gptme/gptme

35.7

Adjusted Score

35.7

Raw Score

100%

Time Factor

2026-08-29

Last Push

4.4K

Stars

Python

Language

467.8K

Lines of Code

1.4K

Files

13.1K

Pattern Hits

2026-08-29

Scan Date

0.12

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 173MEDIUM 1705LOW 11178

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 13056 distinct pattern matches across 26 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 Identifiers8713 hits · 7570 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_server_path_traversal.py	50	def test_events_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	55	def test_step_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	63	def test_tool_confirm_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	71	def test_rerun_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	76	def test_elicit_respond_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	84	def test_interrupt_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	96	def test_browse_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	101	def test_upload_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	106	def test_file_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	112	def test_preview_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	120	def test_download_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW⚡	tests/test_server_path_traversal.py	285	def test_user_avatar_rejects_non_image_extension(	CODE
LOW⚡	tests/test_server_path_traversal.py	303	def test_user_avatar_accepts_image_extension(	CODE
LOW⚡	tests/test_server_path_traversal.py	335	def test_agent_avatar_by_path_rejects_non_image_extension(	CODE
LOW⚡	tests/test_server_path_traversal.py	385	def test_conversation_agent_avatar_rejects_non_image_extension(	CODE
LOW	tests/test_server_path_traversal.py	38	def _assert_traversal_rejected(response):	CODE
LOW	tests/test_server_path_traversal.py	132	def test_post_message_rejects_traversal(self, client: FlaskClient, payload: str):	CODE
LOW	tests/test_server_path_traversal.py	170	def test_rejects_traversal_in_files(	CODE
LOW	tests/test_server_path_traversal.py	183	def test_safe_relative_path_not_rejected_as_traversal(	CODE
LOW	tests/test_server_path_traversal.py	220	def test_name_with_dots_no_traversal(self, client: FlaskClient):	CODE
LOW	tests/test_server_path_traversal.py	254	def test_generate_rejects_branch_traversal(self, client: FlaskClient, payload: str):	CODE
LOW	tests/test_server_path_traversal.py	266	def test_valid_branch_accepted(self, client: FlaskClient):	CODE
LOW	tests/test_server_path_traversal.py	319	def test_user_avatar_rejects_dotfile_without_image_ext(	CODE
LOW	tests/test_server_path_traversal.py	361	def test_agent_avatar_by_path_accepts_image_extension(	CODE
LOW	tests/test_server_path_traversal.py	427	def test_user_avatar_rejects_disguised_non_image(	CODE
LOW	tests/test_server_path_traversal.py	470	def test_rejects_absolute_path_outside_cwd(	CODE
LOW	tests/test_server_path_traversal.py	484	def test_rejects_home_directory_path(	CODE
LOW	tests/test_server_path_traversal.py	498	def test_rejects_traversal_in_path(	CODE
LOW	tests/test_server_path_traversal.py	525	def test_rejects_non_object_json_body(self, client: FlaskClient, body: object):	CODE
LOW	tests/test_server_path_traversal.py	537	def test_rejects_name_that_slugifies_to_empty(	CODE
LOW	tests/test_server_path_traversal.py	611	def test_whitespace_only_id_rejected_unit(self):	CODE
LOW	tests/test_server_path_traversal.py	626	def test_control_or_edge_whitespace_id_rejected_unit(self, bad_id: str):	CODE
LOW	tests/test_server_path_traversal.py	638	def test_multibyte_over_limit_rejected(self):	CODE
LOW	tests/test_server_path_traversal.py	661	def test_too_long_branch_unit(self):	CODE
LOW	tests/test_server_path_traversal.py	696	def test_multibyte_branch_over_limit_rejected(self):	CODE
LOW	tests/test_server_path_traversal.py	716	def test_too_long_branch_via_endpoint(self, client: FlaskClient):	CODE
LOW	tests/test_server_path_traversal.py	738	def test_rejects_traversal_payloads(self):	CODE
LOW	tests/test_server_path_traversal.py	772	def test_rejects_non_string_branch(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	41	def test_matches_simple_at_ref(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	44	def test_matches_hyphenated_name(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	47	def test_matches_multiple_refs(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	53	def test_no_false_positive_on_email_address(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	64	def test_no_messages_yields_nothing(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	68	def test_no_at_ref_yields_nothing(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	73	def test_at_ref_matches_server_yields_hint(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	85	def test_at_ref_no_matching_server_yields_nothing(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	90	def test_no_loaded_servers_yields_nothing(self):	CODE
LOW⚡	tests/test_hooks_mcp_namespace_hint.py	96	def test_multiple_servers_in_one_message(self):	CODE
LOW	tests/test_hooks_mcp_namespace_hint.py	109	def test_only_last_user_message_checked(self):	CODE
LOW	tests/test_hooks_mcp_namespace_hint.py	120	def test_tool_without_description(self):	CODE
LOW	tests/test_hooks_mcp_namespace_hint.py	136	def test_hint_format_contains_full_tool_name(self):	CODE
LOW	tests/test_commands_llm.py	21	def test_cmd_model_appends_replacement_prompt_and_persists_generation(tmp_path: Path):	CODE
LOW	tests/test_commands_llm.py	51	def test_cmd_tools_load_appends_replacement_prompt(tmp_path: Path):	CODE
LOW	tests/test_commands_llm.py	76	def test_cmd_context_reports_context_savings(tmp_path: Path):	CODE
LOW	tests/test_shell_issue729.py	26	def test_stderr_redirect_with_pipe(shell):	CODE
LOW	tests/test_shell_issue729.py	49	def test_jq_complex_syntax_nested_pipe(shell):	CODE
LOW	tests/test_shell_issue729.py	87	def test_jq_function_in_filter(shell):	CODE
LOW	tests/test_shell_issue729.py	118	def test_stdout_and_stderr_redirect(shell):	CODE
LOW	tests/test_dspy_basic.py	138	def test_tool_metric_empty_messages():	CODE
LOW	tests/test_dspy_basic.py	154	def test_prompt_optimizer_init(mock_dspy):	CODE
8653 more matches not shown…

Decorative Section Separators1426 hits · 5004 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	tests/test_hooks_mcp_namespace_hint.py	35	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_hooks_mcp_namespace_hint.py	37	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_hooks_mcp_namespace_hint.py	58	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_hooks_mcp_namespace_hint.py	60	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_computer_api.py	348	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_computer_api.py	350	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	119	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	121	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	181	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	183	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	213	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	215	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	297	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_api.py	299	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_mcp_adapter.py	366	# ============================================================================	COMMENT
MEDIUM	tests/test_mcp_adapter.py	368	# ============================================================================	COMMENT
MEDIUM⚡	tests/test_lessons.py	42	# ──────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_lessons.py	44	# ──────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_lessons.py	67	# ──────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_lessons.py	69	# ──────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_lessons.py	93	# ──────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_lessons.py	95	# ──────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_lessons.py	651	# ──────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_lessons.py	653	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	137	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	139	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	161	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	163	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	202	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	204	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	404	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	406	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	491	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	493	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	777	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	779	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	814	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_lessons.py	816	# ──────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_eval_behavioral_solutions.py	1379	# ── Parametrized solution verification ───────────────────────────────────────	STRING
MEDIUM	tests/test_eval_behavioral_solutions.py	1451	# ── Smoke tests for individual scenarios ─────────────────────────────────────	STRING
MEDIUM	tests/test_computer_x11_integration.py	33	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_x11_integration.py	35	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_x11_integration.py	63	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_x11_integration.py	65	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_x11_integration.py	199	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_computer_x11_integration.py	201	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_computer_x11_integration.py	219	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_computer_x11_integration.py	221	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_backtrack.py	17	# ── helpers ──────────────────────────────────────────────────────────────────	COMMENT
MEDIUM	tests/test_backtrack.py	50	# ── ConvCheckpoint primitives ─────────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_backtrack.py	100	# ── /backtrack command ────────────────────────────────────────────────────────	COMMENT
MEDIUM⚡	tests/test_k_best.py	108	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_k_best.py	110	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_k_best.py	228	# ---------------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/test_k_best.py	230	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_k_best.py	12	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_k_best.py	14	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_k_best.py	36	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_k_best.py	38	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_k_best.py	183	# ---------------------------------------------------------------------------	COMMENT
1366 more matches not shown…

