letta-ai/letta

37.6

Adjusted Score

37.6

Raw Score

100%

Time Factor

2026-07-03

Last Push

23.8K

Stars

Python

Language

263.2K

Lines of Code

1.1K

Files

6.8K

Pattern Hits

2026-07-14

Scan Date

0.27

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 4HIGH 292MEDIUM 1120LOW 5376

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 6792 distinct pattern matches across 24 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 Identifiers2976 hits · 2500 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_modal_sandbox_v2.py	18	def test_execute_tool_dynamic_success(self):	CODE
LOW	tests/test_modal_sandbox_v2.py	42	def test_execute_tool_dynamic_with_error(self):	STRING
LOW	tests/test_modal_sandbox_v2.py	92	def test_execute_with_stdout_capture(self):	STRING
LOW	tests/test_modal_sandbox_v2.py	124	async def test_register_and_get_deployment(self):	CODE
LOW	tests/test_modal_sandbox_v2.py	374	def test_version_hash_calculation(self, mock_tool, mock_user, mock_sandbox_config):	CODE
LOW	tests/test_modal_sandbox_v2.py	450	async def test_run_with_mocked_modal(self, mock_tool, mock_user, mock_sandbox_config):	CODE
LOW	tests/test_modal_sandbox_v2.py	537	def test_detect_async_function(self, mock_user):	CODE
LOW⚡	tests/test_context_window_calculator.py	16	def test_returns_none_for_missing_tag(self):	CODE
LOW⚡	tests/test_context_window_calculator.py	21	def test_returns_none_for_missing_opening_tag(self):	CODE
LOW⚡	tests/test_context_window_calculator.py	26	def test_returns_none_for_unclosed_tag(self):	CODE
LOW⚡	tests/test_context_window_calculator.py	31	def test_handles_multiline_content(self):	CODE
LOW⚡	tests/test_context_window_calculator.py	36	def test_handles_nested_content(self):	CODE
LOW⚡	tests/test_context_window_calculator.py	41	def test_handles_empty_content(self):	CODE
LOW⚡	tests/test_context_window_calculator.py	46	def test_extracts_first_occurrence_with_duplicate_tags(self):	CODE
LOW⚡	tests/test_context_window_calculator.py	56	def test_extracts_standard_agent_sections(self):	CODE
LOW	tests/test_context_window_calculator.py	106	def test_extracts_tool_usage_rules(self):	CODE
LOW	tests/test_context_window_calculator.py	181	def test_returns_correct_dict_keys(self):	CODE
LOW	tests/test_context_window_calculator.py	194	def test_no_base_instructions_tag_extracts_preamble(self):	CODE
LOW	tests/test_context_window_calculator.py	216	def test_nested_memory_filesystem_not_extracted_as_top_level(self):	CODE
LOW	tests/test_context_window_calculator.py	244	def test_top_level_memory_filesystem_outside_memory_blocks(self):	CODE
LOW	tests/test_context_window_calculator.py	266	def test_letta_code_agent_real_format(self):	CODE
LOW	tests/test_context_window_calculator.py	320	def test_git_enabled_agent_bare_file_blocks_captured_as_core_memory(self):	CODE
LOW	tests/test_context_window_calculator.py	354	def test_git_enabled_agent_no_bare_blocks(self):	CODE
LOW	tests/test_context_window_calculator.py	367	def test_extract_top_level_tag_dual_occurrence_nested_first(self):	CODE
LOW	tests/test_context_window_calculator.py	380	def test_extract_system_prompt_pure_text_no_tags(self):	CODE
LOW	tests/test_context_window_calculator.py	388	def test_git_backed_memory_without_memory_blocks_wrapper(self):	CODE
LOW	tests/test_context_window_calculator.py	478	async def test_calculate_context_window_standard_agent(self):	CODE
LOW	tests/test_context_window_calculator.py	519	async def test_calculate_context_window_skips_empty_sections(self):	CODE
LOW	tests/test_context_window_calculator.py	546	async def test_calculate_context_window_all_sections(self):	CODE
LOW	tests/test_context_window_calculator.py	598	async def test_calculate_context_window_git_enabled_agent(self):	CODE
LOW	tests/test_context_window_calculator.py	88	def test_extracts_git_enabled_agent_sections(self):	STRING
LOW	tests/test_context_window_calculator.py	121	def test_extracts_directories(self):	STRING
LOW	tests/test_context_window_calculator.py	141	def test_handles_react_agent_no_memory_blocks(self):	STRING
LOW	tests/test_context_window_calculator.py	157	def test_handles_all_sections_present(self):	STRING
LOW	tests/test_context_window_calculator.py	176	def test_handles_empty_string(self):	STRING
LOW	tests/test_context_window_calculator.py	423	def test_legacy_memory_blocks_wrapper(self):	STRING
LOW	tests/test_openai_ws_session.py	56	async def test_iteration_without_context_manager(self):	CODE
LOW	tests/test_openai_ws_session.py	77	async def test_aclose_is_idempotent(self):	CODE
LOW	tests/test_openai_ws_session.py	84	async def test_stream_after_close_raises(self):	CODE
LOW	tests/test_openai_ws_session.py	100	async def test_stream_responses_yields_events_and_stops_on_terminal(self):	CODE
LOW	tests/test_openai_ws_session.py	162	def test_letta_llm_stream_adapter_stores_flag(self):	CODE
LOW	tests/test_openai_ws_session.py	178	def test_default_flag_is_false(self):	CODE
LOW	tests/test_openai_ws_session.py	193	async def test_aclose_default_is_noop(self):	CODE
LOW⚡	tests/integration_test_turbopuffer.py	163	def test_should_use_tpuf_with_settings():	CODE
LOW⚡	tests/integration_test_turbopuffer.py	356	async def test_native_only_operations(server, default_user, sarah_agent, disable_turbopuffer):	CODE
LOW⚡	tests/integration_test_turbopuffer.py	1899	async def test_archive_namespace_tracking(server, default_user, enable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	190	async def test_archive_creation_with_tpuf_enabled(server, default_user, enable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	200	async def test_archive_creation_with_tpuf_disabled(server, default_user, disable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	211	async def test_dual_write_and_query_with_real_tpuf(server, default_user, sarah_agent, enable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	287	async def test_turbopuffer_metadata_attributes(default_user, enable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	390	async def test_hybrid_search_with_real_tpuf(default_user, enable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	496	async def test_tag_filtering_with_real_tpuf(default_user, enable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	623	async def test_temporal_filtering_with_real_tpuf(default_user, enable_turbopuffer):	CODE
LOW	tests/integration_test_turbopuffer.py	761	def test_should_use_tpuf_for_messages_settings():	CODE
LOW	tests/integration_test_turbopuffer.py	803	def test_message_text_extraction(server, default_user):	CODE
LOW	tests/integration_test_turbopuffer.py	946	async def test_message_embedding_without_config(server, default_user, sarah_agent, enable_message_embedding):	CODE
LOW	tests/integration_test_turbopuffer.py	984	async def test_generic_reciprocal_rank_fusion():	CODE
LOW	tests/integration_test_turbopuffer.py	1085	async def test_message_dual_write_with_real_tpuf(enable_message_embedding, default_user):	CODE
LOW	tests/integration_test_turbopuffer.py	1147	async def test_message_vector_search_with_real_tpuf(enable_message_embedding, default_user):	CODE
LOW	tests/integration_test_turbopuffer.py	1208	async def test_message_hybrid_search_with_real_tpuf(enable_message_embedding, default_user):	CODE
2916 more matches not shown…

Decorative Section Separators504 hits · 1736 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	conf.yaml	56	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	58	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	63	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	65	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	71	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	73	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	106	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	108	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	112	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	114	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	117	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	119	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	126	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	128	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	133	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	135	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	143	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	145	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	153	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	155	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	160	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	162	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	166	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	168	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	172	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	174	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	178	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	180	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	183	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	185	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	200	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	202	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	209	# =============================================================================	COMMENT
MEDIUM⚡	conf.yaml	211	# =============================================================================	COMMENT
MEDIUM	conf.yaml	12	# =============================================================================	COMMENT
MEDIUM	conf.yaml	14	# =============================================================================	COMMENT
MEDIUM	conf.yaml	27	# =============================================================================	COMMENT
MEDIUM	conf.yaml	29	# =============================================================================	COMMENT
MEDIUM	conf.yaml	92	# =============================================================================	COMMENT
MEDIUM	conf.yaml	94	# =============================================================================	COMMENT
MEDIUM	conf.yaml	231	# =============================================================================	COMMENT
MEDIUM	conf.yaml	233	# =============================================================================	COMMENT
MEDIUM	conf.yaml	334	# =============================================================================	COMMENT
MEDIUM	conf.yaml	336	# =============================================================================	COMMENT
MEDIUM	conf.yaml	370	# =============================================================================	COMMENT
MEDIUM	conf.yaml	372	# =============================================================================	COMMENT
MEDIUM⚡	tests/integration_test_client_side_tools.py	47	# ------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_client_side_tools.py	49	# ------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_client_side_tools.py	58	# ------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_client_side_tools.py	60	# ------------------------------	COMMENT
MEDIUM	tests/integration_test_client_side_tools.py	16	# ------------------------------	COMMENT
MEDIUM	tests/integration_test_client_side_tools.py	18	# ------------------------------	COMMENT
MEDIUM	tests/test_utils.py	70	# -----------------------------------------------------------------------	COMMENT
MEDIUM	tests/test_utils.py	77	# -----------------------------------------------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_send_message.py	37	# ------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_send_message.py	39	# ------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_send_message.py	1089	# ------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_send_message.py	1091	# ------------------------------	COMMENT
MEDIUM⚡	tests/integration_test_send_message.py	2579	# ============================	COMMENT
MEDIUM⚡	tests/integration_test_send_message.py	2581	# ============================	COMMENT
444 more matches not shown…

Self-Referential Comments436 hits · 1203 pts

Severity	File	Line	Snippet	Context
MEDIUM	tests/locust_test.py	21	# Create a user and get the token	COMMENT
MEDIUM	tests/test_modal_sandbox_v2.py	140	# Create a mock actor	COMMENT
MEDIUM	tests/test_modal_sandbox_v2.py	182	# Create a mock actor	COMMENT
MEDIUM	tests/test_modal_sandbox_v2.py	242	# Create a mock actor	COMMENT
MEDIUM	tests/test_modal_sandbox_v2.py	287	# Create a mock actor	COMMENT
MEDIUM	tests/test_modal_sandbox_v2.py	463	# Create a mock that has a remote attribute	COMMENT
MEDIUM	tests/test_openai_ws_session.py	200	# Create a concrete subclass just for testing	COMMENT
MEDIUM⚡	tests/integration_test_turbopuffer.py	1901	# Create an archive	COMMENT
MEDIUM	tests/integration_test_turbopuffer.py	214	# Create a TPUF-enabled archive	COMMENT
MEDIUM	tests/integration_test_turbopuffer.py	634	# Create a unique archive ID for this test	COMMENT
MEDIUM	tests/integration_test_turbopuffer.py	2172	# Create a conversation	COMMENT
MEDIUM	tests/integration_test_client_side_tools.py	222	# Create the server-side tool	STRING
MEDIUM⚡	tests/test_utils.py	708	# Create a dict that when serialized will exceed limit	COMMENT
MEDIUM	tests/test_secret.py	198	# Create a Secret from encrypted value	COMMENT
MEDIUM⚡	tests/integration_test_tool_execution_sandbox.py	251	# Create the SandboxConfigManager	COMMENT
MEDIUM⚡	tests/integration_test_tool_execution_sandbox.py	261	# Create the sandbox configuration	COMMENT
MEDIUM⚡	tests/integration_test_tool_execution_sandbox.py	409	# Create a tool that uses the injected client and agent_id to actually clear a memory block	COMMENT
MEDIUM	tests/integration_test_tool_execution_sandbox.py	456	# Create the tool	STRING
MEDIUM	tests/integration_test_tool_execution_sandbox.py	479	# Create the tool in the system	STRING
MEDIUM	tests/integration_test_tool_execution_sandbox.py	482	# Create an agent with a memory block	STRING
MEDIUM	tests/integration_test_tool_execution_sandbox.py	500	# Create the sandbox for the tool	STRING
MEDIUM⚡	tests/integration_test_async_tool_sandbox.py	321	# Create the SandboxConfigManager	COMMENT
MEDIUM⚡	tests/integration_test_async_tool_sandbox.py	331	# Create the sandbox configuration	COMMENT
MEDIUM	tests/integration_test_async_tool_sandbox.py	1246	# Create the tool with proper schema	STRING
MEDIUM	tests/integration_test_async_tool_sandbox.py	1282	# Create the sandbox and verify client injection is detected	STRING
MEDIUM	tests/integration_test_async_tool_sandbox.py	1343	# Create the sandbox and verify client injection is detected	STRING
MEDIUM	tests/integration_test_send_message.py	2385	# # Create a streaming request	COMMENT
MEDIUM	tests/integration_test_override_model.py	373	# Create a conversation	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	111	# Create a realistic-looking tool return with repeated data	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1387	# # Create a new run for the repaired agent	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1391	# # Create a new agent loop with the repaired agent state	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1451	# # Create an input message to trigger the agent	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1457	# # Create a proper run record in the database	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1466	# # Create the agent loop using LettaAgentV3	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1485	# # Create a mock response with the required structure	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1634	# # Create an input message to trigger the agent	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1640	# # Create a proper run record in the database	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1649	# # Create the agent loop using LettaAgentV3	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1666	# # Create a mock response with the required structure	COMMENT
MEDIUM	tests/integration_test_summarizer.py	333	# Create a small conversation with ~5 messages	COMMENT
MEDIUM	tests/integration_test_summarizer.py	396	# Create a large tool return	COMMENT
MEDIUM	tests/integration_test_summarizer.py	589	# # Create an extremely large tool return (100k chars)	COMMENT
MEDIUM	tests/integration_test_summarizer.py	694	# Create a conversation with enough messages to trigger summarization	COMMENT
MEDIUM	tests/integration_test_summarizer.py	792	# Create a conversation with enough messages to summarize	COMMENT
MEDIUM	tests/integration_test_summarizer.py	868	# Create a conversation with enough messages to summarize	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1048	# Create the V3 agent loop	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1127	# Create a real summarizer config using the default factory	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1229	# Create a real summarizer config using the default factory	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1343	# # Create a run for the agent using RunManager	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1347	# # Create the agent loop using LettaAgentV3	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1789	# Create a summarizer config with "all" mode	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1862	# Create a summarizer config with "self" mode	COMMENT
MEDIUM	tests/integration_test_summarizer.py	2232	# Create a conversation with enough messages to summarize	COMMENT
MEDIUM	tests/test_sonnet_nonnative_reasoning_buffering.py	153	# Create the stream	COMMENT
MEDIUM⚡	…/integration_test_typescript_tool_execution_sandbox.py	292	# Create a TypeScript tool that has 'agent_state' in its parameters	COMMENT
MEDIUM	…/integration_test_typescript_tool_execution_sandbox.py	435	# Create a tool that checks if the client variable exists	COMMENT
MEDIUM	…/integration_test_typescript_tool_execution_sandbox.py	480	# Create a tool that uses the 'lodash' npm package	COMMENT
MEDIUM	tests/test_agent_serialization_v2.py	337	# Create a mock MCP tool	COMMENT
MEDIUM	tests/test_internal_agents_count.py	46	# Create a simple AgentState-like object for tracking	COMMENT
MEDIUM	tests/test_letta_agent_batch.py	432	# Create the mock for results	COMMENT
376 more matches not shown…

Cross-File Repetition188 hits · 940 pts

Severity	File	Line	Snippet	Context
HIGH	tests/test_multi_agent.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/conftest.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/integration_test_tool_execution_sandbox.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/integration_test_async_tool_sandbox.py	0	fixture to create and return the default organization.	STRING
HIGH	…/integration_test_typescript_tool_execution_sandbox.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/test_embeddings.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/test_agent_serialization_v2.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/test_letta_agent_batch.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/integration_test_token_counters.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/integration_test_modal.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/test_managers.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/test_agent_serialization.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/test_server_providers.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/managers/test_provider_manager.py	0	fixture to create and return the default organization.	STRING
HIGH	.github/scripts/model-sweep/conftest.py	0	fixture to create and return the default organization.	STRING
HIGH	tests/test_multi_agent.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/conftest.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/integration_test_tool_execution_sandbox.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/integration_test_async_tool_sandbox.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	…/integration_test_typescript_tool_execution_sandbox.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/test_embeddings.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/test_agent_serialization_v2.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/test_letta_agent_batch.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/integration_test_token_counters.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/integration_test_modal.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/test_managers.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/test_agent_serialization.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/test_server_providers.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/managers/test_provider_manager.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	.github/scripts/model-sweep/conftest.py	0	fixture to create and return the default user within the default organization.	STRING
HIGH	tests/conftest.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_cancellation.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_async_tool_sandbox.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_send_message.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_override_model.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_multi_agent.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/test_long_running_agents.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_sleeptime_agent.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_send_message_v2.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_mcp.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/sdk/conftest.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/sdk/mcp_servers_test.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/test_agent_serialization.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/integration_test_builtin_tools.py	0	provides the url for the letta server. if letta_server_url is not set, starts the server in a background thread and poll	STRING
HIGH	tests/conftest.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_async_tool_sandbox.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_send_message.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_override_model.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_system_prompt_prefix_caching.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_multi_agent.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/test_long_running_agents.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_sleeptime_agent.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_builtin_tools.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/integration_test_mcp.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/sdk/conftest.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/sdk/mcp_servers_test.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	.github/scripts/model-sweep/conftest.py	0	creates and returns a synchronous letta rest client for testing.	STRING
HIGH	tests/conftest.py	0	fetches the current weather for a given location. parameters: location (str): the location to get the weather for. retur	STRING
HIGH	tests/integration_test_chat_completions.py	0	fetches the current weather for a given location. parameters: location (str): the location to get the weather for. retur	STRING
HIGH	tests/test_letta_agent_batch.py	0	fetches the current weather for a given location. parameters: location (str): the location to get the weather for. retur	STRING
128 more matches not shown…

Excessive Try-Catch Wrapping805 hits · 825 pts

Severity	File	Line	Snippet	Context
LOW	test_watchdog_hang.py	88	except Exception as e:	CODE
LOW⚡	tests/integration_test_turbopuffer.py	154	except Exception:	CODE
LOW⚡	tests/integration_test_turbopuffer.py	346	except Exception as e:	CODE
LOW⚡	tests/integration_test_turbopuffer.py	350	except Exception:	CODE
LOW⚡	tests/integration_test_turbopuffer.py	1894	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	48	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	490	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	618	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	757	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	1141	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	1202	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	1265	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	1337	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	2398	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	2480	except Exception:	CODE
LOW	tests/integration_test_turbopuffer.py	2536	except Exception:	CODE
LOW	tests/test_crypto_utils.py	92	except Exception as e:	CODE
LOW	tests/conftest.py	79	except Exception:	CODE
LOW	tests/conftest.py	85	except Exception:	CODE
LOW	tests/integration_test_tool_execution_sandbox.py	452	except Exception as e:	CODE
LOW	tests/integration_test_cancellation.py	124	except Exception:	CODE
LOW	tests/integration_test_async_tool_sandbox.py	360	except Exception as e:	CODE
LOW	tests/integration_test_async_tool_sandbox.py	1242	except Exception as e:	CODE
LOW	tests/integration_test_send_message.py	373	except Exception:	CODE
LOW	tests/integration_test_send_message.py	511	except Exception:	CODE
LOW	tests/integration_test_send_message.py	667	except Exception:	CODE
LOW	tests/integration_test_send_message.py	703	except Exception:	CODE
LOW	tests/integration_test_send_message.py	859	except Exception:	CODE
LOW	tests/integration_test_send_message.py	1200	except Exception as e:	CODE
LOW	tests/integration_test_send_message.py	2026	except Exception as e:	CODE
LOW	tests/integration_test_send_message.py	2202	except Exception as e:	CODE
LOW	tests/mcp_test.py	216	except Exception as e:	CODE
LOW	tests/mcp_test.py	257	except Exception as e:	CODE
LOW	tests/mcp_test.py	277	except Exception as e:	CODE
LOW	tests/integration_test_override_model.py	117	except Exception as e:	CODE
LOW⚡	tests/integration_test_override_model.py	141	except Exception as e:	CODE
LOW	tests/test_sonnet_nonnative_reasoning_buffering.py	87	except Exception:	CODE
LOW	tests/test_sonnet_nonnative_reasoning_buffering.py	213	except Exception as e:	CODE
LOW	tests/integration_test_system_prompt_prefix_caching.py	33	except Exception:	CODE
LOW	tests/test_provider_trace_agents.py	94	except Exception:	CODE
LOW	tests/test_provider_trace_agents.py	160	except Exception:	CODE
LOW	tests/test_internal_agents_count.py	59	except Exception:	CODE
LOW⚡	tests/integration_test_usage_tracking.py	126	except Exception as e:	CODE
LOW⚡	tests/integration_test_human_in_the_loop.py	381	except Exception:	CODE
LOW	tests/integration_test_human_in_the_loop.py	368	except Exception:	CODE
LOW	tests/integration_test_human_in_the_loop.py	508	except Exception as e:	CODE
LOW	tests/integration_test_human_in_the_loop.py	770	except Exception as e:	CODE
LOW	tests/integration_test_human_in_the_loop.py	1035	except Exception as e:	CODE
LOW⚡	tests/test_temporal_metrics_local.py	115	except Exception:	CODE
LOW	tests/integration_test_send_message_v2.py	138	except Exception:	CODE
LOW	tests/integration_test_send_message_v2.py	209	except Exception:	CODE
LOW	tests/integration_test_send_message_v2.py	262	except Exception:	CODE
LOW	tests/integration_test_chat_completions.py	183	except Exception as e:	CODE
LOW	tests/integration_test_modal.py	428	except Exception as e:	CODE
LOW	tests/test_managers.py	11220	except Exception:	CODE
LOW	tests/test_sources.py	1599	except Exception as e:	CODE
LOW	tests/test_tool_schema_parsing.py	121	except Exception as e:	CODE
LOW	tests/test_tool_schema_parsing.py	194	except Exception as e:	CODE
LOW	tests/test_tool_schema_parsing.py	315	except Exception as e:	CODE
LOW	tests/test_tool_schema_parsing.py	676	except Exception as e:	CODE
745 more matches not shown…

Docstring Block Structure82 hits · 410 pts

Severity	File	Line	Snippet	Context
HIGH	tests/conftest.py	242	Fetches the current weather for a given location. Args: location (str): The location to ge	STRING
HIGH	tests/test_letta_agent_batch.py	56	Fetches the current weather for a given location. Args: location: The location to get the	STRING
HIGH	tests/integration_test_chat_completions.py	71	Fetches the current weather for a given location. Args: location (str): The location to ge	STRING
HIGH	tests/performance_tests/test_agent_mass_update.py	97	Fetches the current weather for a given location. Parameters: location (str): The location	STRING
HIGH	.github/scripts/model-sweep/conftest.py	68	Fetches the current weather for a given location. Parameters: location (str): The location	STRING
HIGH	letta/local_llm/grammars/gbnf_grammar_generator.py	518	Generate GBNF Grammar from Pydantic Models. This method takes a list of Pydantic models and uses them to gener	STRING
HIGH	letta/server/rest_api/streaming_response.py	161	Wraps a stream generator to provide real-time run cancellation checking. This wrapper periodically checks for	STRING
HIGH	letta/schemas/message.py	943	Parse tool response JSON and extract message and status. Args: response_text: Raw JSON response tex	STRING
HIGH	letta/schemas/providers/chatgpt_oauth.py	197	Perform OAuth token refresh with OpenAI's token endpoint. Args: creds: Current credentials containi	STRING
HIGH	letta/functions/functions.py	328	Dynamically imports a function from a specified module. Args: module_name (str): The name of the modul	STRING
HIGH	letta/functions/functions.py	354	Dynamically loads a specific function from a module and generates its JSON schema. Args: module_name (	STRING
HIGH	letta/functions/typescript_parser.py	10	Derives the OpenAI JSON schema for a given TypeScript function source code. This parser extracts the function signa	STRING
HIGH	letta/functions/async_composio_toolset.py	44	Execute an action asynchronously using the Composio API Args: action (str): The name of th	STRING
HIGH	letta/functions/schema_generator.py	527	Extracts the 'Examples:' section from a Google-style docstring. Args: docstring (Optional[str]): The full d	STRING
HIGH	letta/functions/ast_parsers.py	56	Resolve a type annotation string into a Python type. Previously, primitive support for int, float, str, dict, l	STRING
HIGH	letta/functions/ast_parsers.py	91	Parse the source code to extract annotations for a given function name. Args: source_code (str): The P	STRING
HIGH	letta/functions/function_sets/files.py	11	Open one or more files and load their contents into files section in core memory. Maximum of 5 files can be opened simul	STRING
HIGH	letta/functions/function_sets/builtin.py	43	Search the web using Exa's AI-powered search engine and retrieve relevant content. Examples: web_search("T	STRING
HIGH	letta/functions/function_sets/base.py	23	Memory management tool with various sub-commands for memory block operations. Args: command (str): The	STRING
HIGH	letta/functions/function_sets/base.py	95	Search prior conversation history using hybrid search (text + semantic similarity). Args: query (Optio	STRING
HIGH	letta/functions/function_sets/base.py	165	Add information to long-term archival memory for later retrieval. Use this tool to store facts, knowledge, or	STRING
HIGH	letta/functions/function_sets/base.py	203	Search archival memory using semantic similarity to find relevant information. This tool searches your long-te	STRING
HIGH	letta/functions/function_sets/base.py	312	The memory_replace command allows you to replace a specific string in a memory block with a new string. This is use	STRING
HIGH	letta/functions/function_sets/base.py	392	The memory_insert command allows you to insert text at a specific location in a memory block. Args: la	STRING
HIGH	letta/orm/sqlite_functions.py	75	Validates and transforms embeddings to ensure correct dimensionality. Args: embedding: Input embedding	STRING
HIGH	letta/orm/sqlalchemy_base.py	441	The primary accessor for an ORM record. Async version of read method. Args: db_session: the database	STRING
HIGH	letta/orm/sqlalchemy_base.py	844	Get the count of rows that match the provided filters. Args: db_session: SQLAlchemy session	STRING
HIGH	letta/llm_api/llm_client.py	19	Create an LLM client based on the model endpoint type. Args: provider: The model endpoint	STRING
HIGH	letta/llm_api/chatgpt_oauth_client.py	106	Get the ChatGPT OAuth provider and credentials with automatic refresh if needed. Args: llm_config:	STRING
HIGH	letta/llm_api/anthropic_client.py	382	Sends a batch request to the Anthropic API using the provided agent messages and tools mappings. Args:	STRING
HIGH	letta/helpers/tpuf_client.py	803	Generic query execution for Turbopuffer. Args: namespace_name: Turbopuffer namespace to query	STRING
HIGH	letta/helpers/message_helper.py	16	Async helper to fetch image from URL without blocking the event loop. Retries once on timeout to handle transie	STRING
HIGH	letta/helpers/crypto_utils.py	105	Encrypt a string using AES-256-GCM (synchronous version). WARNING: This performs CPU-intensive PBKDF2	STRING
HIGH	letta/helpers/crypto_utils.py	154	Encrypt a string using AES-256-GCM (async version). Runs the CPU-intensive PBKDF2 key derivation in a	STRING
HIGH	letta/helpers/crypto_utils.py	203	Decrypt a string that was encrypted using AES-256-GCM (synchronous version). WARNING: This performs CP	STRING
HIGH	letta/helpers/crypto_utils.py	254	Decrypt a string that was encrypted using AES-256-GCM (async version). Runs the CPU-intensive PBKDF2 k	STRING
HIGH	letta/services/provider_manager.py	970	Get an LLMConfig from a model handle. Args: handle: The model handle to look up actor:	STRING
HIGH	letta/services/provider_manager.py	1097	Get an EmbeddingConfig from a model handle. Args: handle: The model handle to look up a	STRING
HIGH	letta/services/agent_generate_completion_manager.py	77	Generate a completion directly from the LLM provider using the agent's configuration. This method make	STRING
HIGH	letta/services/tool_manager.py	472	Bulk create or update multiple tools in a single database transaction. Uses optimized PostgreSQL bulk	STRING
HIGH	letta/services/tool_manager.py	1397	Search tools using Turbopuffer semantic search. Args: actor: User performing the search	STRING
HIGH	letta/services/block_manager.py	800	Bulk-update the `value` field for multiple blocks in one transaction. Args: updates: mappi	STRING
HIGH	letta/services/archive_manager.py	293	Create a passage in an archive. Args: archive_id: ID of the archive to add the passage to	STRING
HIGH	letta/services/archive_manager.py	383	Create multiple passages in an archive. Args: archive_id: ID of the archive to add the passages to	STRING
HIGH	letta/services/job_manager.py	329	Get messages associated with a job using cursor-based pagination. This is a wrapper around get_job_mess	STRING
HIGH	letta/services/job_manager.py	387	Get steps associated with a job using cursor-based pagination. This is a wrapper around get_job_message	STRING
HIGH	letta/services/job_manager.py	434	Verify that a job exists and the user has the required access. Args: session: The database	STRING
HIGH	letta/services/job_manager.py	553	Get all steps associated with a job. Args: job_id: The ID of the job to get steps for	STRING
HIGH	letta/services/agent_serialization_manager.py	383	Export agents and their related entities to AgentFileSchema format. Args: agent_ids: List	STRING
HIGH	letta/services/agent_serialization_manager.py	509	Import AgentFileSchema into the database. Args: schema: The agent file schema to import	STRING
HIGH	letta/services/agent_manager.py	1975	Lists all sources attached to an agent with pagination. Args: agent_id: ID of the agent to	STRING
HIGH	letta/services/agent_manager.py	2758	Attaches a tool to an agent. Args: agent_id: ID of the agent to attach the tool to.	STRING
HIGH	letta/services/agent_manager.py	2901	Attaches missing core file tools to an agent. Args: agent_state: The current agent state w	STRING
HIGH	letta/services/agent_manager.py	2962	Detach all core file tools from an agent. Args: agent_state: The current agent state with	STRING
HIGH	letta/services/message_manager.py	910	Most performant query to list messages by directly querying the Message table. This function filters b	STRING
HIGH	letta/services/message_manager.py	1276	Search messages across entire organization using Turbopuffer. Args: actor: User performing	STRING
HIGH	letta/services/step_manager.py	286	Update the transaction ID for a step. Args: actor: The user making the request step_id:	STRING
HIGH	letta/services/step_manager.py	340	Update the stop reason for a step. Args: actor: The user making the request step_id: Th	STRING
HIGH	letta/services/step_manager.py	378	Update a step with error information. Args: actor: The user making the request step_id:	STRING
HIGH	letta/services/step_manager.py	426	Update a step with success status and final usage statistics. Args: actor: The user making the requ	STRING
22 more matches not shown…

Deep Nesting449 hits · 360 pts

Severity	File	Line	Context
LOW	tests/test_modal_sandbox_v2.py	450	CODE
LOW	tests/conftest.py	24	CODE
LOW	tests/conftest.py	70	CODE
LOW	tests/integration_test_cancellation.py	91	CODE
LOW	tests/integration_test_async_tool_sandbox.py	68	CODE
LOW	tests/integration_test_send_message.py	215	CODE
LOW	tests/integration_test_send_message.py	786	CODE
LOW	tests/integration_test_send_message.py	895	CODE
LOW	tests/integration_test_send_message.py	983	CODE
LOW	tests/integration_test_send_message.py	1009	CODE
LOW	tests/integration_test_send_message.py	984	CODE
LOW	tests/mcp_test.py	240	CODE
LOW	tests/mcp_test.py	260	CODE
LOW	tests/mcp_test.py	280	CODE
LOW	tests/integration_test_override_model.py	43	CODE
LOW	tests/test_sonnet_nonnative_reasoning_buffering.py	134	CODE
LOW	tests/integration_test_agent_tool_graph.py	501	CODE
LOW	tests/test_letta_agent_batch.py	362	CODE
LOW	tests/test_letta_agent_batch.py	448	CODE
LOW	tests/test_letta_agent_batch.py	615	CODE
LOW	tests/test_letta_agent_batch.py	781	CODE
LOW	tests/integration_test_multi_agent.py	18	CODE
LOW	tests/integration_test_multi_agent.py	135	CODE
LOW	tests/integration_test_multi_agent.py	210	CODE
LOW	tests/integration_test_multi_agent.py	325	CODE
LOW	tests/test_sdk_client.py	1069	CODE
LOW	tests/test_sdk_client.py	1154	CODE
LOW	tests/test_sdk_client.py	1899	CODE
LOW	tests/test_sdk_client.py	2164	CODE
LOW	tests/test_long_running_agents.py	18	CODE
LOW	tests/integration_test_human_in_the_loop.py	1175	CODE
LOW	tests/test_file_processor.py	226	CODE
LOW	tests/integration_test_sleeptime_agent.py	16	CODE
LOW	tests/integration_test_send_message_v2.py	283	CODE
LOW	tests/integration_test_send_message_v2.py	528	CODE
LOW	tests/test_managers.py	5081	CODE
LOW	tests/test_tool_schema_parsing.py	202	CODE
LOW	tests/test_agent_serialization.py	39	CODE
LOW	tests/test_agent_serialization.py	268	CODE
LOW	tests/test_agent_serialization.py	486	CODE
LOW	tests/integration_test_modal_sandbox_v2.py	632	CODE
LOW	tests/integration_test_builtin_tools.py	24	CODE
LOW	tests/integration_test_builtin_tools.py	161	CODE
LOW	tests/integration_test_clickhouse_llm_traces.py	180	CODE
LOW	tests/integration_test_mcp.py	27	CODE
LOW	tests/test_prompt_caching.py	605	CODE
LOW	tests/test_prompt_caching.py	697	CODE
LOW	tests/test_prompt_caching.py	974	CODE
LOW	tests/performance_tests/test_agent_mass_creation.py	253	CODE
LOW	tests/managers/test_cancellation.py	1143	CODE
LOW	tests/managers/test_agent_manager.py	1876	CODE
LOW	tests/managers/test_agent_manager.py	1990	CODE
LOW	tests/managers/test_mcp_manager.py	238	CODE
LOW	tests/managers/test_tool_manager.py	1207	CODE
LOW	tests/sdk/conftest.py	13	CODE
LOW	tests/sdk/conftest.py	64	CODE
LOW	tests/sdk/conftest.py	111	CODE
LOW	tests/sdk/conftest.py	151	CODE
LOW	tests/sdk/mcp_servers_test.py	31	CODE
LOW	tests/sdk/search_test.py	294	CODE
389 more matches not shown…

Modern AI Meta-Vocabulary113 hits · 360 pts

Severity	File	Line	Snippet	Context
MEDIUM	tests/integration_test_turbopuffer.py	411	# Create simple embeddings for testing (normally you'd use a real embedding model)	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1319	# and memory blocks are very large, potentially consuming most of the context window.	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1323	# # Override context window to be small so we trigger summarization	COMMENT
MEDIUM⚡	tests/integration_test_summarizer.py	1394	# # Now the agent should be able to respond without context window errors	COMMENT
MEDIUM	tests/integration_test_summarizer.py	1060	# this model's context window so that we always trigger the	COMMENT
MEDIUM	tests/test_providers.py	160	# All MiniMax models have 200K context window	COMMENT
MEDIUM⚡	tests/test_providers.py	297	# Test both sync and async context window methods	COMMENT
MEDIUM⚡	tests/integration_test_system_prompt_prefix_caching.py	58	# Step 1: Get initial context window, system prompt, and human block value	COMMENT
MEDIUM⚡	tests/integration_test_system_prompt_prefix_caching.py	156	# Step 1: Get initial context window	COMMENT
MEDIUM⚡	tests/integration_test_system_prompt_prefix_caching.py	183	# Step 5: Get context window and verify core memory was recompiled	COMMENT
MEDIUM⚡	tests/test_sdk_client.py	660	# List groups (most agents won't have groups unless in a multi-agent setup)	COMMENT
MEDIUM⚡	tests/test_managers.py	901	# Non-production environment should include all multi-agent tools	COMMENT
MEDIUM⚡	tests/test_managers.py	918	# Production environment should exclude local-only multi-agent tools	COMMENT
MEDIUM⚡	tests/test_managers.py	922	# But should include standard multi-agent tools	COMMENT
MEDIUM⚡	tests/test_managers.py	927	# Non-production environment should include all multi-agent tools	COMMENT
MEDIUM⚡	tests/test_managers.py	936	# Upsert only multi-agent tools	COMMENT
MEDIUM⚡	tests/test_managers.py	941	# Should only have non-local multi-agent tools	COMMENT
MEDIUM⚡	tests/test_managers.py	946	# Should have all multi-agent tools	COMMENT
MEDIUM⚡	tests/test_managers.py	5136	# Adjust expected names for multi-agent tools in production	COMMENT
MEDIUM	tests/test_managers.py	1046	# Get context window and check for basic appearances	COMMENT
MEDIUM	tests/test_managers.py	1169	# test with small context window model (8k)	COMMENT
MEDIUM	tests/test_managers.py	1189	# test with medium context window model (32k)	COMMENT
MEDIUM	tests/test_managers.py	1209	# test with large context window model (128k)	COMMENT
MEDIUM	tests/test_sources.py	282	# Get agent state to retrieve file config defaults (dynamic based on model context window)	COMMENT
MEDIUM	tests/test_sources.py	352	# Get agent state to retrieve file config defaults (dynamic based on model context window)	COMMENT
MEDIUM⚡	tests/integration_test_conversations_sdk.py	537	# We use an extremely long message to trigger a context window error	COMMENT
MEDIUM⚡	tests/integration_test_conversations_sdk.py	538	very_long_message = "Hello " * 100000 # Very long message to exceed context window	CODE
MEDIUM⚡	tests/integration_test_conversations_sdk.py	548	pass # Expected to fail due to context window exceeded	CODE
MEDIUM	tests/manual_test_multi_agent_broadcast_large.py	58	# Create "manager" agent with multi-agent broadcast tool	COMMENT
MEDIUM	tests/test_client.py	754	# get the sleeptime agent from the multi-agent group	COMMENT
MEDIUM	tests/test_server_providers.py	497	assert model_check.max_context_window == 64000 # Org model's context window	CODE
MEDIUM	tests/test_server_providers.py	1683	assert model.max_context_window == 16384 # Org model's context window	CODE
MEDIUM⚡	tests/managers/test_agent_manager.py	473	# Non-production environment should include all multi-agent tools	COMMENT
MEDIUM⚡	tests/managers/test_agent_manager.py	490	# Production environment should exclude local-only multi-agent tools	COMMENT
MEDIUM⚡	tests/managers/test_agent_manager.py	494	# But should include standard multi-agent tools	COMMENT
MEDIUM⚡	tests/managers/test_agent_manager.py	499	# Non-production environment should include all multi-agent tools	COMMENT
MEDIUM⚡	tests/managers/test_agent_manager.py	508	# Upsert only multi-agent tools	COMMENT
MEDIUM⚡	tests/managers/test_agent_manager.py	513	# Should only have non-local multi-agent tools	COMMENT
MEDIUM⚡	tests/managers/test_agent_manager.py	518	# Should have all multi-agent tools	COMMENT
MEDIUM	tests/managers/test_agent_manager.py	610	# Get context window and check for basic appearances	COMMENT
MEDIUM	tests/managers/test_agent_manager.py	1056	# test with small context window model (8k)	COMMENT
MEDIUM	tests/managers/test_agent_manager.py	1077	# test with medium context window model (32k)	COMMENT
MEDIUM	tests/managers/test_agent_manager.py	1098	# test with large context window model (128k)	COMMENT
MEDIUM⚡	tests/managers/test_tool_manager.py	1262	# Adjust expected names for multi-agent tools in production	COMMENT
MEDIUM	tests/managers/test_provider_manager.py	385	# Test context window limit override would be done at server level	COMMENT
MEDIUM	.github/scripts/model-sweep/model_sweep.py	737	# pydantic prevents us for overriding the context window paramter in the passed LLMConfig	COMMENT
MEDIUM	…b/scripts/model-sweep/generate_model_sweep_markdown.py	327	# Get context window and last scanned time	COMMENT
MEDIUM	letta/constants.py	58	# String in the error message for when the context window is too large	COMMENT
MEDIUM	letta/constants.py	77	# context window size	COMMENT
MEDIUM	letta/constants.py	442	# Tool return truncation limit for LLM context window management	COMMENT
MEDIUM	letta/utils.py	1365	# If no context window info, use conservative defaults	COMMENT
MEDIUM	letta/utils.py	1368	# Define defaults based on context window ranges	COMMENT
MEDIUM	letta/agent.py	432	# check if we are going over the context window: this allows for articifial constraints	COMMENT
MEDIUM	letta/agent.py	637	# overflow the agent context window	COMMENT
MEDIUM	letta/agent.py	1279	# context window breakdown (in messages)	COMMENT
MEDIUM	letta/agent.py	1288	# context window breakdown (in tokens)	COMMENT
MEDIUM	letta/agent.py	1411	# context window breakdown (in messages)	COMMENT
MEDIUM	letta/agent.py	1420	# context window breakdown (in tokens)	COMMENT
MEDIUM	letta/agent.py	1569	# context window breakdown (in messages)	COMMENT
MEDIUM	letta/agent.py	1578	# context window breakdown (in tokens)	COMMENT
53 more matches not shown…

Redundant / Tautological Comments205 hits · 308 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/integration_test_turbopuffer.py	150	# Check if our message ID is in the results	COMMENT
LOW	tests/integration_test_turbopuffer.py	64	# Set environment to DEV for testing	COMMENT
LOW	tests/conftest.py	187	# Set environment to DEV for testing	COMMENT
LOW	tests/integration_test_send_message.py	378	# Check if next message is LettaStopReason to detect this case	COMMENT
LOW	tests/integration_test_send_message.py	831	# Check if there are tool calls in the response	COMMENT
LOW	tests/integration_test_send_message.py	905	# Check if chunks are raw SSE strings (from background streaming)	COMMENT
LOW	tests/integration_test_summarizer.py	1567	# # Check if content has system_alert type with the summary message (object form)	COMMENT
LOW	tests/integration_test_summarizer.py	1745	# # Check if content has system_alert type with the summary message (object form)	COMMENT
LOW	tests/test_sonnet_nonnative_reasoning_buffering.py	200	# Check if this is inner thoughts	COMMENT
LOW	tests/test_sonnet_nonnative_reasoning_buffering.py	251	# Check if we detected large bursts	COMMENT
LOW	tests/test_sdk_client.py	631	# Check if modify method exists	COMMENT
LOW⚡	tests/test_temporal_metrics_local.py	109	# Check if DogStatsD is listening	COMMENT
LOW	tests/integration_test_send_message_v2.py	752	# Check if tool calls are grouped in a single message (parallel) or separate messages (sequential)	COMMENT
LOW	tests/integration_test_send_message_v2.py	777	# Check if we have true parallel tool calling	COMMENT
LOW	tests/integration_test_send_message_v2.py	780	# Check if the single message contains multiple tool calls	COMMENT
LOW	tests/test_tool_schema_parsing.py	150	# Check if we expect the conversion to fail	COMMENT
LOW	tests/integration_test_multi_modal_tool_returns.py	398	# Check if this is a tool return message that might contain our image	COMMENT
LOW	tests/test_server_providers.py	2383	# Check if provider already exists	COMMENT
LOW	tests/test_server_providers.py	2633	# Set last_synced to indicate models are already synced	COMMENT
LOW	tests/test_prompt_caching.py	827	# Check if ANY of the follow-up messages had a cache hit	COMMENT
LOW	tests/test_prompt_caching.py	961	# Check if ANY of the follow-up messages had a cache hit	COMMENT
LOW	tests/managers/test_cancellation.py	1280	# Check if message has tool_returns (new format) or tool_return (old format)	COMMENT
LOW	tests/managers/test_agent_manager.py	2008	# Check if expected type is in the union	COMMENT
LOW	otel/start-otel-collector.sh	92	# Check if update is needed	COMMENT
LOW	sandbox/node_server.py	66	# Check if there was an error	COMMENT
LOW	.github/scripts/model-sweep/conftest.py	166	# Check if already running	COMMENT
LOW	…d06594144ef3_add_and_migrate_encrypted_columns_for_.py	37	# Check if encryption key is available	COMMENT
LOW	…rsions/8149a781ac1b_backfill_encrypted_columns_for_.py	27	# Check if encryption key is available	COMMENT
LOW	…c/versions/74e860718e0d_add_archival_memory_sharing.py	57	# Check if archives table already exists	COMMENT
LOW	…/versions/eff256d296cb_mcp_encrypted_data_migration.py	30	# Check if encryption key is available	COMMENT
LOW	letta/streaming_utils.py	287	# Check if this same chunk also contains the terminating quote (and optional delimiter)	COMMENT
LOW	letta/utils.py	1167	# Check if coro is an async generator instead of a coroutine	COMMENT
LOW	letta/agent.py	553	# Check if inner thoughts is in the function call arguments (possible apparently if you are using Azure)	COMMENT
LOW	letta/agent.py	1217	# Check if there's a summary message in the message queue	COMMENT
LOW	letta/agent.py	1350	# Check if there's a summary message in the message queue	COMMENT
LOW	letta/agent.py	1488	# Check if there's a summary message in the message queue	COMMENT
LOW	letta/local_llm/function_parser.py	46	# Check if it's a user message or system message	COMMENT
LOW	letta/local_llm/utils.py	50	# Check if the file exists	COMMENT
LOW⚡	letta/local_llm/settings/settings.py	29	# Check if settings_dir folder exists (if not, create it)	COMMENT
LOW	letta/local_llm/grammars/gbnf_grammar_generator.py	473	# Check if the field is optional (not required)	COMMENT
LOW	letta/local_llm/grammars/gbnf_grammar_generator.py	85	if not parts: # Check if the list is empty	CODE
LOW	letta/agents/letta_agent_v3.py	289	# Check if we should use SGLang native adapter for multi-turn RL training.	COMMENT
LOW	letta/agents/letta_agent_v3.py	358	# Check if step was cancelled - break out of the step loop	COMMENT
LOW	letta/agents/letta_agent_v3.py	424	# Set context_tokens to expose actual context window usage (vs accumulated prompt_tokens)	COMMENT
LOW	letta/agents/letta_agent_v3.py	502	# Check if we should use SGLang native adapter for multi-turn RL training	COMMENT
LOW	letta/agents/letta_agent_v3.py	615	# Check if step was cancelled - break out of the step loop	COMMENT
LOW	letta/agents/letta_agent_v3.py	696	# Set context_tokens to expose actual context window usage (vs accumulated prompt_tokens)	COMMENT
LOW	letta/agents/letta_agent_v3.py	1184	# Check if there's a fallback route for the current model	COMMENT
LOW	letta/agents/letta_agent_v3.py	1625	# Check if there are required-before-exit tools that haven't been called	COMMENT
LOW	letta/agents/letta_agent_v3.py	1999	# Check if the LLM hit max_tokens (finish_reason == "length")	COMMENT
LOW⚡	letta/server/server.py	1812	# Check if the server name is already in the config	COMMENT
LOW⚡	letta/server/server.py	1883	# Check if the server name is already in the config	COMMENT
LOW	letta/server/startup.sh	23	# Check if we're configured for external Redis	COMMENT
LOW	letta/server/startup.sh	38	# Check if we're configured for external Postgres	COMMENT
LOW	letta/server/ws_api/server.py	110	# Check if a port argument is provided	COMMENT
LOW	letta/server/ws_api/interface.py	37	if self.clients: # Check if there are any clients connected	CODE
LOW	letta/server/rest_api/streaming_response.py	132	# Check if there's anything left	COMMENT
LOW⚡	letta/server/rest_api/proxy_helpers.py	319	# Set max_tokens to budget_tokens + reasonable buffer for response	COMMENT
LOW	letta/server/rest_api/proxy_helpers.py	28	# Check if entire message is a policy spec (starts with tag)	COMMENT
LOW	letta/server/rest_api/proxy_helpers.py	33	# Check if policy spec is appended (with prefix)	COMMENT
145 more matches not shown…

Unused Imports320 hits · 291 pts

Severity	File	Line	Context
LOW	tests/test_openai_ws_session.py	5	CODE
LOW	tests/integration_test_tool_execution_sandbox.py	13	CODE
LOW	tests/integration_test_async_tool_sandbox.py	18	CODE
LOW	tests/integration_test_async_tool_sandbox.py	383	CODE
LOW	tests/integration_test_send_message.py	32	CODE
LOW	tests/integration_test_modal.py	14	CODE
LOW	…sions/b888f21b151f_add_vector_db_provider_to_source.py	26	CODE
LOW	…sions/b888f21b151f_add_vector_db_provider_to_source.py	26	CODE
LOW	letta/__init__.py	25	CODE
LOW	letta/__init__.py	26	CODE
LOW	letta/__init__.py	27	CODE
LOW	letta/__init__.py	28	CODE
LOW	letta/__init__.py	29	CODE
LOW	letta/__init__.py	30	CODE
LOW	letta/__init__.py	31	CODE
LOW	letta/__init__.py	31	CODE
LOW	letta/__init__.py	31	CODE
LOW	letta/__init__.py	32	CODE
LOW	letta/__init__.py	33	CODE
LOW	letta/__init__.py	34	CODE
LOW	letta/__init__.py	34	CODE
LOW	letta/__init__.py	34	CODE
LOW	letta/__init__.py	34	CODE
LOW	letta/__init__.py	34	CODE
LOW	letta/__init__.py	41	CODE
LOW	letta/__init__.py	42	CODE
LOW	letta/__init__.py	43	CODE
LOW	letta/__init__.py	44	CODE
LOW	letta/__init__.py	45	CODE
LOW	letta/__init__.py	46	CODE
LOW	letta/__init__.py	47	CODE
LOW	letta/__init__.py	19	CODE
LOW	letta/errors.py	7	CODE
LOW	letta/errors.py	8	CODE
LOW	letta/serialize_schemas/__init__.py	1	CODE
LOW	letta/agents/base_agent.py	18	CODE
LOW	letta/agents/agent_loop.py	12	CODE
LOW	letta/agents/letta_agent.py	9	CODE
LOW	letta/agents/letta_agent.py	51	CODE
LOW	letta/agents/base_agent_v2.py	14	CODE
LOW	letta/agents/base_agent_v2.py	14	CODE
LOW	letta/agents/base_agent_v2.py	15	CODE
LOW	letta/agents/voice_sleeptime_agent.py	4	CODE
LOW	letta/agents/letta_agent_v3.py	73	CODE
LOW	letta/server/rest_api/dependencies.py	13	CODE
LOW	letta/server/rest_api/routers/v1/tags.py	8	CODE
LOW	letta/server/rest_api/routers/v1/internal_agents.py	5	CODE
LOW	letta/server/rest_api/routers/v1/users.py	11	CODE
LOW	letta/server/rest_api/routers/v1/git_http.py	20	CODE
LOW	letta/server/rest_api/routers/v1/internal_runs.py	10	CODE
LOW	letta/server/rest_api/routers/v1/archives.py	12	CODE
LOW	letta/server/rest_api/routers/v1/identities.py	20	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	52	CODE
LOW	letta/server/rest_api/routers/v1/voice.py	9	CODE
LOW	letta/server/rest_api/routers/v1/jobs.py	9	CODE
LOW	letta/server/rest_api/routers/v1/providers.py	12	CODE
LOW	letta/server/rest_api/routers/v1/organizations.py	9	CODE
LOW	letta/server/rest_api/routers/v1/llms.py	10	CODE
LOW	letta/server/rest_api/routers/v1/internal_templates.py	10	CODE
LOW	letta/server/rest_api/routers/v1/runs.py	24	CODE
260 more matches not shown…

Verbosity Indicators128 hits · 185 pts

Severity	File	Line	Snippet	Context
LOW	tests/integration_test_client_side_tools.py	92	# Step 1: User asks for the secret code - agent should call the tool	COMMENT
LOW	tests/integration_test_client_side_tools.py	109	# Step 2: Provide the tool return (simulating client-side execution)	COMMENT
LOW	tests/integration_test_client_side_tools.py	152	# Step 3: Ask the agent what the secret code was (testing memory/context)	COMMENT
LOW	tests/integration_test_client_side_tools.py	173	# Step 4: Validate the full conversation history makes sense	COMMENT
LOW	tests/integration_test_client_side_tools.py	242	# Step 1: Call the tool WITH client_tools specified - should pause for approval	STRING
LOW	tests/integration_test_client_side_tools.py	261	# Step 2: Provide client-side tool return	STRING
LOW	tests/integration_test_client_side_tools.py	286	# Step 3: Verify the client value was used, not the server value	STRING
LOW	tests/integration_test_client_side_tools.py	306	# Step 4: Test that WITHOUT client_tools, server tool executes directly	STRING
LOW	tests/integration_test_client_side_tools.py	354	# Step 1: Trigger the client-side tool call	STRING
LOW	tests/integration_test_client_side_tools.py	366	# Step 2: Provide an error response	STRING
LOW⚡	tests/test_mcp_encryption.py	251	# Step 1: Create OAuth session (without tokens initially)	COMMENT
LOW	tests/test_mcp_encryption.py	264	# Step 2: Update session with tokens (simulating OAuth callback)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	58	# Step 1: Get initial context window, system prompt, and human block value	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	109	# Step 5: Verify system prompt still hasn't changed	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	114	# Step 6: Manually update a block via the API	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	156	# Step 1: Get initial context window	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	161	# Step 2: Get the human block	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	170	# Step 3: Update block via agent-specific endpoint (triggers recompilation)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	179	# Step 4: Verify the block was updated	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	183	# Step 5: Get context window and verify core memory was recompiled	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	192	# Step 6: System prompt template should NOT change (only memory content changes)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	205	# Step 1: Get initial system prompt	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	209	# Step 2: Manually update a block via the API	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	224	# Step 3: Reset messages (this should trigger system prompt rebuild)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	227	# Step 4: Verify system prompt HAS changed and includes the new memory	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	72	# Step 2: Tell the agent to update its memory using the memory tool	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	84	# Step 3: Verify block was modified but system prompt hasn't changed	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	97	# Step 4: Send another message to the agent	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	129	# Step 7: Send another message and verify system prompt still hasn't changed	COMMENT
LOW⚡	tests/test_sdk_client.py	367	# Step 0: create an agent with no tags	COMMENT
LOW⚡	tests/test_sdk_client.py	380	# Step 1: Add multiple tags to the agent	COMMENT
LOW⚡	tests/test_sdk_client.py	386	# Step 2: Retrieve tags for the agent and verify they match the added tags	COMMENT
LOW⚡	tests/test_sdk_client.py	392	# Step 3: Retrieve agents by each tag to ensure the agent is associated correctly	COMMENT
LOW⚡	tests/test_sdk_client.py	397	# Step 4: Delete a specific tag from the agent and verify its removal	COMMENT
LOW⚡	tests/test_sdk_client.py	406	# Step 5: Delete all remaining tags from the agent	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1351	# Step 1: Send message that triggers approval request	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1358	# Step 2: Start approval stream and wait for first chunk (sentinel approach)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1395	# Step 4: Accumulate remaining chunks (already collected above)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1397	# Step 5: Verify we got chunks AND a cancelled stop reason	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1405	# Step 6: Verify run status is cancelled	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1412	# Step 7: Verify that approval response message is persisted	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1418	# Step 8: Attempt retry with same response	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1438	# Step 9: Accumulate chunks	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1441	# Step 10: Verify we got chunks AND an end_turn stop reason	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1449	# Step 11: Verify keep-alive message was sent	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1471	# Step 1: Send message that triggers approval request	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1478	# Step 2: Start approval stream and wait for first chunk (sentinel approach)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1515	# Step 4: Verify we got cancelled	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1520	# Step 6: Verify tool return message is persisted	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1525	# Step 7: Call compact with mode='all' to evict all messages from context	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1531	# Step 8: Verify only system and summary messages remain in context (should be 2)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1538	# Step 9: Retry the original approval response - should succeed via idempotency check	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1558	# Step 10: Accumulate chunks	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1561	# Step 11: Verify we got chunks AND an end_turn stop reason (not an error)	COMMENT
LOW	tests/integration_test_human_in_the_loop.py	1379	# Step 3: Wait for first chunk before triggering cancellation	COMMENT
LOW	tests/integration_test_human_in_the_loop.py	1499	# Step 3: Wait for first chunk before triggering cancellation	COMMENT
LOW⚡	tests/test_managers.py	6569	# Step 1: Create block	COMMENT
LOW⚡	tests/test_managers.py	6575	# Step 2: Update to v2, checkpoint => seq=2	COMMENT
LOW⚡	tests/test_managers.py	6581	# Step 3: Update to v3, checkpoint => seq=3	COMMENT
LOW⚡	tests/test_managers.py	6732	# Step 1: create initial block => seq=1	COMMENT
68 more matches not shown…

Structural Annotation Overuse128 hits · 179 pts

Severity	File	Line	Snippet	Context
LOW	tests/integration_test_client_side_tools.py	92	# Step 1: User asks for the secret code - agent should call the tool	COMMENT
LOW	tests/integration_test_client_side_tools.py	109	# Step 2: Provide the tool return (simulating client-side execution)	COMMENT
LOW	tests/integration_test_client_side_tools.py	152	# Step 3: Ask the agent what the secret code was (testing memory/context)	COMMENT
LOW	tests/integration_test_client_side_tools.py	173	# Step 4: Validate the full conversation history makes sense	COMMENT
LOW	tests/integration_test_client_side_tools.py	242	# Step 1: Call the tool WITH client_tools specified - should pause for approval	STRING
LOW	tests/integration_test_client_side_tools.py	261	# Step 2: Provide client-side tool return	STRING
LOW	tests/integration_test_client_side_tools.py	286	# Step 3: Verify the client value was used, not the server value	STRING
LOW	tests/integration_test_client_side_tools.py	306	# Step 4: Test that WITHOUT client_tools, server tool executes directly	STRING
LOW	tests/integration_test_client_side_tools.py	354	# Step 1: Trigger the client-side tool call	STRING
LOW	tests/integration_test_client_side_tools.py	366	# Step 2: Provide an error response	STRING
LOW⚡	tests/test_mcp_encryption.py	251	# Step 1: Create OAuth session (without tokens initially)	COMMENT
LOW	tests/test_mcp_encryption.py	264	# Step 2: Update session with tokens (simulating OAuth callback)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	58	# Step 1: Get initial context window, system prompt, and human block value	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	109	# Step 5: Verify system prompt still hasn't changed	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	114	# Step 6: Manually update a block via the API	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	156	# Step 1: Get initial context window	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	161	# Step 2: Get the human block	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	170	# Step 3: Update block via agent-specific endpoint (triggers recompilation)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	179	# Step 4: Verify the block was updated	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	183	# Step 5: Get context window and verify core memory was recompiled	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	192	# Step 6: System prompt template should NOT change (only memory content changes)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	205	# Step 1: Get initial system prompt	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	209	# Step 2: Manually update a block via the API	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	224	# Step 3: Reset messages (this should trigger system prompt rebuild)	COMMENT
LOW⚡	tests/integration_test_system_prompt_prefix_caching.py	227	# Step 4: Verify system prompt HAS changed and includes the new memory	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	72	# Step 2: Tell the agent to update its memory using the memory tool	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	84	# Step 3: Verify block was modified but system prompt hasn't changed	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	97	# Step 4: Send another message to the agent	COMMENT
LOW	tests/integration_test_system_prompt_prefix_caching.py	129	# Step 7: Send another message and verify system prompt still hasn't changed	COMMENT
LOW⚡	tests/test_sdk_client.py	367	# Step 0: create an agent with no tags	COMMENT
LOW⚡	tests/test_sdk_client.py	380	# Step 1: Add multiple tags to the agent	COMMENT
LOW⚡	tests/test_sdk_client.py	386	# Step 2: Retrieve tags for the agent and verify they match the added tags	COMMENT
LOW⚡	tests/test_sdk_client.py	392	# Step 3: Retrieve agents by each tag to ensure the agent is associated correctly	COMMENT
LOW⚡	tests/test_sdk_client.py	397	# Step 4: Delete a specific tag from the agent and verify its removal	COMMENT
LOW⚡	tests/test_sdk_client.py	406	# Step 5: Delete all remaining tags from the agent	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1351	# Step 1: Send message that triggers approval request	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1358	# Step 2: Start approval stream and wait for first chunk (sentinel approach)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1395	# Step 4: Accumulate remaining chunks (already collected above)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1397	# Step 5: Verify we got chunks AND a cancelled stop reason	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1405	# Step 6: Verify run status is cancelled	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1412	# Step 7: Verify that approval response message is persisted	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1418	# Step 8: Attempt retry with same response	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1438	# Step 9: Accumulate chunks	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1441	# Step 10: Verify we got chunks AND an end_turn stop reason	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1449	# Step 11: Verify keep-alive message was sent	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1471	# Step 1: Send message that triggers approval request	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1478	# Step 2: Start approval stream and wait for first chunk (sentinel approach)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1515	# Step 4: Verify we got cancelled	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1520	# Step 6: Verify tool return message is persisted	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1525	# Step 7: Call compact with mode='all' to evict all messages from context	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1531	# Step 8: Verify only system and summary messages remain in context (should be 2)	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1538	# Step 9: Retry the original approval response - should succeed via idempotency check	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1558	# Step 10: Accumulate chunks	COMMENT
LOW⚡	tests/integration_test_human_in_the_loop.py	1561	# Step 11: Verify we got chunks AND an end_turn stop reason (not an error)	COMMENT
LOW	tests/integration_test_human_in_the_loop.py	1379	# Step 3: Wait for first chunk before triggering cancellation	COMMENT
LOW	tests/integration_test_human_in_the_loop.py	1499	# Step 3: Wait for first chunk before triggering cancellation	COMMENT
LOW⚡	tests/test_managers.py	6569	# Step 1: Create block	COMMENT
LOW⚡	tests/test_managers.py	6575	# Step 2: Update to v2, checkpoint => seq=2	COMMENT
LOW⚡	tests/test_managers.py	6581	# Step 3: Update to v3, checkpoint => seq=3	COMMENT
LOW⚡	tests/test_managers.py	6732	# Step 1: create initial block => seq=1	COMMENT
68 more matches not shown…

AI Slop Vocabulary48 hits · 109 pts

Severity	File	Line	Snippet	Context
MEDIUM⚡	tests/integration_test_summarizer.py	1307	### in order to leverage caching (for self compaction)	COMMENT
MEDIUM⚡	tests/test_agent_serialization_v2.py	772	"""Test file export functionality with comprehensive validation"""	STRING
MEDIUM⚡	tests/integration_test_usage_tracking.py	449	# Use Anthropic Sonnet 4.5 for this test as it has the most comprehensive caching	COMMENT
MEDIUM⚡	tests/test_managers.py	1279	# Create an agent using the comprehensive fixture.	COMMENT
MEDIUM⚡	tests/test_managers.py	1297	# Create an agent using the comprehensive fixture.	COMMENT
MEDIUM	tests/test_managers.py	3612	"""Test comprehensive tag functionality for passages."""	STRING
MEDIUM	tests/managers/conftest.py	364	"""Create a comprehensive test agent with all features."""	STRING
MEDIUM	tests/managers/test_passage_manager.py	815	"""Test comprehensive tag functionality for passages."""	STRING
MEDIUM⚡	tests/managers/test_agent_manager.py	1210	# Create an agent using the comprehensive fixture.	COMMENT
MEDIUM⚡	tests/managers/test_agent_manager.py	1228	# Create an agent using the comprehensive fixture.	COMMENT
MEDIUM	…empty_archival/agent_state/2024-01-11_12_44_32_PM.json	1	{"model": "gpt-4", "system": "You are Letta, the latest version of Limnal Corporation's digital companion, developed in	CODE
MEDIUM	…empty_archival/agent_state/2024-01-11_12_44_33_PM.json	1	{"model": "gpt-4", "system": "You are Letta, the latest version of Limnal Corporation's digital companion, developed in	CODE
MEDIUM	…nt_test_attach/agent_state/2024-01-11_12_42_19_PM.json	1	{"model": "gpt-4", "system": "You are Letta, the latest version of Limnal Corporation's digital companion, developed in	CODE
MEDIUM	…nt_test_attach/agent_state/2024-01-11_12_42_17_PM.json	1	{"model": "gpt-4", "system": "You are Letta, the latest version of Limnal Corporation's digital companion, developed in	CODE
MEDIUM	…nts/agent_test/agent_state/2024-01-11_12_43_59_PM.json	1	{"model": "gpt-4", "system": "You are Letta, the latest version of Limnal Corporation's digital companion, developed in	CODE
MEDIUM	…nts/agent_test/agent_state/2024-01-11_12_43_57_PM.json	1	{"model": "gpt-4", "system": "You are Letta, the latest version of Limnal Corporation's digital companion, developed in	CODE
MEDIUM	letta/streaming_utils.py	268	# Not in message yet: accumulate until we see '<json_key>': (robust to split fragments)	COMMENT
MEDIUM⚡	letta/agent.py	1740	# If it exists in the input message, elevate it to the 'message' level	COMMENT
LOW	letta/agent.py	950	# Fallback if for some reason context_window is missing, just set to the default	COMMENT
LOW	letta/agent.py	1134	# Fallback if for some reason context_window is missing, just set to the default	COMMENT
MEDIUM	letta/agent.py	314	"""Get response from LLM API with robust retry mechanism."""	STRING
LOW	letta/serialize_schemas/marshmallow_agent.py	117	# no user messages, just return system message	COMMENT
LOW	letta/serialize_schemas/marshmallow_agent.py	152	# no user messages, just return system message	COMMENT
MEDIUM	letta/agents/letta_agent_v3.py	1238	# NOTE: we no longer refresh the system prompt before compaction so we can leverage cach	COMMENT
MEDIUM	letta/agents/letta_agent_v3.py	1461	# NOTE: we no longer refresh the system prompt before compaction so we can leverage cache for self m	COMMENT
MEDIUM	letta/server/rest_api/interface.py	307	# To make streamed arguments robust, we add a JSON-aware incremental	COMMENT
LOW	letta/server/rest_api/interface.py	1120	# Example where we just pass through the raw stream from the underlying OpenAI SSE stream	COMMENT
LOW	letta/server/rest_api/auth_token.py	15	# user is admin so we just return the default uuid	COMMENT
LOW	letta/server/rest_api/routers/v1/tools.py	557	# TODO: don't do this in the future (just return MCPServer)	COMMENT
LOW	letta/server/rest_api/routers/v1/tools.py	612	# TODO: don't do this in the future (just return MCPServer)	COMMENT
LOW	letta/schemas/sandbox_config.py	45	# If `data` is not a dict (e.g., it's another Pydantic model), just return it	COMMENT
LOW	letta/schemas/sandbox_config.py	73	# If `data` is not a dict (e.g., it's another Pydantic model), just return it	COMMENT
LOW	letta/schemas/message.py	641	# TODO remove the cast and just return the native type	COMMENT
LOW	letta/schemas/message.py	728	# If assistant mode is off, just create one ToolCallMessage with all tool calls	COMMENT
LOW⚡	letta/schemas/providers/anthropic.py	138	# just use a cheap model to count some tokens - as of 5/7/2025 this is faster than fetching the list of mode	COMMENT
LOW	letta/schemas/providers/base.py	96	# Simplified asyncio handling - just use asyncio.run()	COMMENT
LOW	letta/schemas/providers/base.py	120	# Simplified asyncio handling - just use asyncio.run()	COMMENT
LOW	letta/functions/typescript_parser.py	180	# Handle union types (simplified - just use string)	COMMENT
LOW	letta/functions/mcp_client/types.py	94	# If the result still contains unreplaced templates, just return original value	COMMENT
MEDIUM	letta/orm/sqlalchemy_base.py	278	# TODO (cliandy): can make this more robust down the line	COMMENT
MEDIUM	letta/llm_api/openai_client.py	574	# don't raise error since this isn't robust against edge cases	COMMENT
LOW	letta/llm_api/openai_client.py	646	# For text or json_object, just pass the type	COMMENT
MEDIUM⚡	letta/services/mcp_manager.py	1130	# TODO: @jnjpng make this check more robust and remove direct os.getenv	COMMENT
MEDIUM⚡	letta/services/mcp_server_manager.py	1326	# TODO: @jnjpng make this check more robust and remove direct os.getenv	COMMENT
MEDIUM	letta/services/tool_sandbox/modal_sandbox_v2.py	418	# Start with a more robust base image with development tools	COMMENT
LOW	letta/interfaces/anthropic_streaming_interface.py	150	# This interface doesn't track cache tokens, so we just use the raw values	COMMENT
LOW	letta/interfaces/anthropic_streaming_interface.py	702	# This interface doesn't track cache tokens, so we just use the raw values	COMMENT
MEDIUM	letta/interfaces/openai_streaming_interface.py	476	# Minimal, robust extraction: only emit the value of "message".	COMMENT

Over-Commented Block161 hits · 108 pts

Severity	File	Line	Snippet	Context
LOW	conf.yaml	1	# Letta Configuration File	COMMENT
LOW	conf.yaml	161	# Letta Client (LETTA_DEFAULT_*)	COMMENT
LOW	conf.yaml	281	# api_key: ""	COMMENT
LOW	conf.yaml	301		COMMENT
LOW	conf.yaml	321	# sglang:	COMMENT
LOW	conf.yaml	341	# sandbox_template_id: ""	COMMENT
LOW	conf.yaml	361		COMMENT
LOW	conf.yaml	381	# health_metrics_enabled: false	COMMENT
LOW	conf.yaml	401	#	COMMENT
LOW	alembic.ini	1	# A generic, single database configuration.	COMMENT
LOW	alembic.ini	21	# Any required deps can installed by adding `alembic[tz]` to the pip requirements	COMMENT
LOW	alembic.ini	41	# The path separator used here should be the separator specified by "version_path_separator" below.	COMMENT
LOW	alembic.ini	61	# output_encoding = utf-8	COMMENT
LOW	tests/locust_test.py	81	# response = requests.post(f"{self.base_url}/api/agents/{agent_id}/messages", json=request.model_dump(), head	COMMENT
LOW	tests/locust_test.py	101	# try:	COMMENT
LOW	tests/test_crypto_utils.py	421	This simulates the real-world scenario of existing encrypted values	COMMENT
LOW	tests/integration_test_async_tool_sandbox.py	41		COMMENT
LOW	tests/integration_test_send_message.py	1901	assert_greeting_with_assistant_message_response(messages_from_db, model_handle, model_settings, from_db=True)	COMMENT
LOW	tests/integration_test_send_message.py	1921	# client.agents.update(agent_id=agent_state.id, model=model_handle, model_settings=model_settings)	COMMENT
LOW	tests/integration_test_send_message.py	2281	# # ids=[c.model for c in TESTED_LLM_CONFIGS],	COMMENT
LOW	tests/integration_test_send_message.py	2301	# # Verify the job was created	COMMENT
LOW	tests/integration_test_send_message.py	2321	# def test_job_cancellation_endpoint_validation(	COMMENT
LOW	tests/integration_test_send_message.py	2341	# disable_e2b_api_key: Any,	COMMENT
LOW	tests/integration_test_send_message.py	2361	# # Try to cancel the completed job - should fail	COMMENT
LOW	tests/integration_test_send_message.py	2381	# This verifies that jobs continue even if the client "disconnects" (simulated by not consuming the stream).	COMMENT
LOW	tests/integration_test_send_message.py	2401	# )	COMMENT
LOW	tests/integration_test_send_message.py	2561		COMMENT
LOW	tests/integration_test_summarizer.py	41		COMMENT
LOW	tests/integration_test_summarizer.py	581	# TESTED_LLM_CONFIGS,	COMMENT
LOW	tests/integration_test_summarizer.py	601	# content=[	COMMENT
LOW	tests/integration_test_summarizer.py	621	# ],	COMMENT
LOW	tests/integration_test_summarizer.py	641	# # Find tool return messages in the result and verify truncation occurred	COMMENT
LOW	tests/integration_test_summarizer.py	1301	)	COMMENT
LOW	tests/integration_test_summarizer.py	1321	# """	COMMENT
LOW	tests/integration_test_summarizer.py	1341	# agent_state = await server.agent_manager.create_agent_async(agent_create, actor=actor)	COMMENT
LOW	tests/integration_test_summarizer.py	1361	# )	COMMENT
LOW	tests/integration_test_summarizer.py	1381	# # Reload agent state after repairs	COMMENT
LOW	tests/integration_test_summarizer.py	1401	# # Verify we got a valid response after repair	COMMENT
LOW	tests/integration_test_summarizer.py	1421	# from unittest.mock import patch	COMMENT
LOW	tests/integration_test_summarizer.py	1441	# messages.append(	COMMENT
LOW	tests/integration_test_summarizer.py	1461	# id=test_run_id,	COMMENT
LOW	tests/integration_test_summarizer.py	1481	# from unittest.mock import MagicMock	COMMENT
LOW	tests/integration_test_summarizer.py	1501	# response=mock_response,	COMMENT
LOW	tests/integration_test_summarizer.py	1521	# stream_tokens=True,	COMMENT
LOW	tests/integration_test_summarizer.py	1541	# # Verify a summary message exists with the correct format	COMMENT
LOW	tests/integration_test_summarizer.py	1561	# if "prior messages" in text_to_check and "hidden" in text_to_check and "summary of the previous" in	COMMENT
LOW	tests/integration_test_summarizer.py	1581	# # Verify we attempted multiple invocations (the failing one + retry after summarization)	COMMENT
LOW	tests/integration_test_summarizer.py	1601	# Note: This test only runs with OpenAI since it uses OpenAI-specific error handling.	COMMENT
LOW	tests/integration_test_summarizer.py	1621	# content=[TextContent(type="text", text=f"User message {i}: This is test message number {i}.")],	COMMENT
LOW	tests/integration_test_summarizer.py	1641	# run_manager = RunManager()	COMMENT
LOW	tests/integration_test_summarizer.py	1661	# if call_count == 1:	COMMENT
LOW	tests/integration_test_summarizer.py	1681	# message="This model's maximum context length is 8000 tokens. However, your messages resulted in 12000 t	COMMENT
LOW	tests/integration_test_summarizer.py	1701	# run_id=test_run_id,	COMMENT
LOW	tests/integration_test_summarizer.py	1721	# # "prior messages ... have been hidden" and "summary of the previous"	COMMENT
LOW	tests/integration_test_summarizer.py	1741	# summary_message_text = parsed.get("message")	COMMENT
LOW	tests/test_providers.py	261	assert len(embedding_models) == 0 # embedding models currently not supported by SGLang	COMMENT
LOW	tests/integration_test_agent_tool_graph.py	861	# Check order	COMMENT
LOW	tests/integration_test_agent_tool_graph.py	881	# -- flip_coin	COMMENT
LOW	tests/integration_test_agent_tool_graph.py	901	# child_output_mapping={	COMMENT
LOW	tests/integration_test_agent_tool_graph.py	921	# found_secret_word = False	COMMENT
101 more matches not shown…

Cross-Language Confusion18 hits · 105 pts

Severity	File	Line	Snippet	Context
HIGH	…/integration_test_typescript_tool_execution_sandbox.py	595	assert "let client: any = null;" in script	CODE
HIGH	…/integration_test_typescript_tool_execution_sandbox.py	615	result, agent_state = parse_typescript_result('{"results": 42, "agent_state": null}')	CODE
HIGH	tests/managers/test_message_manager.py	1087	arguments='{"message": {"status": "success", "data": {"count": 42, "items": ["a", "b"]}, "meta": null}}'	CODE
HIGH	letta/streaming_utils.py	106	# numbers, null, or a nested object/array) streamed in next, the stream	COMMENT
HIGH	letta/local_llm/json_parser.py	63	- if (function call is not send_message && prev message['role'] == user): insert heartbeat	STRING
HIGH	letta/server/rest_api/routers/v1/groups.py	168	include_return_message_types: Optional[List[MessageType]] = Query(None, description="Message types to include in res	CODE
HIGH	letta/server/rest_api/routers/v1/agents.py	1601	include_return_message_types: Optional[List[MessageType]] = Query(None, description="Message types to include in res	CODE
HIGH	letta/server/rest_api/routers/v1/messages.py	46	None, description="Message types to include in response. When null, all message types are returned."	CODE
HIGH	letta/server/rest_api/routers/v1/conversations.py	214	include_return_message_types: Optional[List[MessageType]] = Query(None, description="Message types to include in res	CODE
HIGH	letta/schemas/providers.py	283	# "param": null,	COMMENT
HIGH	letta/schemas/providers/openai.py	184	# "param": null,	COMMENT
HIGH	letta/functions/schema_validator.py	173	# They represent union types (e.g., string \| null)	COMMENT
HIGH	letta/functions/helpers.py	288	# If we have exactly one type and null, make it Optional	COMMENT
HIGH	letta/llm_api/helpers.py	39	# If it's simple primitives only (string, null, integer, boolean, etc), flatten to type array	COMMENT
HIGH	letta/services/mcp_server_manager.py	9	from sqlalchemy import delete, desc, null, select	CODE
HIGH	letta/services/tool_sandbox/typescript_generator.py	140	"let client: any = null;",	CODE
HIGH	letta/services/tool_sandbox/typescript_generator.py	209	lines.append("const _output = { results: _result, agent_state: null };")	CODE
HIGH	letta/services/provider_trace_backends/clickhouse.py	96	# OpenAI Responses API returns {"error": null} on success	COMMENT

AI Structural Patterns126 hits · 97 pts

Severity	File	Line	Context
LOW	tests/test_agent_serialization_v2.py	184	CODE
LOW	tests/sdk/conftest.py	256	CODE
LOW	letta/database_utils.py	43	CODE
LOW	letta/agent.py	300	CODE
LOW	letta/agent.py	857	CODE
LOW	letta/local_llm/chat_completion_proxy.py	33	CODE
LOW	letta/local_llm/grammars/gbnf_grammar_generator.py	1014	CODE
LOW	letta/local_llm/grammars/gbnf_grammar_generator.py	1050	CODE
LOW	…m/llm_chat_completion_wrappers/configurable_wrapper.py	11	CODE
LOW	letta/local_llm/llm_chat_completion_wrappers/chatml.py	32	CODE
LOW	letta/local_llm/llm_chat_completion_wrappers/llama3.py	31	CODE
LOW	letta/agents/letta_agent.py	83	CODE
LOW	letta/agents/letta_agent.py	1714	CODE
LOW	letta/agents/base_agent_v2.py	51	CODE
LOW	letta/agents/base_agent_v2.py	76	CODE
LOW	letta/agents/letta_agent_v2.py	191	CODE
LOW	letta/agents/letta_agent_v2.py	300	CODE
LOW	letta/agents/letta_agent_v2.py	1062	CODE
LOW	letta/agents/letta_agent_v3.py	222	CODE
LOW	letta/agents/letta_agent_v3.py	444	CODE
LOW	letta/agents/letta_agent_v3.py	895	CODE
LOW	letta/agents/letta_agent_v3.py	1595	CODE
LOW	letta/agents/letta_agent_v3.py	2077	CODE
LOW	letta/server/server.py	894	CODE
LOW	letta/server/server.py	936	CODE
LOW	letta/server/server.py	986	CODE
LOW	letta/server/rest_api/utils.py	382	CODE
LOW	letta/server/rest_api/dependencies.py	38	CODE
LOW	letta/server/rest_api/routers/v1/steps.py	22	CODE
LOW	letta/server/rest_api/routers/v1/tags.py	15	CODE
LOW	letta/server/rest_api/routers/v1/tools.py	63	CODE
LOW	letta/server/rest_api/routers/v1/tools.py	171	CODE
LOW	letta/server/rest_api/routers/v1/tools.py	476	CODE
LOW	letta/server/rest_api/routers/v1/internal_runs.py	30	CODE
LOW	letta/server/rest_api/routers/v1/archives.py	91	CODE
LOW	letta/server/rest_api/routers/v1/groups.py	18	CODE
LOW	letta/server/rest_api/routers/v1/groups.py	150	CODE
LOW	letta/server/rest_api/routers/v1/identities.py	26	CODE
LOW	letta/server/rest_api/routers/v1/identities.py	158	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	125	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	218	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	296	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	415	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	465	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	1142	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	1510	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	1579	CODE
LOW	letta/server/rest_api/routers/v1/agents.py	2348	CODE
LOW	letta/server/rest_api/routers/v1/mcp_servers.py	212	CODE
LOW	letta/server/rest_api/routers/v1/jobs.py	17	CODE
LOW	letta/server/rest_api/routers/v1/providers.py	18	CODE
LOW	letta/server/rest_api/routers/v1/messages.py	31	CODE
LOW	letta/server/rest_api/routers/v1/messages.py	211	CODE
LOW	letta/server/rest_api/routers/v1/internal_blocks.py	27	CODE
LOW	letta/server/rest_api/routers/v1/internal_blocks.py	125	CODE
LOW	letta/server/rest_api/routers/v1/folders.py	119	CODE
LOW	letta/server/rest_api/routers/v1/folders.py	441	CODE
LOW	letta/server/rest_api/routers/v1/blocks.py	28	CODE
LOW	letta/server/rest_api/routers/v1/blocks.py	112	CODE
LOW	letta/server/rest_api/routers/v1/blocks.py	192	CODE
66 more matches not shown…

Modern Structural Boilerplate51 hits · 59 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/integration_test_send_message.py	35	logger = logging.getLogger(__name__)	CODE
LOW	tests/mcp_test.py	36	async def set_tokens(self, tokens: OAuthToken) -> None:	CODE
LOW	tests/mcp_test.py	42	async def set_client_info(self, client_info: OAuthClientInformationFull) -> None:	CODE
LOW	tests/integration_test_override_model.py	24	logger = logging.getLogger(__name__)	CODE
LOW⚡	tests/integration_test_usage_tracking.py	32	logger = logging.getLogger(__name__)	CODE
LOW⚡	tests/integration_test_human_in_the_loop.py	12	logger = logging.getLogger(__name__)	CODE
LOW⚡	tests/integration_test_send_message_v2.py	27	logger = logging.getLogger(__name__)	CODE
LOW	tests/integration_test_conversations_sdk.py	15	logger = logging.getLogger(__name__)	CODE
LOW⚡	tests/test_prompt_caching.py	28	logger = logging.getLogger(__name__)	CODE
LOW	tests/helpers/endpoints_helper.py	12	logger = logging.getLogger(__name__)	CODE
LOW	letta/log_context.py	7	def set_log_context(key: str, value: Any) -> None:	CODE
LOW	letta/log_context.py	24	def update_log_context(**kwargs: Any) -> None:	CODE
LOW	letta/types/__init__.py	7	__all__ = ["JsonDict", "JsonValue"]	CODE
LOW⚡	letta/agents/letta_agent.py	894	async def _update_agent_last_run_metrics(self, completion_time: datetime, duration_ms: float) -> None:	CODE
LOW⚡	letta/agents/letta_agent_v2.py	1464	async def _update_agent_last_run_metrics(self, completion_time: datetime, duration_ms: float) -> None:	CODE
LOW	letta/agents/letta_agent_batch.py	386	async def _update_request_statuses_async(self, updates: List[RequestStatusUpdateInfo]) -> None:	CODE
LOW	letta/server/rest_api/routers/v1/git_http.py	54	def set_server_instance(server) -> None:	CODE
LOW	letta/server/rest_api/middleware/__init__.py	5	__all__ = ["CheckPasswordMiddleware", "LoggingMiddleware", "RequestIdMiddleware"]	CODE
LOW	letta/otel/db_pool_monitoring.py	37	def _setup_pool_listeners(self, pool: Pool, engine_name: str, pool_mode: str) -> None:	CODE
LOW	letta/otel/sqlalchemy_instrumentation_integration.py	21	logger = logging.getLogger(__name__)	CODE
LOW	letta/schemas/tool_rule.py	10	logger = logging.getLogger(__name__)	CODE
LOW	letta/schemas/mcp.py	8	logger = logging.getLogger(__name__)	CODE
LOW⚡	letta/schemas/mcp.py	94	def set_token_secret(self, secret: Secret) -> None:	CODE
LOW⚡	letta/schemas/mcp.py	98	def set_custom_headers_secret(self, secret: Secret) -> None:	CODE
LOW⚡	letta/schemas/mcp_server.py	207	def set_access_token_secret(self, secret: Secret) -> None:	CODE
LOW⚡	letta/schemas/mcp_server.py	211	def set_refresh_token_secret(self, secret: Secret) -> None:	CODE
LOW⚡	letta/schemas/mcp_server.py	215	def set_client_secret_secret(self, secret: Secret) -> None:	CODE
LOW⚡	letta/schemas/mcp_server.py	219	def set_authorization_code_secret(self, secret: Secret) -> None:	CODE
LOW⚡	letta/schemas/providers/chatgpt_oauth.py	264	async def _update_stored_credentials(self, creds: ChatGPTOAuthCredentials, actor: Optional["User"] = None) -> None:	CODE
LOW	letta/schemas/providers/__init__.py	27	__all__ = [	CODE
LOW	letta/prompts/system_prompts/__init__.py	29	__all__ = ["SYSTEM_PROMPTS"]	CODE
LOW⚡	letta/orm/base.py	20	def set_updated_at(self, timestamp: Optional[datetime] = None) -> None:	CODE
LOW⚡	letta/orm/base.py	30	def _set_created_and_updated_by_fields(self, actor_id: str) -> None:	CODE
LOW	letta/monitoring/__init__.py	5	__all__ = [	CODE
LOW	letta/llm_api/openai_ws_session.py	30	logger = logging.getLogger(__name__)	CODE
LOW	letta/llm_api/baseten_client.py	16	logger = logging.getLogger(__name__)	CODE
LOW	letta/helpers/tpuf_client.py	25	logger = logging.getLogger(__name__)	CODE
LOW	letta/services/provider_manager.py	244	async def update_provider_last_synced_async(self, provider_id: str, actor: Optional[PydanticUser] = None) -> None:	CODE
LOW	letta/services/webhook_service.py	6	logger = logging.getLogger(__name__)	CODE
LOW	letta/services/sandbox_credentials_service.py	9	logger = logging.getLogger(__name__)	CODE
LOW	letta/services/credit_verification_service.py	8	logger = logging.getLogger(__name__)	CODE
LOW	letta/services/lettuce/__init__.py	6	__all__ = ["LettuceClient"]	CODE
LOW	letta/services/llm_router/__init__.py	15	__all__ = [	CODE
LOW	letta/services/memory_repo/__init__.py	12	__all__ = [	CODE
LOW	letta/services/memory_repo/storage/__init__.py	6	__all__ = [	CODE
LOW	letta/services/mcp/oauth_utils.py	56	async def set_tokens(self, tokens: OAuthToken) -> None:	CODE
LOW	letta/services/mcp/oauth_utils.py	83	async def set_client_info(self, client_info: OAuthClientInformationFull) -> None:	CODE
LOW⚡	letta/services/mcp/oauth_utils.py	139	async def update_session_status(self, status: OAuthSessionStatus) -> None:	CODE
LOW⚡	letta/services/mcp/oauth_utils.py	179	async def set_authorization_url(self, url: str) -> None:	CODE
LOW	letta/services/provider_trace_backends/__init__.py	15	__all__ = [	CODE
LOW	…erfaces/openai_chat_completions_streaming_interface.py	74	def _update_tool_call_info(self, tool_call: Any) -> None:	CODE

Hallucination Indicators4 hits · 40 pts

Severity	File	Line	Snippet	Context
CRITICAL	letta/orm/sqlalchemy_base.py	270	query = query.join(cls.identities).filter(cls.identities.property.mapper.class_.identifier_key.in_(identifie	CODE
CRITICAL	letta/jobs/llm_batch_job_polling.py	55	response = await server.anthropic_async_client.beta.messages.batches.retrieve(batch_id_str)	CODE
CRITICAL	letta/jobs/llm_batch_job_polling.py	80	results = await server.anthropic_async_client.beta.messages.batches.results(batch_resp_id)	CODE
CRITICAL	letta/services/block_manager.py	404	BlockModel.identities.property.mapper.class_.identifier_key.in_(identifier_keys)	CODE

Synthetic Comment Markers4 hits · 25 pts

Severity	File	Line	Snippet	Context
HIGH	letta/services/tool_sandbox/local_sandbox.py	149	print(f"Auto-generated code for debugging:\n\n{code}")	CODE
HIGH	letta/services/tool_executor/tool_execution_sandbox.py	49	# This is the variable name in the auto-generated code that contains the function results	COMMENT
HIGH⚡	letta/services/tool_executor/tool_execution_sandbox.py	167	logger.error(f"Logging out tool {self.tool_name} auto-generated code for debugging: \n\n{code}")	CODE
HIGH	letta/services/tool_executor/tool_execution_sandbox.py	233	logger.debug(f"Tool {self.tool_name} auto-generated code for debugging: \n\n{code}")	CODE

Overly Generic Function Names20 hits · 12 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_modal_sandbox_v2.py	338	def test_function(x: int, y: int) -> int:	CODE
LOW	tests/test_modal_sandbox_v2.py	395	mock_tool.source_code = "def test_function(x, y): return x * y"	CODE
LOW	tests/test_modal_sandbox_v2.py	408	mock_tool.source_code = "def test_function(x, y): return x + y" # Reset	CODE
LOW	tests/test_utils.py	92	def my_function(a: int, b: float, c: str, d: list, e: dict, f: CustomClass = None):	CODE
LOW	tests/conftest.py	111	def test_function(e2b_sandbox_mode, ...):	STRING
LOW	tests/conftest.py	177	def test_function(turbopuffer_mode, ...):	STRING
LOW	tests/test_sdk_client.py	1722	# def main_function(x: int, y: int) -> int:	COMMENT
LOW	tests/test_sdk_client.py	1752	# def process_data(data: str, count: int) -> str:	COMMENT
LOW	tests/test_sdk_client.py	1639	def main_function(x: int, y: int) -> int:	STRING
LOW⚡	tests/managers/test_tool_manager.py	1805	def test_function(arg1: str) -> str:	CODE
LOW	tests/managers/test_tool_manager.py	1834	def test_function(arg1: str, arg2: int) -> str:	CODE
LOW	tests/managers/test_tool_manager.py	1870	def test_function(arg1: str, arg2: int = 5) -> str:	CODE
LOW	tests/managers/test_tool_manager.py	1903	def test_function(old_arg: str) -> str:	CODE
LOW	tests/managers/test_tool_manager.py	1938	def test_function():	CODE
LOW	tests/managers/test_tool_manager.py	1957	def test_function(arg1: str) -> str:	CODE
LOW	tests/managers/test_tool_manager.py	1986	def test_function() -> str:	CODE
LOW	tests/managers/test_tool_manager.py	2019	def test_function() -> str:	CODE
LOW	tests/managers/test_tool_manager.py	2034	def test_function(new_arg: str) -> str:	CODE
LOW	tests/managers/test_tool_manager.py	2067	source_code="def test_function(): pass",	CODE
LOW	letta/validators.py	136	def my_function(agent_id: str, folder_id: str):	STRING

TODO Padding7 hits · 11 pts

Severity	File	Line	Snippet	Context
LOW	letta/server/rest_api/routers/v1/zai.py	114	# TODO: Add caching layer to avoid duplicate memory searches	COMMENT
LOW⚡	letta/server/rest_api/routers/v1/anthropic.py	118	# TODO: Add caching layer to avoid duplicate memory searches	COMMENT
LOW	letta/schemas/model.py	227	# TODO: implement support for this	COMMENT
LOW	letta/schemas/model.py	244	# TODO: implement support for these	COMMENT
LOW	letta/schemas/model.py	316	# TODO: implement support for these	COMMENT
LOW	letta/services/block_manager.py	833	# TODO: implement for async	COMMENT
LOW	letta/services/helpers/agent_manager_helper.py	588	# TODO: add tests to check if this works for ascending order but nulls are stil last?	COMMENT

Dead Code5 hits · 10 pts

Severity	File	Line	Context
MEDIUM	tests/managers/test_cancellation.py	1504	CODE
MEDIUM	tests/managers/test_cancellation.py	1576	CODE
MEDIUM	tests/managers/test_cancellation.py	1629	CODE
MEDIUM	tests/managers/test_cancellation.py	1681	CODE
MEDIUM	tests/managers/test_cancellation.py	1733	CODE

Slop Phrases5 hits · 9 pts

Severity	File	Line	Snippet	Context
LOW	letta/local_llm/llm_chat_completion_wrappers/chatml.py	24	# Don't forget to use send_message, otherwise the user won't see your message"""	COMMENT
LOW	letta/local_llm/llm_chat_completion_wrappers/llama3.py	23	# Don't forget to use send_message, otherwise the user won't see your message"""	COMMENT
LOW⚡	letta/server/server.py	1847	# Write out the file, and make sure to in include the top-level mcpConfig (wrapped to avoid blocking event loop)	COMMENT
LOW⚡	letta/server/server.py	1891	# Write out the file, and make sure to in include the top-level mcpConfig (wrapped to avoid blocking event loop)	COMMENT
LOW	letta/llm_api/anthropic_client.py	973	# make sure to check for overflow errors, regardless of error type	COMMENT

Fake / Example Data9 hits · 8 pts

Severity	File	Line	Snippet	Context
LOW	…/integration_test_typescript_tool_execution_sandbox.py	425	assert result.func_return == "John Doe"	CODE
LOW	tests/sdk/mcp_servers_test.py	1072	"content": 'Use the create_person tool to create a person named "John Doe", age 30, '	CODE
LOW	tests/sdk/mcp_servers_test.py	1073	'email "john@example.com", with address at "123 Main St", city "New York", zip "10001".',	CODE
LOW	tests/sdk/mcp_servers_test.py	1089	assert "John Doe" in person_return.tool_return, "Expected person name in response"	CODE
LOW	tests/data/api_server.go	360	userService.CreateUser("John Doe", "john@example.com")	CODE
LOW	tests/data/test.md	78	1. Lorem ipsum dolor sit amet	CODE
LOW	tests/data/test.md	78	1. Lorem ipsum dolor sit amet	CODE
LOW	scripts/migrate_tools.py	10	fake_user = User(id="user-00000000-0000-4000-8000-000000000000", name="fake", organization_id=org.id)	CODE
LOW	scripts/migrate_tools.py	12	ToolManager().upsert_base_tools(actor=fake_user)	CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings