sum 251+; 1,115 lines of code)
• 1 complex files (McCabe index sum 126-250; 726 lines of code)
• 5 medium complex files (McCabe index sum 51-125; 1,436 lines of code)
• 14 slightly complex files (McCabe index sum 26-50; 4,509 lines of code)
• 126 simple files (McCabe index sum 1-25; 17,113 lines of code)
• 149 main files have no units (e.g. languages without unit analysis, or files without functions) and are not classified.

The McCabe index of a file grows with its size: per 100 lines tells a large but plain file from a dense one, and most complex unit whether the complexity sits in one unit or is spread over many.

File	# lines	# units	McCabe index (sum)	per 100 lines	most complex unit
stats.py in project/evmbench/scripts	1115	45	255	23	51
replay.py in project/swelancer/runtime_scripts	726	24	180	25	39
apply_patch.py in project/paperbench/paperbench/solvers	359	24	116	32	21
tasks.py in project/paperbench/paperbench/rubric	285	29	87	31	11
type_helpers.py in project/common/preparedness_turn_completer/preparedness_turn_completer	199	18	74	37	10
audit.py in project/evmbench/evmbench	173	20	55	32	6
app.py in project/paperbench/paperbench/gui	420	14	53	13	10
backfill_agent_tokens.py in project/evmbench/scripts	159	9	50	31	10
patch.py in project/evmbench/evmbench/nano/grade	477	7	49	10	14
script.js in project/paperbench/paperbench/gui/static	345	18	48	14	4
count_disclosures.py in project/evmbench/scripts	238	8	41	17	23
config.py in project/evmbench/evmbench/ploit	179	10	38	21	17
utils.py in project/paperbench/paperbench	171	21	37	22	6
library_config.py in project/common/nanoeval/nanoeval	173	25	36	21	8
local.py in project/common/alcatraz/alcatraz/clusters	1590	15	34	2	7
email_handler.py in project/swelancer/runtime_utils	97	8	33	34	12
monitor.py in project/paperbench/paperbench/monitor	132	11	29	22	12
docker_build.py in project/evmbench	118	5	28	24	13
simple.py in project/paperbench/paperbench/judge	563	10	28	5	9
utils.py in project/evmbench/evmbench	102	13	27	26	15
build_images.py in project/swelancer/scripts	165	11	27	16	11
monitor.py in project/common/nanoeval/nanoeval	179	5	24	13	16
utils.py in project/common/preparedness_turn_completer/preparedness_turn_completer	125	5	24	19	17
bootstrap.py in project/evmbench/evmbench	136	10	23	17	6
converters.py in project/common/preparedness_turn_completer/preparedness_turn_completer/oai_responses_turn_completer	383	18	23	6	6
graded_task_node.py in project/paperbench/paperbench/judge	109	10	22	20	4
smoke_test_agent_clis.py in project/evmbench/scripts	299	7	21	7	5
eval.py in project/common/nanoeval/nanoeval	183	11	21	11	5
utils.py in project/paperbench/paperbench/judge	117	5	21	18	9
judge_eval_perf_cost.py in project/paperbench/experiments/judge_eval	177	6	20	11	7
eval.py in project/paperbench/paperbench/nano	284	7	19	7	7
setup.rs in project/evmbench/ploit/crates/core/src	450	9	18	4	7
user_tool.py in project/swelancer/issues/32236_652	116	3	18	16	11
task_to_alcatraz_config.py in project/common/nanoeval_alcatraz/nanoeval_alcatraz	67	1	18	27	18
judge_eval_perf_tables.py in project/paperbench/experiments/judge_eval	124	5	18	15	8
utils.py in project/paperbench/paperbench/nano	68	3	18	26	7
agent.py in project/evmbench/evmbench/agents	127	4	17	13	10
json_recorder.py in project/common/nanoeval/nanoeval	124	15	16	13	2
metrics.py in project/paperbench/paperbench	231	11	16	7	3
completer.py in project/paperbench/paperbench/solvers/basicagent	68	5	16	24	8
registry.py in project/paperbench/paperbench/judge/judge_eval	73	7	16	22	7
token_usage.py in project/paperbench/paperbench/judge	43	7	16	37	5
task.py in project/evmbench/evmbench/nano	510	5	15	3	4
eval.py in project/swelancer/swelancer	548	8	15	3	3
recorder_protocol.py in project/common/nanoeval/nanoeval	62	14	15	24	2
result.rs in project/evmbench/ploit/crates/grader/src	238	10	14	6	3
manager.rs in project/evmbench/ploit/crates/config/src	278	4	14	5	9
_multiprocessing_utils.py in project/common/nanoeval/nanoeval	54	5	14	26	6
structs.py in project/paperbench/paperbench/nano	180	6	14	8	5
process.rs in project/evmbench/ploit/crates/utils/src	104	4	13	13	4

• File complexity is the sum of the McCabe indexes (conditional complexity) of the units (methods, functions...) in a file.
• Files are classified in five categories based on that sum: 1-25 (simple files), 26-50 (slightly complex files), 51-125 (medium complex files), 126-250 (complex files), 251+ (very complex files). The default bands are five times the unit complexity bands, as a file is a group of units.
• Only main code files with units are classified. Files without units (languages without unit analysis, or files without functions) have no complexity to measure and are left out rather than counted as simple.
• A very complex file holds many decisions, which often means it has too many responsibilities: a candidate to split, and a place where changes need more testing.