Cross-Language Confusion115 hits · 638 pts

Severity	File	Line	Snippet	Context
HIGH	tests/test_shell_issue729.py	4	1. File descriptor redirects (2>/dev/null, 2>&1, etc.)	STRING
HIGH⚡	tests/test_codeblock.py	160	pwd && ls -la	CODE
HIGH⚡	tests/test_codeblock.py	166	assert codeblocks[0] == Codeblock("shell", "pwd && ls -la")	CODE
HIGH	tests/test_shell_output_mixing_issue408.py	106	ret1, out1, err1 = shell.run("sleep 0.1 && echo 'delayed output'")	CODE
HIGH	tests/test_server_client.py	74	b'data: {"type": "connected", "session_id": null}'	CODE
HIGH⚡	tests/test_tools_shell_background.py	553	job = start_background_job("echo out && echo err >&2")	CODE
HIGH	tests/test_tools_shell_background.py	186	job = start_background_job("echo hello && echo world")	CODE
HIGH	tests/test_tools_shell_background.py	202	job = start_background_job("sleep 0.1 && echo done")	CODE
HIGH	tests/test_shell_issue772.py	5	2. Combinations of \|\| with pipe operators (\|)	STRING
HIGH⚡	tests/test_shell_issue772.py	27	Issue: cat .env.local 2>/dev/null \|\| cat .env.example \| head -20	STRING
HIGH⚡	tests/test_shell_issue772.py	31	The shell tool was treating the first \| in \|\| as a pipe operator,	STRING
HIGH⚡	tests/test_shell_issue772.py	40	code = f"cat /nonexistent 2>/dev/null \|\| cat {temp_file} \| head -3"	CODE
HIGH⚡	tests/test_shell_issue772.py	62	Verify that \|\| operator works correctly in basic cases.	STRING
HIGH⚡	tests/test_shell_issue772.py	64	code = "false \|\| echo 'fallback'"	CODE
HIGH	tests/test_shell_issue772.py	85	code = f"cat /nonexistent 2>/dev/null \|\| cat {temp_file}"	CODE
HIGH⚡	tests/test_shell_issue772.py	102	"""Test multiple \|\| operators in sequence.	STRING
HIGH⚡	tests/test_shell_issue772.py	104	Verify that chains of \|\| operators are handled correctly.	STRING
HIGH⚡	tests/test_shell_issue772.py	106	code = "false \|\| false \|\| echo 'third try'"	CODE
HIGH⚡	tests/test_shell_issue772.py	115	"""Test combination of \|\| and && operators.	STRING
HIGH⚡	tests/test_shell_issue772.py	115	"""Test combination of \|\| and && operators.	STRING
HIGH⚡	tests/test_shell_issue772.py	119	code = "true && echo 'success' \|\| echo 'failure'"	CODE
HIGH⚡	tests/test_shell_issue772.py	133	cat file1 2>/dev/null \|\| cat file2 \| head -20	STRING
HIGH⚡	tests/test_shell_issue772.py	136	- stderr redirect (2>/dev/null)	STRING
HIGH	tests/test_shell_issue772.py	147	code = f"cat /nonexistent 2>/dev/null \|\| cat {temp_file} \| head -5"	CODE
HIGH	tests/test_execenv.py	410	env.run("cat input.txt > output.txt && echo ' modified' >> output.txt")	CODE
HIGH⚡	tests/test_tools_shell_multiline.py	13	gh pr checks "$pr" --json name,conclusion,workflowName --jq '.[] \| select(.conclusion != null) \| "\\(.workflowName): \	CODE
HIGH⚡	tests/test_tools_shell_validation.py	205	assert _find_first_unquoted_pipe("cmd1 \|\| cmd2") is None	CODE
HIGH⚡	tests/test_tools_shell_validation.py	208	result = _find_first_unquoted_pipe("cmd1 \|\| cmd2 \| cmd3")	CODE
HIGH⚡	tests/test_tools_shell_validation.py	211	cmd = "cmd1 \|\| cmd2 \| cmd3"	CODE
HIGH⚡	tests/test_tools_shell_validation.py	1215	assert is_allowlisted("ls -la && pwd")	CODE
HIGH	tests/test_sandbox.py	1020	# (suppressed with 2>/dev/null) so pre-existing lower hard limits are	COMMENT
HIGH	tests/test_sandbox.py	1031	# Hard limit is best-effort (2>/dev/null) so a pre-existing lower	COMMENT
HIGH	tests/test_reduce.py	859	content="I'll run two things.\n```shell\necho a && echo b\n```",	CODE
HIGH	tests/test_reduce.py	925	content="I'll run two things.\n```shell\necho a && echo b\n```",	CODE
HIGH	tests/test_reduce.py	987	content="I'll run two things.\n```shell\necho a && echo b\n```",	CODE
HIGH⚡	tests/test_util_review.py	1429	self._review_jsonl('```json\n{"findings": null}\n```')	CODE
HIGH⚡	tests/test_util_review.py	1424	"""When findings key is not a list (null, int, dict), the block is skipped."""	STRING
HIGH	tests/test_util_review.py	363	"""Non-list findings value (null, int, dict) must not crash from_dict."""	STRING
HIGH	tests/test_tools_shell.py	1275	# stdin is a pipe (not /dev/null, not terminal)	COMMENT
HIGH	tests/test_tools_shell.py	1325	# stdin is a pipe (not /dev/null, not terminal)	COMMENT
HIGH	tests/test_tools_shell.py	1638	job = start_background_job("sleep 0.1 && echo 'done'")	CODE
HIGH	tests/test_tools_shell.py	1641	assert job.command == "sleep 0.1 && echo 'done'"	CODE
HIGH	tests/test_tools_shell.py	1719	start_background_job("sleep 0.5 && echo 'job1'")	CODE
HIGH	tests/test_tools_shell.py	1720	start_background_job("sleep 0.5 && echo 'job2'")	CODE
HIGH	tests/test_tools_shell.py	205	cat /tmp/test_space.sh && rm /tmp/test_space.sh"""	STRING
HIGH	tests/test_tools_shell.py	216	cat /tmp/test_space2.sh && rm /tmp/test_space2.sh"""	STRING
HIGH	tests/test_tools_shell.py	1556	"""curl http://example.com \| bash && cat << EOF	STRING
HIGH	scripts/demo_capture.py	121	"playwright (pip install playwright && playwright install chromium)"	CODE
HIGH	scripts/demo_capture.py	163	f"cd {tmpdir} && gptme --non-interactive --no-confirm {model_flag} '{prompt}' 2>&1 \|\| true",	CODE
HIGH	scripts/demo_capture.py	23	- playwright (pip install playwright && playwright install chromium)	STRING
HIGH	gptme/sandbox.py	223	"ulimit -v N && BASH_ENV='' exec bash"]``	STRING
HIGH	gptme/sandbox.py	243	prefix = f"ulimit -H -v {kib} 2>/dev/null; ulimit -v {kib} && "	CODE
HIGH	gptme/sandbox.py	276	["bash", "-c", f"ulimit -H -v {kib} 2>/dev/null; ulimit -v {kib}"],	CODE
HIGH	gptme/tools/_browser_playwright.py	439	if (el.className && typeof el.className === 'string')	STRING
HIGH	gptme/tools/tmux.py	576	"new-session cd ~/bob && gptme --non-interactive 'What is your latest work?'",	CODE
HIGH	gptme/tools/computer.py	2177	"(fix: pip install playwright && playwright install chromium)"	STRING
HIGH	gptme/tools/computer.py	2235	"pip install playwright && playwright install chromium",	STRING
HIGH	gptme/tools/python.py	105	_ipython.push({func.__name__: func})	CODE
HIGH	gptme/tools/python.py	199	_ipython.push(registered_functions)	CODE
HIGH	gptme/util/gh.py	744	'.[] \| select(.event == "cross-referenced" and .source.issue.pull_request != null) '	CODE
55 more matches not shown…

Unused Imports731 hits · 604 pts

Severity	File	Line	Context
LOW	tests/test_computer_api.py	7	CODE
LOW	tests/test_computer_video_frames_cli.py	7	CODE
LOW	tests/test_computer_x11_integration.py	17	CODE
LOW	tests/test_backtrack.py	8	CODE
LOW	tests/test_util_clipboard_paste.py	24	CODE
LOW	tests/test_k_best.py	3	CODE
LOW	tests/conftest.py	22	CODE
LOW	tests/test_acp_agent.py	28	CODE
LOW	tests/test_dirs.py	14	CODE
LOW	tests/test_computer_audit.py	8	CODE
LOW	tests/test_workspace_snapshot.py	12	CODE
LOW	tests/test_cli_status.py	3	CODE
LOW	tests/test_computer_actions.py	8	CODE
LOW	tests/test_computer_screenshot_cli.py	7	CODE
LOW	tests/test_computer_task.py	8	CODE
LOW	tests/test_eventlog.py	3	CODE
LOW	tests/test_mcp_server.py	3	CODE
LOW	tests/test_mcp_server.py	586	CODE
LOW	tests/test_provider_plugins.py	7	CODE
LOW	tests/test_terminal_startup_latency.py	8	CODE
LOW	tests/test_circuit_breaker.py	11	CODE
LOW	tests/test_anti_slop.py	3	CODE
LOW	tests/test_timings.py	22	CODE
LOW	tests/test_acp_client.py	12	CODE
LOW	tests/test_acp_client.py	32	CODE
LOW	tests/test_model_selection_trace_export.py	3	CODE
LOW	tests/test_cli_from_turn.py	3	CODE
LOW	tests/test_tools_hashline_edit.py	3	CODE
LOW	tests/test_computer_record.py	8	CODE
LOW	tests/test_local_discovery.py	7	CODE
LOW	tests/test_tools_patch_anchored.py	3	CODE
LOW	tests/test_tasks_api.py	18	CODE
LOW	tests/test_execute_msg_hook_call_id.py	17	CODE
LOW	tests/test_computer_doctor_cmd.py	7	CODE
LOW	tests/test_acp_session_runtime.py	3	CODE
LOW	tests/test_cmd_service.py	7	CODE
LOW	tests/test_util_hooks.py	3	CODE
LOW	tests/test_tools_request_tool_change.py	1	CODE
LOW	tests/test_computer_demo_cmd.py	8	CODE
LOW	tests/test_gptme_provider.py	8	CODE
LOW	tests/test_acp_protocol_logging.py	7	CODE
LOW	tests/test_github_ci_self_heal.py	1	CODE
LOW	tests/test_browser_state.py	3	CODE
LOW	tests/test_computer_run_task_cli.py	8	CODE
LOW	tests/test_tools__anchored.py	7	CODE
LOW	tests/test_snapshot_command.py	3	CODE
LOW	tests/test_snapshot_command.py	14	CODE
LOW	tests/test_treeofthoughts.py	8	CODE
LOW	tests/test_tools_convert.py	3	CODE
LOW	tests/test_computer_gate.py	11	CODE
LOW	tests/test_computer_use_integration.py	18	CODE
LOW	tests/test_computer_milestone_integration.py	23	CODE
LOW	tests/test_util_cli_review_watch.py	3	CODE
LOW	tests/test_computer_docker_config.py	15	CODE
LOW	tests/test_skill_commands.py	184	CODE
LOW	tests/test_acp_e2e_smoke.py	13	CODE
LOW	tests/test_computer_settle.py	12	CODE
LOW	tests/test_llm_openai.py	6	CODE
LOW	tests/test_llm_openai.py	7	CODE
LOW	tests/test_computer_risk_streaming.py	16	CODE
671 more matches not shown…

Excessive Try-Catch Wrapping587 hits · 594 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_eval_behavioral_solutions.py	925	except Exception:	STRING
LOW	tests/test_eval_behavioral_solutions.py	1435	except Exception as e:	STRING
LOW	tests/conftest.py	141	except Exception:	CODE
LOW	tests/conftest.py	418	except Exception as e:	CODE
LOW	tests/conftest.py	449	except Exception as e:	CODE
LOW	tests/conftest.py	497	except Exception:	CODE
LOW	tests/conftest.py	741	except Exception as e:	CODE
LOW	tests/test_acp_agent.py	413	except Exception as e:	CODE
LOW	tests/test_chat_history.py	69	except Exception as e:	CODE
MEDIUM	tests/test_chat_history.py	70	print(f"Error generating chat history: {e}")	CODE
MEDIUM	tests/test_circuit_breaker.py	345	def do_fail():	CODE
MEDIUM	tests/test_circuit_breaker.py	371	def do_fail():	CODE
LOW⚡	tests/test_circuit_breaker.py	348	except Exception as e:	CODE
LOW	tests/test_circuit_breaker.py	414	except Exception:	CODE
LOW	tests/test_hooks_server_confirm.py	211	except Exception as e:	CODE
LOW	tests/test_hooks_server_confirm.py	219	except Exception as e:	CODE
MEDIUM	tests/test_hooks_server_confirm.py	204	def register_and_wait(tool_id):	CODE
LOW	tests/test_llm_anthropic.py	377	except Exception:	CODE
MEDIUM	tests/test_subagent_unit.py	5139	def thread_target():	CODE
LOW	tests/test_subagent_unit.py	5151	except Exception as e:	CODE
MEDIUM	tests/test_eval_behavioral.py	1022	print(f"Error: negative value {value}")	CODE
MEDIUM	tests/test_eval_behavioral.py	1028	print(f"Error: {e}")	CODE
MEDIUM	tests/test_eval_behavioral.py	984	def process_record(record):	CODE
LOW⚡	tests/test_eval_behavioral.py	1680	except Exception:	CODE
LOW	tests/test_eval_behavioral.py	2949	except Exception as err:	STRING
LOW⚡	tests/test_mcp_discovery.py	160	except Exception as e:	CODE
MEDIUM	tests/test_computer_use_integration.py	37	def _playwright_available() -> bool:	CODE
LOW⚡	tests/test_computer_use_integration.py	184	except Exception:	CODE
LOW	tests/test_computer_use_integration.py	61	except Exception:	CODE
MEDIUM	tests/test_computer_milestone_integration.py	34	def _playwright_available() -> bool:	CODE
LOW⚡	tests/test_computer_milestone_integration.py	81	except Exception:	CODE
LOW	tests/test_computer_milestone_integration.py	54	except Exception:	CODE
LOW⚡	tests/test_server_bad_input.py	346	except Exception:	CODE
LOW	tests/test_llm_openai.py	688	except Exception:	CODE
LOW	tests/test_tools_subagent.py	1225	except Exception:	CODE
LOW	tests/test_tools_subagent.py	1274	except Exception:	CODE
LOW	tests/test_server_session_models.py	714	except Exception as e:	CODE
LOW	tests/test_server_session_models.py	723	except Exception as e:	CODE
LOW	tests/test_server_session_models.py	756	except Exception as e:	CODE
LOW	tests/test_server_session_models.py	764	except Exception as e:	CODE
LOW	tests/test_oauth_openrouter.py	88	except Exception as e:	CODE
MEDIUM	tests/test_oauth_openrouter.py	81	def target() -> None:	CODE
MEDIUM	tests/test_hooks_registry.py	911	def registerer():	CODE
MEDIUM	tests/test_hooks_registry.py	920	def unregisterer():	CODE
LOW	tests/test_hooks_registry.py	917	except Exception as e:	CODE
LOW	tests/test_hooks_registry.py	924	except Exception as e:	CODE
LOW	tests/test_tui.py	520	except Exception as e:	CODE
MEDIUM	scripts/auto_rename_logs.py	44	print(f"Error: Logs directory not found: {logs_dir}")	CODE
LOW	scripts/auto_rename_logs.py	142	except Exception as e:	CODE
MEDIUM	scripts/auto_rename_logs.py	143	print(f"Error processing {conv_dir}: {e}")	CODE
LOW	scripts/treeofthoughts.py	285	except Exception:	CODE
MEDIUM	scripts/treeofthoughts.py	454	print(f"Error: workspace {workspace} is not a directory", file=sys.stderr)	CODE
MEDIUM	scripts/github_bot.py	72	print(f"Error: {name} environment variable is required", file=sys.stderr)	CODE
MEDIUM	scripts/github_bot.py	642	print("Error in resolve mode: provide GITHUB_EVENT_PATH or --issue")	STRING
MEDIUM	scripts/github_bot.py	726	print("Error: Either GITHUB_EVENT_PATH or --comment-body required")	STRING
LOW	scripts/build_changelog.py	587	except Exception as e:	STRING
LOW	scripts/analyze_compression.py	116	except Exception as e:	CODE
LOW	scripts/analyze_compression.py	220	except Exception as e:	CODE
LOW	scripts/demo_capture.py	195	except Exception as e:	CODE
LOW	scripts/demo_capture.py	275	except Exception as e:	CODE
527 more matches not shown…

Self-Referential Comments176 hits · 543 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	tests/test_server_path_traversal.py	400	# Create the logdir and a valid conversation log so the	COMMENT
MEDIUM	tests/test_shell_issue729.py	58	# Create a test JSON file	COMMENT
MEDIUM	tests/test_cost_awareness_delayed_warning.py	72	# Create a messages list with a user message	COMMENT
MEDIUM	tests/test_cost_awareness_delayed_warning.py	97	# Create a messages list with an assistant message	COMMENT
MEDIUM	tests/test_eval_behavioral_solutions.py	205	# Create the shared validation module	STRING
MEDIUM	tests/test_replay.py	293	# Create a working context that's nearly full (simulated)	COMMENT
MEDIUM	tests/test_utils.py	16	# Create a test image file that is 0.8MB raw (within 1MB limit)	COMMENT
MEDIUM	tests/test_utils.py	53	# Create a file that is 1.2MB (exceeds 1MB limit)	COMMENT
MEDIUM	tests/conftest.py	530	# Create a temporary file with the given content	COMMENT
MEDIUM	tests/test_acp_agent.py	1031	# Create a fake log directory with a conversation.jsonl	COMMENT
MEDIUM	tests/test_acp_agent.py	1652	# Create a minimal log dir so new_session can resume from disk	COMMENT
MEDIUM	tests/test_tools_chats.py	13	# Create a test conversation containing "python"	COMMENT
MEDIUM	tests/test_server_v2.py	1889	# Create a conversation with a non-test prefix so it isn't filtered out	COMMENT
MEDIUM	tests/test_server_v2.py	1922	# Create a real conversation with a non-test prefix so it isn't filtered	COMMENT
MEDIUM	tests/test_util_cli_skills.py	174	# Create a skill so the index isn't empty (tests the "not found" path, not "empty index")	COMMENT
MEDIUM	tests/test_util_cli_skills.py	206	# Create a skill where metadata.name differs from the H1 title	COMMENT
MEDIUM⚡	tests/test_demo_capture.py	48	# Create a fake file	COMMENT
MEDIUM	tests/test_cc_memory.py	90	# Create a fake CC memory file	COMMENT
MEDIUM	tests/test_agent.py	451	# Create a plist file	COMMENT
MEDIUM	tests/test_agent.py	771	# Create a fake workspace	COMMENT
MEDIUM	tests/test_agent.py	991	# Create a repo with multiple commits	COMMENT
MEDIUM	tests/test_tools_save.py	110	# Create a subdirectory to work from	COMMENT
MEDIUM⚡	tests/test_tools_save.py	132	# Create a directory outside the cwd	COMMENT
MEDIUM⚡	tests/test_tools_save.py	141	# Create a symlink in work_dir pointing to outside_dir	COMMENT
MEDIUM	tests/test_tools_save.py	188	# Create a symlink in work_dir pointing to outside_dir	COMMENT
MEDIUM⚡	tests/test_misc_subprocess_timeouts.py	198	# Create a file so glob has something to find	COMMENT
MEDIUM	tests/test_llm_utils.py	361	# Create a file with a tilde path won't work in CI, but we can test	COMMENT
MEDIUM	tests/test_tool_use.py	239	"import numpy as np\nimport pandas as pd\n\n# Create a simple dataset\n"	CODE
MEDIUM⚡	tests/test_chat_history.py	30	# Create a few mock conversations	COMMENT
MEDIUM	tests/test_plugins.py	463	# Create a hook module with register() function	COMMENT
MEDIUM	tests/test_agent_service_timeouts.py	100	# Create a temp file to simulate a log file	COMMENT
MEDIUM	tests/test_util_gh_mocked.py	587	# Create a body that exceeds 1000 tokens (4000 chars)	COMMENT
MEDIUM	tests/test_util_gh_mocked.py	616	# Create a body with identifiable start and end	COMMENT
MEDIUM	tests/test_util_gh_mocked.py	665	# Create a mock response with a very long review comment	COMMENT
MEDIUM	tests/test_util_gh_mocked.py	720	# Create a long comment with a suggestion in the middle (where truncation would remove it)	COMMENT
MEDIUM	tests/test_util_gh_mocked.py	842	# Create a diff larger than default budget (4000 tokens = 16000 chars)	COMMENT
MEDIUM	tests/test_custom_providers.py	94	# Create a minimal tool spec	COMMENT
MEDIUM	tests/test_prompt.py	123	# Create a symlink	COMMENT
MEDIUM	tests/test_prompt.py	148	# Create a new file in tmp_path	COMMENT
MEDIUM⚡	tests/test_util_safety.py	202	# Create a report with artificially high-risk segments	COMMENT
MEDIUM⚡	tests/test_message.py	534	# Create a message - role with special chars is unusual but should be safe	COMMENT
MEDIUM	tests/test_message.py	246	# Create a test file in tmp_path	COMMENT
MEDIUM	tests/test_llm_anthropic.py	336	# Create a mock that looks like an Anthropic API 500 error	COMMENT
MEDIUM⚡	tests/test_prompt_skills_summary.py	17	# Create a mock skill file	COMMENT
MEDIUM	tests/test_prompt_skills_summary.py	56	# Create a mock skill file with long description	COMMENT
MEDIUM⚡	tests/test_prompts.py	211	# Create a temp workspace with a file	COMMENT
MEDIUM⚡	tests/test_prompts.py	216	# Create a file outside workspace	COMMENT
MEDIUM⚡	tests/test_prompts.py	221	# Create a gptme.toml with path traversal attempt	COMMENT
MEDIUM	tests/test_execenv.py	236	# Create a mock log file	COMMENT
MEDIUM⚡	tests/test_tasks_api.py	1023	# Create a temp task file	COMMENT
MEDIUM	tests/test_tasks_api.py	219	# Create a fake repo subdir with .git	COMMENT
MEDIUM	tests/test_tools_browser.py	393	# Create a minimal mock for pypdf	COMMENT
MEDIUM	tests/test_computer_doctor_cmd.py	736	# Create a tmp_path home with no .Xdefaults	COMMENT
MEDIUM	tests/test_acp_session_runtime.py	193	# Create a conversation via the API so LogManager can find it	COMMENT
MEDIUM	tests/test_acp_session_runtime.py	204	# Create a session backed by the dummy ACP runtime	COMMENT
MEDIUM	tests/test_acp_session_runtime.py	917	# Create a session with a dead ACP runtime	COMMENT
MEDIUM⚡	tests/test_lessons_matcher.py	374	# Create a lesson file	COMMENT
MEDIUM	tests/test_lessons_matcher.py	422	# Create a lesson file	COMMENT
MEDIUM⚡	tests/test_cmd_service.py	966	# Create a subdir whose name contains a character systemd forbids but	COMMENT
MEDIUM	tests/test_server_v2_sessions.py	2080	# Create a conversation via PUT, then manually remove the session to simulate no-session case	COMMENT
116 more matches not shown…

Deep Nesting448 hits · 398 pts

Severity	File	Line	Context
LOW	tests/test_eval_behavioral_solutions.py	55	CODE
LOW	tests/conftest.py	254	CODE
LOW	tests/conftest.py	454	CODE
LOW	tests/conftest.py	575	CODE
LOW	tests/conftest.py	708	CODE
LOW	tests/conftest.py	583	CODE
LOW	tests/conftest.py	717	CODE
LOW	tests/test_tree.py	4	CODE
LOW	tests/test_onboard.py	20	CODE
LOW	tests/test_plugins.py	369	CODE
LOW	tests/test_server_parent_death_watcher.py	45	CODE
LOW	tests/test_util_gh_mocked.py	194	CODE
LOW	tests/test_util_gh_mocked.py	228	CODE
LOW	tests/test_util_gh_mocked.py	712	CODE
LOW	tests/test_util_gh_mocked.py	201	CODE
LOW	tests/test_util_gh_mocked.py	745	CODE
LOW	tests/test_hooks_server_confirm.py	301	CODE
LOW	tests/test_hooks_server_confirm.py	329	CODE
LOW	tests/test_computer_docker_config.py	28	CODE
LOW	tests/test_tools_subagent.py	1176	CODE
LOW	tests/test_tools_subagent.py	1230	CODE
LOW	tests/test_computer_semantic_fixtures.py	183	CODE
LOW	docs/conf.py	25	CODE
LOW	scripts/check_model_freshness.py	95	CODE
LOW	scripts/auto_rename_logs.py	32	CODE
LOW	scripts/github_ci_self_heal.py	294	CODE
LOW	scripts/github_ci_self_heal.py	305	CODE
LOW	scripts/treeofthoughts.py	183	CODE
LOW	scripts/generate_sounds.py	371	CODE
LOW	scripts/build_changelog.py	207	CODE
LOW	scripts/build_changelog.py	327	CODE
LOW	scripts/build_changelog.py	454	CODE
LOW	scripts/check_rst_formatting.py	26	CODE
LOW	scripts/check_rst_formatting.py	170	CODE
LOW	scripts/analyze_compression.py	33	CODE
LOW	scripts/analyze_compression.py	132	CODE
LOW	scripts/analyze_compression.py	236	CODE
LOW	scripts/analyze_compression.py	329	CODE
LOW	scripts/analyze_compression.py	502	CODE
LOW	scripts/demo_capture.py	133	CODE
LOW	scripts/demo_capture.py	200	CODE
LOW	scripts/demo_capture.py	286	CODE
LOW	scripts/demo_capture.py	485	CODE
LOW	gptme/dirs.py	109	CODE
LOW	gptme/checkpoint.py	240	CODE
LOW	gptme/prompt_queue.py	128	CODE
LOW	gptme/model_attestation.py	129	CODE
LOW	gptme/message.py	544	CODE
LOW	gptme/message.py	615	CODE
LOW	gptme/__version__.py	16	CODE
LOW	gptme/__version__.py	60	CODE
LOW	gptme/status_provider.py	77	CODE
LOW	gptme/sandbox.py	548	CODE
LOW	gptme/chat.py	201	CODE
LOW	gptme/chat.py	391	CODE
LOW	gptme/chat.py	695	CODE
LOW	gptme/circuit_breaker.py	108	CODE
LOW	gptme/k_best.py	69	CODE
LOW	gptme/k_best.py	99	CODE
LOW	gptme/k_best.py	101	CODE
388 more matches not shown…

Modern Structural Boilerplate297 hits · 298 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_eval_behavioral_solutions.py	586	def update_metadata(self, key: str, info: dict[str, Any]) -> None:	STRING
LOW	tests/conftest.py	46	logger = logging.getLogger(__name__)	CODE
LOW	tests/test_workspace_snapshot.py	344	def set_git_dates(timestamp: int) -> None:	CODE
LOW	tests/test_server_tts_api.py	12	def _set_openrouter_key(monkeypatch: pytest.MonkeyPatch, value: str \| None) -> None:	CODE
LOW	tests/test_server_v2_auto_stepping.py	12	logger = logging.getLogger(__name__)	CODE
LOW	tests/test_acp_session_runtime.py	60	async def set_session_model(self, session_id: str, model_id: str) -> None:	CODE
LOW	tests/thread_leak.py	30	__all__ = [	CODE
LOW⚡	tests/test_subagent_unit.py	5908	def _set_agent_id(self, agent_id: str) -> None:	CODE
LOW	tests/test_eval_behavioral.py	1055	logger = logging.getLogger(__name__)	CODE
LOW	tests/test_server_v2_tool_confirmation.py	11	logger = logging.getLogger(__name__)	CODE
LOW	scripts/auto_rename_logs.py	14	logger = logging.getLogger(__name__)	CODE
LOW	scripts/github_bot.py	77	def set_github_env(name: str, value: str) -> None:	CODE
LOW	scripts/build_changelog.py	32	logger = logging.getLogger(__name__)	CODE
LOW	scripts/analyze_compression.py	25	logger = logging.getLogger(__name__)	CODE
LOW	scripts/train/collect.py	19	logger = logging.getLogger(__name__)	CODE
LOW	gptme/dirs.py	11	logger = logging.getLogger(__name__)	CODE
LOW	gptme/prompt_queue.py	17	logger = logging.getLogger(__name__)	CODE
LOW	gptme/model_attestation.py	19	logger = logging.getLogger(__name__)	CODE
LOW	gptme/model_attestation.py	274	def set_selection_trace(trace: ModelSelectionTrace \| None) -> None:	CODE
LOW	gptme/session.py	18	logger = logging.getLogger(__name__)	CODE
LOW	gptme/__init__.py	13	__all__ = [	CODE
LOW	gptme/telemetry.py	32	__all__ = [	CODE
LOW	gptme/telemetry.py	49	logger = logging.getLogger(__name__)	CODE
LOW	gptme/message.py	30	logger = logging.getLogger(__name__)	CODE
LOW	gptme/message.py	44	def set_output_format(fmt: str) -> None:	CODE
LOW	gptme/status_provider.py	31	logger = logging.getLogger(__name__)	CODE
LOW	gptme/sandbox.py	47	logger = logging.getLogger(__name__)	CODE
LOW	gptme/chat.py	51	logger = logging.getLogger(__name__)	CODE
LOW	gptme/circuit_breaker.py	38	logger = logging.getLogger(__name__)	CODE
LOW	gptme/circuit_breaker.py	286	__all__ = [	CODE
LOW	gptme/k_best.py	38	logger = logging.getLogger(__name__)	CODE
LOW	gptme/init.py	36	logger = logging.getLogger(__name__)	CODE
LOW	gptme/init.py	412	def _set_console_compact(enabled: bool) -> None:	CODE
LOW	gptme/profiles.py	41	logger = logging.getLogger(__name__)	CODE
LOW	gptme/executor.py	25	logger = logging.getLogger(__name__)	CODE
LOW	gptme/workspace_snapshot.py	30	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/_browser_lynx.py	12	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/shell_validation.py	24	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/shell_flags.py	34	__all__ = ["FlagSpec", "PERMITTED_FLAGS", "flags_permitted"]	CODE
LOW	gptme/tools/_browser_perplexity.py	9	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/clarify.py	23	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/lessons.py	34	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/gh.py	47	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/convert.py	31	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/_browser_playwright.py	39	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/shell.py	118	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/shell.py	1040	def set_workspace_cwd(cwd: str) -> None:	CODE
LOW	gptme/tools/shell.py	1075	def set_shell(shell: ShellSession) -> None:	CODE
LOW	gptme/tools/tmux.py	31	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/__init__.py	42	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/__init__.py	45	__all__ = [	CODE
LOW⚡	gptme/tools/__init__.py	85	def _set_available_tools_cache(tools: list[ToolSpec] \| None) -> None:	CODE
LOW	gptme/tools/__init__.py	494	def set_tools(tools: list[ToolSpec]) -> None:	CODE
LOW	gptme/tools/computer.py	117	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/_browser_thread.py	16	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/_browser_thread.py	44	def set_storage_state_override(path: Path \| None) -> None:	CODE
LOW	gptme/tools/elicit.py	34	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/browser.py	389	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/autocommit.py	33	logger = logging.getLogger(__name__)	CODE
LOW	gptme/tools/autocommit.py	176	__all__ = ["tool"]	STRING
237 more matches not shown…

Redundant / Tautological Comments186 hits · 283 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_dspy_basic.py	13	# Check if DSPy is available and handle import errors gracefully	COMMENT
LOW	tests/conftest.py	287	# Check if Anthropic quota is exhausted to skip API tests gracefully.	COMMENT
LOW⚡	tests/test_chat_history.py	23	# Check if feature is enabled	COMMENT
LOW	tests/test_hybrid_lessons.py	12	# Check if hybrid matching is available	COMMENT
LOW	tests/test_server_v2_hooks.py	76	# Check if test SESSION_START hook message is in the log	COMMENT
LOW	tests/test_server_v2_hooks.py	125	# Check if test STEP_PRE hook message is in the log	COMMENT
LOW	tests/test_server_v2_hooks.py	228	# Check if test TURN_POST hook message is in the log	COMMENT
LOW	tests/test_tools_computer.py	38	# Check if xrandr is available and can run successfully	COMMENT
LOW	tests/test_auto_compact.py	362	# Check if original had <think>, if so, compacted should too	COMMENT
LOW⚡	tests/test_tools_morph.py	242	# Set cwd to a subdirectory so the file path escapes it	COMMENT
LOW	tests/test_cli.py	1338	# Set timeout to 5 minutes to avoid Anthropic's streaming recommendation	COMMENT
LOW	tests/test_dspy_hybrid.py	5	# Check if DSPy is available and handle import errors gracefully	COMMENT
LOW	tests/test_tools_shell.py	1265	# Check if stdin is /dev/null	COMMENT
LOW	tests/test_tools_shell.py	1315	# Check if stdin is /dev/null	COMMENT
LOW	scripts/github_bot.py	454	# Check if there are changes	COMMENT
LOW	scripts/check_rst_formatting.py	65	# Check if we're exiting a code block (line with same or less indentation that has content)	COMMENT
LOW	scripts/check_rst_formatting.py	99	# Check if this is the start of a new list (not continuing an existing one)	COMMENT
LOW	scripts/check_rst_formatting.py	106	# Check if there's a blank line between this and the parent list item	COMMENT
LOW	scripts/check_rst_formatting.py	119	# Check if bullet list needs blank line before it	COMMENT
LOW	scripts/check_rst_formatting.py	128	# Check if there's no blank line before this list	COMMENT
LOW	scripts/check_poetry_lock_sync.sh	13	# Check if poetry.lock is in sync with pyproject.toml	COMMENT
LOW	scripts/demo_capture.py	109	# Check if playwright is importable	COMMENT
LOW	scripts/demo_capture.py	375	# Check if server is already running	COMMENT
LOW	scripts/demo_capture.py	418	# Check if already running	COMMENT
LOW	scripts/demo_capture.py	616	# Print results	COMMENT
LOW	scripts/build_server_executable.sh	6	# Check if we're in the project root	COMMENT
LOW	scripts/build_server_executable.sh	24	# Check if build was successful	COMMENT
LOW	webui/server.py	36	# Check if this looks like an asset request	COMMENT
LOW	gptme/chat.py	453	# Check if user declined execution - return to prompt without generating response	COMMENT
LOW	gptme/chat.py	490	# Check if there are any runnable tools left	COMMENT
LOW	gptme/chat.py	520	# Check if there's an interrupt or decline message after the last assistant message	COMMENT
LOW	gptme/chat.py	549	# Check if we should prompt for input or generate response directly	COMMENT
LOW	gptme/codeblock.py	261	# Check if this line starts with backticks (potential opening or closing)	COMMENT
LOW	gptme/codeblock.py	265	# Check if this is a bare fence (only backticks on the line)	COMMENT
LOW	gptme/codeblock.py	330	# Check if this line is a bare fence (only backticks)	COMMENT
LOW	gptme/codeblock.py	336	# Check if there's content after it (allowing blank lines)	COMMENT
LOW	gptme/init.py	162	# Check if model has provider/model format	COMMENT
LOW	gptme/init.py	165	# Check if it's a built-in provider or custom provider	COMMENT
LOW	gptme/info.py	260	# Check if editable via direct_url.json	COMMENT
LOW	gptme/info.py	280	# Check if installed via pipx (lives in ~/.local/pipx/)	COMMENT
LOW	gptme/info.py	298	# Check if any of the indicator packages are installed	COMMENT
LOW	gptme/tools/save.py	160	# Check if file exists and store original content for comparison	COMMENT
LOW	gptme/tools/save.py	165	# Check if folder exists	COMMENT
LOW	gptme/tools/save.py	191	# Check if this was an inefficient overwrite (minimal changes)	COMMENT
LOW	gptme/tools/save.py	244	# Check if folder exists first	COMMENT
LOW	gptme/tools/shell_validation.py	245	# Check if remaining text is the delimiter. Include the	COMMENT
LOW	gptme/tools/shell_validation.py	252	# Check if the line from pos to newline_idx is just the delimiter.	COMMENT
LOW	gptme/tools/shell_validation.py	284	# Check if this pipe is inside quotes	COMMENT
LOW	gptme/tools/shell_validation.py	286	# Check if this is part of \|\| (logical OR)	COMMENT
LOW	gptme/tools/shell_validation.py	560	# Check if all commands in the pipeline are allowlisted	COMMENT
LOW	gptme/tools/shell_validation.py	620	# Check if the match is within a safe region (quoted or heredoc)	COMMENT
LOW	gptme/tools/shell_validation.py	680	# Check if command is allowlisted	COMMENT
LOW	gptme/tools/shell_validation.py	707	# Check if disabled via environment variable	COMMENT
LOW	gptme/tools/shell_validation.py	711	# Check if shellcheck is available	COMMENT
LOW	gptme/tools/shell_validation.py	780	# Check if any triggered codes are error codes (should block)	COMMENT
LOW	gptme/tools/gh.py	244	# Check if all checks are done	COMMENT
LOW	gptme/tools/mcp_adapter.py	203	# Check if the tool has inputSchema with properties	COMMENT
LOW	gptme/tools/mcp_adapter.py	444	# Check if server already loaded	COMMENT
LOW	gptme/tools/mcp_adapter.py	448	# Check if server is in config	COMMENT
LOW	gptme/tools/shell.py	424	# Check if -S or -n flags are present (they disable TTY need)	COMMENT
126 more matches not shown…

Docstring Block Structure51 hits · 255 pts

Severity	File	Line	Snippet	Context
HIGH	tests/test_eval_behavioral_solutions.py	714	\ import re def parse_args(input_str): \"\"\"Parse a string of arguments respe	STRING
HIGH	tests/test_eval_behavioral_solutions.py	1041	\ \"\"\"API client for external service.\"\"\" import urllib.request import json	STRING
HIGH	tests/test_eval_behavioral.py	2732	\ import re def parse_args(input_str): """Parse a shell-like argument string into a list of tokens. Handles q	STRING
HIGH	gptme/model_attestation.py	225	Convenience constructor for creating a selection trace. Args: requested_model: The model string as requeste	STRING
HIGH	gptme/session.py	81	Create a new session. Args: session_id: Unique session identifier log: Optional LogMana	STRING
HIGH	gptme/message.py	298	Concatenate two messages of the same role. Merges content with separator, and combines files and file_hashes.	STRING
HIGH	gptme/sandbox.py	517	Return path to the cached CPython WASI binary, downloading if needed. Args: path: Override cache path. Defa	STRING
HIGH	gptme/k_best.py	77	Decorator factory: run the wrapped function K times and return the best result. Args: k: Number of candidat	STRING
HIGH	gptme/tools/_browser_playwright.py	739	Load a previously saved browser session (cookies, localStorage) from a file. This is the in-session complement to `	STRING
HIGH	gptme/tools/_browser_playwright.py	891	Press a keyboard key or shortcut in the current browser page. Dispatches the key event to the active focused elemen	STRING
HIGH	gptme/tools/computer.py	765	Return a text dump of the AT-SPI2 accessibility tree for all desktop apps. Requires the optional ``pyatspi`` packag	STRING
HIGH	gptme/tools/computer.py	843	Find an accessible element by role and name and return its center coordinates. Looks up the element via AT-SPI2, co	STRING
HIGH	gptme/tools/computer.py	939	Return a text dump of the macOS accessibility tree for visible apps. Uses AppleScript via ``osascript`` to query th	STRING
HIGH	gptme/tools/computer.py	1043	Find a UI element on macOS by AX role and name and return its center coordinates. Searches the frontmost applicatio	STRING
HIGH	gptme/tools/computer.py	2431	Start recording the screen to an MP4 file. Uses ``ffmpeg`` with ``x11grab`` (Linux) or ``avfoundation`` (macOS).	STRING
HIGH	gptme/tools/computer.py	2514	Record the screen for a fixed duration and return the output path. Synchronous wrapper around ``start_recording()``	STRING
HIGH	gptme/tools/browser.py	189	Convert PDF pages to images using auto-detected CLI tools. Auto-detects and uses the first available tool: pdftoppm	STRING
HIGH	gptme/tools/browser.py	1020	Load a previously saved browser session (cookies, localStorage) from a file. In-session complement to ``save_browse	STRING
HIGH	gptme/tools/browser.py	1130	Press a keyboard key or shortcut in the current browser page. Dispatches the key event to the focused element (or d	STRING
HIGH	gptme/tools/computer_semantic.py	125	Compare pre/post screenshots and optionally apply a semantic expectation check. Pixel-change detection runs uncondi	STRING
HIGH	gptme/tools/screenshot.py	63	Validate that screenshot path is within allowed directory. Security: Prevents arbitrary file writes via path traver	STRING
HIGH	gptme/tools/base.py	356	Tool specification. Defines a tool that can be used by the agent. Args: name: The name of the tool.	STRING
HIGH	gptme/tools/autocompact/hook.py	35	Get a unique name for the compacted conversation fork. The original conversation stays untouched as the backup	STRING
HIGH	gptme/tools/subagent/batch.py	392	Wait for the first subagent to complete and return its result. Useful for speculative/hedging patterns: spawn N	STRING
HIGH	gptme/tools/subagent/api.py	1159	Inject a steering message into a running subagent's conversation. The message is queued via the subagent's logdir p	STRING
HIGH	gptme/tools/subagent/api.py	1523	Wait for the first of the given subagents to complete. Useful for speculative/hedging patterns: spawn N subagents a	STRING
HIGH	gptme/llm/__init__.py	1175	Get available models from a provider. Args: provider: The provider to get models from Returns:	STRING
HIGH	gptme/llm/utils.py	85	Convert Parameter objects to JSON Schema dictionary format. This utility converts a list of Parameter objects into	STRING
HIGH	gptme/oauth/openrouter.py	146	Run the OpenRouter PKCE OAuth flow and return a durable API key. Args: key_label: Label OpenRouter shows fo	STRING
HIGH	gptme/util/master_context.py	97	Recover truncated content from the master context. Args: logfile: Path to conversation.jsonl b	STRING
HIGH	gptme/util/git_worktree.py	130	Create a git worktree for isolated subagent execution. Args: repo_path: Path to the git repository root.	STRING
HIGH	gptme/util/content.py	2	Check if message content is a command (starts with / and first word has exactly one slash). Examples:	STRING
HIGH	gptme/util/keyword_matching.py	20	Convert a keyword (possibly with wildcards) to a compiled regex pattern. Wildcards: - '*' matches zero or more	STRING
HIGH	gptme/util/install.py	40	Get the install command for a package based on the current environment. Args: package: Package name (e	STRING
HIGH	gptme/util/tool_format.py	195	Format detailed tool information. Args: tool: The tool to format include_examples: Include example	STRING
HIGH	gptme/server/workspace_api.py	160	Safely resolve a path within a workspace. Args: workspace: Base workspace path path: Optional	STRING
HIGH	gptme/server/client.py	53	Create a new session for conversation via events endpoint. Args: conversation_id: The conversation	STRING
HIGH	gptme/server/client.py	104	Take a step in the conversation. Args: conversation_id: The conversation ID session_id:	STRING
HIGH	gptme/server/client.py	140	Stream events from conversation. Args: conversation_id: The conversation ID session_id:	STRING
HIGH	gptme/server/client.py	174	Interrupt the current generation or tool execution. Args: conversation_id: The conversation ID	STRING
HIGH	gptme/server/client.py	206	Confirm, edit, or skip a pending tool execution. Args: conversation_id: The conversation ID	STRING
HIGH	gptme/agent/workspace.py	61	Create an agent workspace by cloning a template repository. This is the recommended approach for creating agen	STRING
HIGH	gptme/mcp/client.py	348	Read a specific resource by URI. Args: uri: The URI of the resource to read. Returns:	STRING
HIGH	gptme/mcp/client.py	428	Get a specific prompt by name with optional arguments. Args: name: The name of the prompt to retrie	STRING
HIGH	gptme/logmanager/conversations.py	553	Delete a conversation by its ID. Args: conv_id: The conversation ID to delete Returns: Tr	STRING
HIGH	gptme/lessons/installer.py	510	Check if all skill dependencies are satisfied. Args: skill_names: Specific skills to check (None = check al	STRING
HIGH	gptme/lessons/parser.py	110	Convert a file glob pattern to likely keywords. Args: glob_pattern: File pattern like '*/.py' or 'src/api	STRING
HIGH	gptme/eval/leaderboard.py	1310	Generate a leaderboard from eval results. Args: results_dir: Path to eval_results directory with timestampe	STRING
HIGH	gptme/eval/dspy/hybrid_optimizer.py	382	Select appropriate optimization strategy based on task characteristics. This function implements automated opt	STRING
HIGH	gptme/eval/dspy/experiments.py	407	Run a complete prompt optimization experiment. Args: experiment_name: Name for the experiment	STRING
HIGH	gptme/eval/dspy/experiments.py	458	Quick test of prompt variations without full optimization. Args: prompt_variations: Dictionary of prom	STRING

Modern AI Meta-Vocabulary33 hits · 98 pts

Severity	File	Line	Snippet	Context
MEDIUM	README.md	83	- [🛡 Guardrails](#-guardrails)	CODE
MEDIUM	tests/test_replay.py	337	# Model with a 100-token context window	COMMENT
MEDIUM	tests/conftest.py	285	# Download the embedding model before running tests.	COMMENT
MEDIUM⚡	tests/test_cmd_service.py	224	# First scaffold a periodic agent (creates testagent.timer)	COMMENT
MEDIUM⚡	tests/test_cmd_service.py	967	# launchd allows (apostrophe), then scaffold into it.	COMMENT
MEDIUM	tests/test_cmd_service.py	659	# to ensure the actual warning path was exercised (not just the scaffold line).	COMMENT
MEDIUM⚡	tests/test_tools_rag.py	50	# Should yield nothing when RAG is not available	COMMENT
MEDIUM	tests/test_tools_rag.py	68	# Should yield nothing when RAG is disabled	COMMENT
MEDIUM⚡	tests/test_tools_token_awareness.py	80	# GPT-4 has a context window of 8192 or 128000 depending on version	COMMENT
MEDIUM⚡	tests/test_init.py	56	# init() — orchestration tests	COMMENT
MEDIUM	docs/howto/automate-task.md	56	## Set up a project scaffold	COMMENT
MEDIUM	gptme/constants.py	58	# This prevents unbounded memory usage and context window overflow	COMMENT
MEDIUM	gptme/tools/shell.py	1757	# starting background execution so guardrails can intercept commands	COMMENT
MEDIUM	gptme/tools/shell.py	1846	# TOOL_CONFIRM hooks (including third-party guardrails) see the complete	COMMENT
MEDIUM	gptme/tools/rag.py	21	workspace_only = true # Whether to only search in the workspace directory, or the whole RAG index	STRING
MEDIUM	gptme/tools/rag.py	22	paths = [] # List of paths to include in the RAG index. Has no effect if workspace_only is true.	STRING
MEDIUM	gptme/tools/rag.py	82	### When to use RAG	COMMENT
MEDIUM	gptme/tools/subagent/hooks.py	105	# immediately, without waiting for the agentic turn to finish naturally.	COMMENT
MEDIUM	gptme/llm/llm_openai_models.py	297	# GPT-5.3 Codex — top-tier agentic coding model	COMMENT
MEDIUM	gptme/llm/llm_openai_models.py	327	# GPT-5.2 Codex — agentic coding variant of 5.2	COMMENT
MEDIUM	gptme/llm/llm_openai_models.py	349	# GPT-5.1 Codex — agentic coding variant of 5.1	COMMENT
MEDIUM	gptme/util/ask_execute.py	129	# execution request: route it through the hook chain again so guardrails	COMMENT
MEDIUM	gptme/agent/cli.py	287	gptme service init # Lightweight alternative: minimal service scaffold	STRING
MEDIUM	gptme/cli/util.py	466	# Initialize RAG	COMMENT
MEDIUM	gptme/cli/util.py	487	# Initialize RAG	COMMENT
MEDIUM	gptme/cli/util.py	519	# Initialize RAG	COMMENT
MEDIUM	gptme/cli/setup.py	464	# Uncomment to enable RAG (Retrieval-Augmented Generation)	COMMENT
MEDIUM	gptme/cli/cmd_init.py	9	gptme init my-project # Create my-project/ with scaffold	STRING
MEDIUM	gptme/logmanager/manager.py	842	# Expired — drop from context window (not from the log on disk).	COMMENT
MEDIUM	gptme/logmanager/manager.py	972	# Enrich with enabled context enhancements (RAG, fresh context)	COMMENT
MEDIUM	gptme/prompts/__init__.py	477	# Sub-module imports for orchestration	COMMENT
MEDIUM	gptme/commands/llm.py	353	# Show context window utilization if model info is available	COMMENT
MEDIUM	gptme/eval/agents/swebench.py	83	# implementation. This is the orchestration skeleton from bjsi's	COMMENT

Verbosity Indicators39 hits · 73 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_server_v2.py	1657	# Step 1: list sessions (webui catalog fetch)	COMMENT
LOW⚡	tests/test_server_v2.py	1664	# Step 2: fetch that specific session by its ID (webui detail fetch)	COMMENT
LOW⚡	tests/test_snapshot_command.py	267	# Step 1: record state before attempt	COMMENT
LOW⚡	tests/test_snapshot_command.py	276	# Step 2: mutate workspace (simulate a failed attempt)	COMMENT
LOW⚡	tests/test_snapshot_command.py	280	# Step 3: restore to before-attempt	COMMENT
LOW	tests/test_server_v2_hooks.py	275	# We need to check if there's an endpoint for this or if it happens automatically	COMMENT
LOW⚡	tests/test_computer_use_integration.py	233	# Step 1: open the form page	COMMENT
LOW⚡	tests/test_computer_use_integration.py	237	# Step 2: fill message and author fields	COMMENT
LOW⚡	tests/test_computer_use_integration.py	241	# Step 3: click the submit button (waits for page load internally)	COMMENT
LOW⚡	tests/test_computer_use_integration.py	244	# Step 4: read the result page	COMMENT
LOW⚡	tests/test_computer_use_integration.py	604	# Step 1: open a page and save the session	COMMENT
LOW⚡	tests/test_computer_use_integration.py	609	# Step 2: reload the state in the same session	COMMENT
LOW⚡	tests/test_computer_use_integration.py	616	# Step 3: open the same page again — should work with restored state	COMMENT
LOW⚡	tests/test_chat.py	498	# Step 1: include_paths extracts the image path	COMMENT
LOW⚡	tests/test_chat.py	503	# Step 2: embed_attached_file_content preserves images	COMMENT
LOW⚡	tests/test_chat.py	509	# Step 3: msgs2dicts preserves files for provider processing	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	117	# Step 1: load the game	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	120	# Step 2: verify initial state	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	129	# Step 3: fire at the enemy — auto-aims so one Space wins	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	132	# Step 4: verify the milestone marker	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	202	# Step 1: load the game	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	205	# Step 2: verify initial state	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	211	# Step 3: click three ore nodes to gather 6 iron ore	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	216	# Step 4: wait for the craft button to become enabled (needs >= 5 ore)	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	219	# Step 5: craft an iron plate	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	222	# Step 6: verify the milestone marker	COMMENT
LOW	gptme/llm/llm_gptme.py	254	# Step 1: Request device authorization	COMMENT
LOW⚡	gptme/llm/llm_gptme.py	272	# Step 2: Display code to user	COMMENT
LOW⚡	gptme/llm/llm_gptme.py	276	# Step 3: Poll for token	COMMENT
LOW	gptme/cli/auth.py	81	# Step 1: Request device authorization	COMMENT
LOW	gptme/cli/auth.py	109	# Step 2: Show the user what to do	COMMENT
LOW⚡	gptme/cli/auth.py	125	# Step 3: Poll for token	COMMENT
LOW	gptme/cli/onboard.py	238	# Step 1: Detect providers	COMMENT
LOW⚡	gptme/cli/onboard.py	268	# Step 2: Select provider	COMMENT
LOW⚡	gptme/cli/onboard.py	278	# Step 3: Test connectivity	COMMENT
LOW	gptme/cli/onboard.py	292	# Step 4: Create config	COMMENT
LOW⚡	gptme/hooks/form_autodetect.py	204	# Step 1: Heuristic detection	COMMENT
LOW⚡	gptme/hooks/form_autodetect.py	210	# Step 2: LLM parsing	COMMENT
LOW⚡	gptme/hooks/form_autodetect.py	216	# Step 3: Create and yield form message	COMMENT

Over-Commented Block78 hits · 73 pts

Severity	File	Line	Snippet	Context
LOW	.pre-commit-config.yaml	41	#- repo: local	COMMENT
LOW	pyproject.toml	401	# Scope --timeout to the test function only, NOT the whole runtest protocol.	COMMENT
LOW	docker-compose.yml	1	# Two services are available:	COMMENT
LOW	docker-compose.yml	41	restart: unless-stopped	COMMENT
LOW	gptme.toml	1	prompt = "This is gptme"	COMMENT
LOW	gptme.toml	21	#paths = [	COMMENT
LOW	gptme.toml	41	#command = "npx"	COMMENT
LOW	tauri/src-tauri/src/lib.rs	61		COMMENT
LOW	tauri/src-tauri/src/lib.rs	701	// Two paths need coverage:	COMMENT
LOW	tauri/src-tauri/src/lib.rs	761	match child.kill() {	COMMENT
LOW	tests/conftest.py	461	yield	COMMENT
LOW	tests/thread_leak.py	41	DICT_RACE_MESSAGE = "dictionary changed size during iteration"	COMMENT
LOW	tests/data/gh-issue-list.txt	1	#301 fix: broken import in tools/shell.py open 2025-01-15	COMMENT
LOW	tests/data/gh-issue-list.txt	21	#321 fix: edge case with piped commands detection open 2025-01-28	COMMENT
LOW	scripts/check_model_freshness.py	41	"llm/models/resolution.py",	COMMENT
LOW	scripts/bump_version.sh	1	#!/bin/bash	COMMENT
LOW	scripts/publish_release.sh	1	#!/bin/bash	COMMENT
LOW	scripts/install.sh	1	#!/usr/bin/env sh	COMMENT
LOW	.github/workflows/release.yml	1	name: Release	COMMENT
LOW	.github/workflows/webui-visual-diff.yml	161	# Pinned to a full commit SHA (not the mutable `v2` tag) because this	COMMENT
LOW	.github/workflows/build.yml	141	labels: ${{ steps.meta-eval.outputs.labels }}	COMMENT
LOW	webui/.pre-commit-config.yaml	41	# types_or: [javascript, jsx, ts, tsx]	COMMENT
LOW	webui/eslint.config.js	61	// {	COMMENT
LOW	webui/wrangler.toml	1	# Cloudflare Pages configuration	COMMENT
LOW	webui/e2e/generation.spec.ts	1	import { test, expect, type Page } from '@playwright/test';	COMMENT
LOW	webui/e2e/generation.spec.ts	21	// chatConfig arrived. Fix: skeleton pill while loading (PR #3441).	COMMENT
LOW	webui/e2e/generation.spec.ts	41	// 1. UI: chat-input is enabled (server is reachable and a conversation can	COMMENT
LOW	webui/e2e/generation.spec.ts	81	// Wait for ConversationContent to actually mount. MainLayout returns null when	COMMENT
LOW	webui/e2e/performance.spec.ts	1	import { test, expect } from '@playwright/test';	COMMENT
LOW	webui/e2e/ui-stability.spec.ts	1	import { test, expect } from '@playwright/test';	COMMENT
LOW	webui/e2e/conversation.spec.ts	141		COMMENT
LOW	webui/src/stores/conversations.ts	61	needsInitialStep: boolean;	COMMENT
LOW	webui/src/utils/workspaceApi.ts	81	//	COMMENT
LOW	webui/src/utils/conversation.ts	1	// This file can be removed or repurposed for other conversation utilities	COMMENT
LOW	webui/src/utils/tauri.ts	1	// Detect if we're running in Tauri environment.	COMMENT
LOW	webui/src/components/ConversationContent.tsx	401	const count = visibleMessageIndicesRef.current.length;	COMMENT
LOW	webui/src/components/ChatInput.tsx	621	defaultModel: apiDefaultModel,	COMMENT
LOW	…/src/components/__tests__/ConversationContent.test.tsx	761	// by the scroll handler. The second rAF must detect the position drift	COMMENT
LOW	…/src/components/__tests__/ConversationContent.test.tsx	821	});	COMMENT
LOW	gptme/sandbox.py	241	# before we get here). The soft limit is then set to the same value and	COMMENT
LOW	gptme/codeblock.py	141	)	COMMENT
LOW	gptme/codeblock.py	221		COMMENT
LOW	gptme/codeblock.py	281	# Only treat it as a real blank if there's more content after.	COMMENT
LOW	gptme/tools/shell_validation.py	361	# predict at validation time, so any `..` in a path is treated as	COMMENT
LOW	gptme/tools/shell_validation.py	581	return False	COMMENT
LOW	gptme/tools/hashline_edit.py	81		COMMENT
LOW	gptme/tools/base.py	781	# message and stamp call_id plus tool provenance on it.	COMMENT
LOW	gptme/tools/subagent/execution.py	201	# with a fresh, empty contextvars context, so _loaded_tools_var here already	COMMENT
LOW	gptme/tools/subagent/execution.py	421	# Subagent._read_log() instead.	COMMENT
LOW	gptme/tools/subagent/execution.py	581	# Progress file for subprocess-mode progress delivery.	COMMENT
LOW	gptme/tools/subagent/batch.py	281	# collected in self.results. Workers will exit their poll loops within	COMMENT
LOW	gptme/tools/subagent/types.py	301	# behaviour: auto-remove if unchanged, preserve branch if changes exist.	COMMENT
LOW	gptme/llm/__init__.py	741	# Buffer chars for the current line before forwarding to on_token.	COMMENT
LOW	gptme/llm/__init__.py	821	just_closed_thinking = True	COMMENT
LOW	gptme/llm/__init__.py	881	# <think> tag characters from leaking to callers before are_thinking	COMMENT
LOW	gptme/llm/__init__.py	921		COMMENT
LOW	gptme/llm/models/data.py	61	"azure",	COMMENT
LOW	gptme/llm/models/data.py	381	"preferred_edit_format": "diff",	COMMENT
LOW	gptme/server/session_step.py	681		COMMENT
LOW	gptme/server/session_step.py	1061	# assistant message may contain multiple tool uses. After completing	COMMENT
18 more matches not shown…

Structural Annotation Overuse38 hits · 72 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_server_v2.py	1657	# Step 1: list sessions (webui catalog fetch)	COMMENT
LOW⚡	tests/test_server_v2.py	1664	# Step 2: fetch that specific session by its ID (webui detail fetch)	COMMENT
LOW⚡	tests/test_snapshot_command.py	267	# Step 1: record state before attempt	COMMENT
LOW⚡	tests/test_snapshot_command.py	276	# Step 2: mutate workspace (simulate a failed attempt)	COMMENT
LOW⚡	tests/test_snapshot_command.py	280	# Step 3: restore to before-attempt	COMMENT
LOW⚡	tests/test_computer_use_integration.py	233	# Step 1: open the form page	COMMENT
LOW⚡	tests/test_computer_use_integration.py	237	# Step 2: fill message and author fields	COMMENT
LOW⚡	tests/test_computer_use_integration.py	241	# Step 3: click the submit button (waits for page load internally)	COMMENT
LOW⚡	tests/test_computer_use_integration.py	244	# Step 4: read the result page	COMMENT
LOW⚡	tests/test_computer_use_integration.py	604	# Step 1: open a page and save the session	COMMENT
LOW⚡	tests/test_computer_use_integration.py	609	# Step 2: reload the state in the same session	COMMENT
LOW⚡	tests/test_computer_use_integration.py	616	# Step 3: open the same page again — should work with restored state	COMMENT
LOW⚡	tests/test_chat.py	498	# Step 1: include_paths extracts the image path	COMMENT
LOW⚡	tests/test_chat.py	503	# Step 2: embed_attached_file_content preserves images	COMMENT
LOW⚡	tests/test_chat.py	509	# Step 3: msgs2dicts preserves files for provider processing	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	117	# Step 1: load the game	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	120	# Step 2: verify initial state	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	129	# Step 3: fire at the enemy — auto-aims so one Space wins	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	132	# Step 4: verify the milestone marker	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	202	# Step 1: load the game	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	205	# Step 2: verify initial state	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	211	# Step 3: click three ore nodes to gather 6 iron ore	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	216	# Step 4: wait for the craft button to become enabled (needs >= 5 ore)	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	219	# Step 5: craft an iron plate	COMMENT
LOW⚡	tests/test_computer_milestone_integration.py	222	# Step 6: verify the milestone marker	COMMENT
LOW	gptme/llm/llm_gptme.py	254	# Step 1: Request device authorization	COMMENT
LOW⚡	gptme/llm/llm_gptme.py	272	# Step 2: Display code to user	COMMENT
LOW⚡	gptme/llm/llm_gptme.py	276	# Step 3: Poll for token	COMMENT
LOW	gptme/cli/auth.py	81	# Step 1: Request device authorization	COMMENT
LOW	gptme/cli/auth.py	109	# Step 2: Show the user what to do	COMMENT
LOW⚡	gptme/cli/auth.py	125	# Step 3: Poll for token	COMMENT
LOW	gptme/cli/onboard.py	238	# Step 1: Detect providers	COMMENT
LOW⚡	gptme/cli/onboard.py	268	# Step 2: Select provider	COMMENT
LOW⚡	gptme/cli/onboard.py	278	# Step 3: Test connectivity	COMMENT
LOW	gptme/cli/onboard.py	292	# Step 4: Create config	COMMENT
LOW⚡	gptme/hooks/form_autodetect.py	204	# Step 1: Heuristic detection	COMMENT
LOW⚡	gptme/hooks/form_autodetect.py	210	# Step 2: LLM parsing	COMMENT
LOW⚡	gptme/hooks/form_autodetect.py	216	# Step 3: Create and yield form message	COMMENT

AI Slop Vocabulary19 hits · 46 pts

Severity	File	Line	Snippet	Context
MEDIUM	pyproject.toml	102	swebench = {version = ">=1.0", optional=true} # for SWE-bench eval harness	CODE
MEDIUM	pyproject.toml	103	terminal-bench = {version = ">=0.1", optional=true, python = ">=3.12"} # for Terminal-Bench eval harness	CODE
MEDIUM⚡	tests/test_mcp_discovery.py	151	# We'll use a try-except to make the test more robust	COMMENT
LOW⚡	tests/test_reduce.py	150	# Should not raise ValueError, just return messages as-is with content preserved	COMMENT
MEDIUM	gptme/tools/autocompact/resume.py	177	resume_prompt = """Please create a comprehensive resume of this conversation that includes:	CODE
MEDIUM	gptme/tools/subagent/__init__.py	141	### ACP Mode (multi-harness support)	COMMENT
MEDIUM	gptme/tools/subagent/api.py	549	# ACP mode: multi-harness support via Agent Client Protocol	COMMENT
MEDIUM	gptme/llm/__init__.py	927	# TODO: make this more robust/general, maybe with a callback that runs on each char/chunk	COMMENT
LOW	gptme/util/reduce.py	124	# but if prev_len == tokens, we are not making progress, so just return the log as-is	COMMENT
MEDIUM	gptme/util/install.py	18	# Check for uvx (comprehensive cross-platform detection to avoid false positives)	COMMENT
MEDIUM	gptme/plugins/__init__.py	482	# Check for uvx (comprehensive cross-platform detection to avoid false positives)	COMMENT
LOW	gptme/server/openapi_docs.py	634	# For now, just use the first non-None type as 200 response	COMMENT
MEDIUM⚡	gptme/server/tasks_api.py	273	"""Get comprehensive task information with derived data."""	STRING
MEDIUM	gptme/eval/dspy/tasks.py	415	"""Create advanced multi-step tasks for comprehensive evaluation."""	STRING
MEDIUM	gptme/eval/dspy/experiments.py	29	"""Generate a comprehensive optimization report."""	STRING
MEDIUM	gptme/eval/swe_extra/swe_bench_test_spec.py	1	# edited from the original swebench/harness/test_spec.py	COMMENT
MEDIUM	gptme/eval/swebench/evaluate.py	246	# NOTE: This is NOT authoritative — use the official SWE-bench harness for	COMMENT
MEDIUM	gptme/eval/swebench/main.py	208	# Run the official harness for the first model's predictions	COMMENT
MEDIUM	gptme/eval/swebench/main.py	209	# (multi-model harness runs can be done by calling this multiple times)	COMMENT

Fake / Example Data36 hits · 36 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_server_path_traversal.py	291	fake_key = tmp_path / "id_rsa"	CODE
LOW⚡	tests/test_server_path_traversal.py	292	fake_key.write_text("PRIVATE KEY")	CODE
LOW⚡	tests/test_server_path_traversal.py	295	mock_config.user.avatar = str(fake_key)	CODE
LOW⚡	tests/test_server_path_traversal.py	343	fake_key = agent_dir / "id_rsa"	CODE
LOW⚡	tests/test_server_path_traversal.py	344	fake_key.write_text("PRIVATE KEY")	CODE
LOW⚡	tests/test_server_path_traversal.py	393	fake_key = agent_dir / "id_rsa"	CODE
LOW⚡	tests/test_server_path_traversal.py	394	fake_key.write_text("PRIVATE KEY")	CODE
LOW	tests/test_agent.py	687	monkeypatch.setenv("GIT_AUTHOR_EMAIL", "test@test.com")	CODE
LOW	tests/test_agent.py	689	monkeypatch.setenv("GIT_COMMITTER_EMAIL", "test@test.com")	CODE
LOW	tests/test_prompts.py	341	["git", "config", "user.email", "test@test.com"],	CODE
LOW	tests/test_prompts.py	617	["git", "config", "user.email", "test@test.com"],	CODE
LOW	tests/test_subagent_worktree_isolation.py	57	["git", "config", "user.email", "test@test.com"],	CODE
LOW	tests/test_agent_doctor.py	62	"GIT_AUTHOR_EMAIL": "test@test.com",	CODE
LOW	tests/test_agent_doctor.py	64	"GIT_COMMITTER_EMAIL": "test@test.com",	CODE
LOW⚡	tests/test_lessons_commands.py	160	assert "placeholder" in result.lower()	CODE
LOW	tests/test_lessons_commands.py	57	keywords=["patch", "placeholder", "error"],	CODE
LOW⚡	tests/test_util_cli.py	1035	"GIT_AUTHOR_EMAIL": "test@test.com",	CODE
LOW⚡	tests/test_util_cli.py	1037	"GIT_COMMITTER_EMAIL": "test@test.com",	CODE
LOW⚡	tests/test_util_cli.py	1043	["git", "config", "user.email", "test@test.com"],	CODE
LOW	tests/test_git_worktree.py	23	["git", "config", "user.email", "test@test.com"],	CODE
LOW	tests/test_config_encoding.py	99	config_file.write_text('[user]\nabout = "placeholder"\n', encoding="utf-8")	CODE
LOW⚡	tests/test_util_keyword_matching.py	251	text = "Lorem ipsum " * 100 + "specific keyword here" + " dolor sit" * 100	CODE
LOW	tests/test_util_review.py	2332	run(["git", "config", "user.email", "test@test.com"])	CODE
LOW	.github/workflows/docs.yml	39	echo -e '[prompt]\nabout_user = "testing"\nresponse_preference = "just testing"\n\n[env]\nOPENAI_API_KEY="sk-d	CODE
LOW	webui/src/democonversations.ts	152	content: `Stress message ${i + 1}: ${'Lorem ipsum dolor sit amet. '.repeat(3)}`,	CODE
LOW	webui/src/democonversations.ts	152	content: `Stress message ${i + 1}: ${'Lorem ipsum dolor sit amet. '.repeat(3)}`,	CODE
LOW	webui/src/components/__tests__/CommandPalette.test.tsx	678	expect(input).toHaveAttribute('placeholder', expect.stringContaining('command'));	CODE
LOW	gptme/tools/_browser_playwright.py	1062	fill_element("[name='email']", "user@example.com")	STRING
LOW	gptme/tools/computer.py	2068	computer("type", text="user@example.com")	CODE
LOW	gptme/tools/computer.py	2072	computer("type", text="user@example.com")	CODE
LOW	gptme/tools/browser.py	1253	fill_element("[name='email']", "user@example.com")	STRING
LOW	gptme/eval/suites/practical7.py	215	"phone": "555-1234",	CODE
LOW	gptme/eval/suites/practical7.py	217	"street": "123 Main St",	CODE
LOW	gptme/eval/suites/practical7.py	230	"street": "123 Main St",	CODE
LOW	gptme/eval/suites/behavioral/noisy_worktree_fix.py	106	assert validate_email("user@example.com") is True	CODE
LOW⚡	gptme/eval/suites/behavioral/add_docstrings.py	174	assert validate_email("user@example.com") == "user@example.com"	CODE

Dead Code15 hits · 30 pts

Severity	File	Line	Context
MEDIUM	tests/test_mcp_server.py	36	CODE
MEDIUM	tests/test_llm_utils.py	481	CODE
MEDIUM	tests/test_hooks.py	64	CODE
MEDIUM	tests/test_hooks.py	69	CODE
MEDIUM	tests/test_hooks.py	74	CODE
MEDIUM	tests/test_llm_openai_subscription.py	357	CODE
MEDIUM	tests/test_hooks_cwd_changed.py	87	CODE
MEDIUM	tests/test_hooks_cwd_changed.py	137	CODE
MEDIUM	gptme/hooks/aw_watcher_agent.py	158	CODE
MEDIUM	gptme/hooks/aw_watcher_agent.py	175	CODE
MEDIUM	gptme/hooks/aw_watcher_agent.py	195	CODE
MEDIUM	gptme/hooks/aw_watcher_agent.py	235	CODE
MEDIUM	gptme/hooks/cwd_changed.py	42	CODE
MEDIUM	gptme/hooks/auto_snapshots.py	253	CODE
MEDIUM	gptme/hooks/auto_snapshots.py	284	CODE

AI Structural Patterns29 hits · 24 pts

Severity	File	Line	Context
LOW	tests/test_util_cli_models.py	191	CODE
LOW	tests/test_agent.py	29	CODE
LOW	tests/test_chats_stats.py	52	CODE
LOW	tests/test_util_cli.py	839	CODE
LOW	tests/test_util_tool_format.py	28	CODE
LOW	tests/test_json_output.py	137	CODE
LOW	tests/test_json_output.py	652	CODE
LOW	tests/test_json_output.py	765	CODE
LOW	gptme/chat.py	55	CODE
LOW	gptme/tools/shell.py	1951	CODE
LOW	gptme/tools/chats.py	646	CODE
LOW	gptme/tools/base.py	401	CODE
LOW	gptme/tools/subagent/execution.py	155	CODE
LOW	gptme/tools/subagent/execution.py	889	CODE
LOW	gptme/tools/subagent/batch.py	491	CODE
LOW	gptme/tools/subagent/batch.py	600	CODE
LOW	gptme/tools/subagent/batch.py	1096	CODE
LOW	gptme/tools/subagent/api.py	98	CODE
LOW	gptme/llm/__init__.py	279	CODE
LOW	gptme/llm/__init__.py	690	CODE
LOW	gptme/llm/models/listing.py	284	CODE
LOW	gptme/util/reduce.py	632	CODE
LOW	gptme/config/cli_setup.py	101	CODE
LOW	gptme/cli/cmd_init.py	187	CODE
LOW	gptme/prompts/__init__.py	385	CODE
LOW	gptme/prompts/__init__.py	504	CODE
LOW	gptme/eval/main.py	542	CODE
LOW	gptme/eval/dspy/prompt_optimizer.py	226	CODE
LOW	gptme/eval/dspy/prompt_optimizer.py	278	CODE

Cross-File Repetition3 hits · 15 pts

Severity	File	Snippet	Context
HIGH	tests/test_eval_behavioral.py	\ def summarize(text): \"\"\"return a dict with word and line counts for the given text.\"\"\" if not text: return {"wor	STRING
HIGH	…vioral/minimal_feature_preserve_default_with_decoys.py	\ def summarize(text): \"\"\"return a dict with word and line counts for the given text.\"\"\" if not text: return {"wor	STRING
HIGH	…eval/suites/behavioral/add_feature_preserve_default.py	\ def summarize(text): \"\"\"return a dict with word and line counts for the given text.\"\"\" if not text: return {"wor	STRING

Overly Generic Function Names15 hits · 14 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_active_context.py	20	(tmp_path / "utils.py").write_text("def helper(): pass\n")	CODE
LOW	tests/test_active_context.py	138	assert "def helper(): pass" in content	CODE
LOW⚡	tests/test_eval_behavioral.py	231	def helper():	CODE
LOW⚡	tests/test_eval_behavioral.py	1665	def helper():	CODE
LOW⚡	tests/test_context.py	353	(subdir / "utils.py").write_text("def helper(): pass")	CODE
LOW⚡	tests/test_context.py	492	(subdir / "utils.py").write_text("def helper(): pass")	CODE
LOW⚡	tests/test_tools_base.py	276	def helper(name: str) -> bool:	CODE
LOW⚡	tests/test_tools_base.py	287	def helper(name: str) -> bool:	CODE
LOW	tests/test_tools_base.py	455	def helper(name: str) -> bool:	CODE
LOW	tests/test_tools_base.py	501	def helper(name: str) -> bool:	CODE
LOW	tests/test_tools_shell.py	388	def test_function(shell):	STRING
LOW	gptme/cli/cmd_computer.py	903	def run_task(task: str, timeout: int, model: str \| None, as_json: bool):	CODE
LOW	gptme/hooks/tests/test_active_context.py	28	(ws / "utils.py").write_text("def helper():\n return 42\n")	CODE
LOW	gptme/eval/dspy/tasks.py	584	"slow_processor.py": """def process_data(data):	STRING
LOW	gptme/eval/dspy/tasks.py	708	def process_data(size):	CODE

Synthetic Comment Markers2 hits · 12 pts

Severity	File	Line	Snippet	Context
HIGH	scripts/github_ci_self_heal.py	262	f"🤖 Auto-generated by gptme-bot self-heal ({model})\n"	CODE
HIGH	gptme/cli/onboard.py	201	# Generated by gptme onboard	COMMENT

Example Usage Blocks8 hits · 12 pts

Severity	File	Line	Snippet	Context
LOW	docker-compose.yml	56	# Usage:	COMMENT
LOW	tauri/scripts/build-appimage.sh	4	# Usage:	COMMENT
LOW	scripts/bump_version.sh	4	# Usage:	COMMENT
LOW	scripts/gh-pr-view-with-pr-comments.sh	9	# Example usage:	COMMENT
LOW	scripts/publish_release.sh	11	# Usage:	COMMENT
LOW	scripts/install.sh	4	# Usage:	COMMENT
LOW	gptme/lessons/installer.py	675	# Example usage	COMMENT
LOW	gptme/eval/dspy/experiments.py	498	# Example usage	COMMENT

TODO Padding6 hits · 12 pts

Severity	File	Line	Snippet	Context
LOW⚡	gptme/eval/suites/behavioral/implement_lru_cache.py	144	# TODO: implement LRU caching	COMMENT
LOW	gptme/eval/suites/behavioral/implement_memoization.py	91	# TODO: implement memoization	COMMENT
LOW⚡	…tme/eval/suites/behavioral/implement_priority_queue.py	97	# TODO: implement priority queue	COMMENT
LOW⚡	…tme/eval/suites/behavioral/implement_priority_queue.py	101	# TODO: implement	COMMENT
LOW⚡	…tme/eval/suites/behavioral/implement_priority_queue.py	111	# TODO: implement	COMMENT
LOW⚡	…tme/eval/suites/behavioral/implement_priority_queue.py	120	# TODO: implement	COMMENT

Slop Phrases3 hits · 8 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	tests/test_anti_slop.py	31	"Certainly! It's worth noting that we delve into a rich tapestry of ideas "	CODE
MEDIUM⚡	tests/test_anti_slop.py	51	"It's worth noting that we delve into a rich tapestry of ideas."	CODE
MEDIUM	tests/test_server_v2_tool_confirmation.py	45	"Done, let me know if you need anything else.",	CODE

Cross-Language Confusion (JS/TS)1 hit · 5 pts

Severity	File	Line	Snippet	Context
HIGH	webui/src/utils/__tests__/highlightUtils.test.ts	37	'// ... existing code ...\ndef updated_fn():\n return True\n// ... existing code ...';	CODE

Magic Placeholder Names1 hit · 5 pts

Severity	File	Line	Snippet	Context
HIGH	gptme/cli/util.py	1391	click.echo(f"Set it with: export {env_var}=your-api-key")	CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings