Home / microsoft / RD-Agent
AI Slop Code RIsk Monitor

microsoft/RD-Agent

Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are committed to automating these high-value generic R&D processes through R&D-Agent, which lets AI drive data-driven AI. 🔗https://aka.ms/RD-Agent-Tech-Report

26.2 Elevated pattern signal View on GitHub

Maintenance and Governance Review

This report summarizes scanner evidence for microsoft/RD-Agent, a Python project with 14,887 GitHub stars. SynthScan examined 115,141 lines of code across 803 source files, recording 1537 pattern matches distributed across 24 syntactic categories. The legacy adjusted pattern signal is 26.2 (Elevated pattern signal); it is not an authorship estimate.

Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.

26.2
Adjusted Signal
26.2
Raw Pattern Score
100%
Time Factor
2026-10-06
Last Push
14.9K
Stars
Python
Language
115.1K
Lines of Code
803
Files
1.5K
Pattern Hits
2026-10-12
Scan Date
0.33
HC Hit Rate

Risk Review · v3

Use each vector to guide manual verification of the underlying findings.

Cognitive Ownership Risk 33.1/100 · MEDIUM

Can maintainers explain intent, assumptions, and trade-offs? 682 finding(s).

Hallucinated Dependency & Supply Chain Risk 14.1/100 · LOW

Are imports declared and package/API names verified? 313 finding(s).

Agent Over-Engineering Ratio 58.5/100 · HIGH

Are wrappers, guards, and repeated structures necessary? 813 finding(s).

PR Review Velocity Impact 75.5/100 · CRITICAL

Which changes need focused human review? 1537 finding(s).

What These Metrics Mean

Adjusted Signal
Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score
Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor
A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits
Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate
High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files
Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.

Severity Breakdown

Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.

CRITICAL 0HIGH 269MEDIUM 177LOW 1091

Directory Score Breakdown

This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.

Pattern Findings

The scanner identified 1537 distinct pattern matches across 24 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. 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 configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.

Cross-File Repetition212 hits · 1060 pts
SeverityFileLineSnippetContext
HIGHrdagent/core/interactor.py0interact with the user to get feedback or confirmation. responsibilities: - present the current state of the experiment STRING
HIGHrdagent/scenarios/data_science/interactor/__init__.py0interact with the user to get feedback or confirmation. responsibilities: - present the current state of the experiment STRING
HIGHrdagent/components/interactor/__init__.py0interact with the user to get feedback or confirmation. responsibilities: - present the current state of the experiment STRING
HIGHrdagent/app/qlib_rd_loop/model.py0auto r&d evolving loop for fintech factors. you can continue running session by .. code-block:: python dotenv run -- pytSTRING
HIGHrdagent/app/qlib_rd_loop/factor.py0auto r&d evolving loop for fintech factors. you can continue running session by .. code-block:: python dotenv run -- pytSTRING
HIGHrdagent/app/qlib_rd_loop/quant.py0auto r&d evolving loop for fintech factors. you can continue running session by .. code-block:: python dotenv run -- pytSTRING
HIGH…ent/components/coder/data_science/ensemble/__init__.py0assign the code list to the evolving item. the code list is aligned with the evolving item's sub-tasks. if a task is notSTRING
HIGH…ent/components/coder/data_science/pipeline/__init__.py0assign the code list to the evolving item. the code list is aligned with the evolving item's sub-tasks. if a task is notSTRING
HIGH…ent/components/coder/data_science/workflow/__init__.py0assign the code list to the evolving item. the code list is aligned with the evolving item's sub-tasks. if a task is notSTRING
HIGHrdagent/components/coder/data_science/model/__init__.py0assign the code list to the evolving item. the code list is aligned with the evolving item's sub-tasks. if a task is notSTRING
HIGH…gent/components/coder/data_science/feature/__init__.py0assign the code list to the evolving item. the code list is aligned with the evolving item's sub-tasks. if a task is notSTRING
HIGH…ponents/coder/data_science/raw_data_loader/__init__.py0assign the code list to the evolving item. the code list is aligned with the evolving item's sub-tasks. if a task is notSTRING
HIGHrdagent/scenarios/data_science/dev/runner/__init__.py0assign the code list to the evolving item. the code list is aligned with the evolving item's sub-tasks. if a task is notSTRING
HIGH…ent/spaceship-titanic_template/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…mplates/playground-series-s4e8/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…/templates/meta_tpl_deprecated/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…lar-playground-series-dec-2021/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…york-city-taxi-fare-prediction/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…plates/playground-series-s3e26/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…/experiment/templates/sf-crime/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…s/forest-cover-type-prediction/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…nt/templates/spaceship-titanic/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…ent/templates/digit-recognizer/fea_share_preprocess.py0this method loads the data, drops the unnecessary columns, and splits it into train and validation sets.STRING
HIGH…ent/spaceship-titanic_template/fea_share_preprocess.py0fits the preprocessor on the training data and returns the fitted preprocessor.STRING
HIGH…mplates/playground-series-s4e8/fea_share_preprocess.py0fits the preprocessor on the training data and returns the fitted preprocessor.STRING
HIGH…/templates/meta_tpl_deprecated/fea_share_preprocess.py0fits the preprocessor on the training data and returns the fitted preprocessor.STRING
HIGH…plates/playground-series-s3e26/fea_share_preprocess.py0fits the preprocessor on the training data and returns the fitted preprocessor.STRING
HIGH…/experiment/templates/sf-crime/fea_share_preprocess.py0fits the preprocessor on the training data and returns the fitted preprocessor.STRING
HIGH…nt/templates/spaceship-titanic/fea_share_preprocess.py0fits the preprocessor on the training data and returns the fitted preprocessor.STRING
HIGH…ent/spaceship-titanic_template/fea_share_preprocess.py0transforms the given dataframe using the fitted preprocessor. ensures the processed data has consistent features across STRING
HIGH…/templates/meta_tpl_deprecated/fea_share_preprocess.py0transforms the given dataframe using the fitted preprocessor. ensures the processed data has consistent features across STRING
HIGH…nt/templates/spaceship-titanic/fea_share_preprocess.py0transforms the given dataframe using the fitted preprocessor. ensures the processed data has consistent features across STRING
HIGH…ent/spaceship-titanic_template/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…mplates/playground-series-s4e8/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…plates/playground-series-s3e14/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…/templates/meta_tpl_deprecated/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…ventilator-pressure-prediction/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…lar-playground-series-dec-2021/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…york-city-taxi-fare-prediction/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…plates/playground-series-s3e11/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…plates/playground-series-s3e16/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…plates/playground-series-s3e26/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…rize-english-language-learning/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…/experiment/templates/sf-crime/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…s/forest-cover-type-prediction/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…lar-playground-series-may-2022/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…nt/templates/spaceship-titanic/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…l-iceberg-classifier-challenge/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…ent/templates/digit-recognizer/fea_share_preprocess.py0this method applies the preprocessing steps to the training, validation, and test datasets.STRING
HIGH…paceship-titanic_template/model/select_randomforest.py0select relevant features. to be used in fit & predict function.STRING
HIGH…periment/spaceship-titanic_template/model/select_nn.py0select relevant features. to be used in fit & predict function.STRING
HIGH…nt/spaceship-titanic_template/model/select_lightgbm.py0select relevant features. to be used in fit & predict function.STRING
HIGH…ent/spaceship-titanic_template/model/select_xgboost.py0select relevant features. to be used in fit & predict function.STRING
HIGH…es/playground-series-s4e9/model/select_randomforest.py0select relevant features. to be used in fit & predict function.STRING
HIGH…nt/templates/playground-series-s4e9/model/select_nn.py0select relevant features. to be used in fit & predict function.STRING
HIGH…plates/playground-series-s4e9/model/select_lightgbm.py0select relevant features. to be used in fit & predict function.STRING
HIGH…mplates/playground-series-s4e9/model/select_xgboost.py0select relevant features. to be used in fit & predict function.STRING
HIGH…es/playground-series-s4e8/model/select_randomforest.py0select relevant features. to be used in fit & predict function.STRING
HIGH…nt/templates/playground-series-s4e8/model/select_nn.py0select relevant features. to be used in fit & predict function.STRING
HIGH…plates/playground-series-s4e8/model/select_lightgbm.py0select relevant features. to be used in fit & predict function.STRING
152 more matches not shown…
Unused Imports304 hits · 300 pts
SeverityFileLineSnippetContext
LOWtest/utils/test_conf.py4CODE
LOWtest/utils/test_env.py1CODE
LOWtest/utils/test_import.py2CODE
LOWtest/utils/env_tpl/read_exp.py2CODE
LOWtest/utils/env_tpl/read_exp.py3CODE
LOWtest/finetune/test_benchmark_tablebench.py6CODE
LOWtest/finetune/test_benchmark.py10CODE
LOWtest/finetune/test_benchmark_api.py10CODE
LOWtest/oai/test_advanced.py7CODE
LOWtest/oai/test_base.py1CODE
LOWtest/oai/test_completion.py3CODE
LOWtest/notebook/testfiles/main2.py5CODE
LOWtest/notebook/testfiles/main2.py14CODE
LOWtest/notebook/testfiles/main2.py16CODE
LOWrdagent/core/evolving_agent.py1CODE
LOWrdagent/core/conf.py1CODE
LOWrdagent/core/evolving_framework.py1CODE
LOWrdagent/core/proposal.py3CODE
LOWrdagent/core/experiment.py1CODE
LOWrdagent/core/utils.py1CODE
LOWrdagent/core/developer.py1CODE
LOWrdagent/core/interactor.py1CODE
LOWrdagent/app/cli.py9CODE
LOWrdagent/app/benchmark/factor/analysis.py8CODE
LOWrdagent/app/benchmark/model/eval.py14CODE
LOWrdagent/app/CI/run.py1CODE
LOWrdagent/app/utils/ws.py1CODE
LOWrdagent/app/utils/health_check.py6CODE
LOWrdagent/app/utils/info.py4CODE
LOWrdagent/app/utils/info.py9CODE
LOWrdagent/app/utils/ws_ft.py1CODE
LOWrdagent/app/utils/ws_ft.py5CODE
LOWrdagent/app/utils/ws_ft.py7CODE
LOWrdagent/app/finetune/data_science/scen.py4CODE
LOWrdagent/app/finetune/data_science/scen.py5CODE
LOWrdagent/app/finetune/data_science/scen.py8CODE
LOWrdagent/app/finetune/data_science/loop.py8CODE
LOWrdagent/app/finetune/data_science/loop.py9CODE
LOWrdagent/app/qlib_rd_loop/factor_from_report.py4CODE
LOWrdagent/app/qlib_rd_loop/factor_from_report.py15CODE
LOWrdagent/app/rl/ui/data_loader.py18CODE
LOWrdagent/utils/env.py44CODE
LOWrdagent/utils/__init__.py21CODE
LOWrdagent/utils/agent/apply_patch.py9CODE
LOWrdagent/utils/agent/__init__.py1CODE
LOWrdagent/utils/agent/workflow.py2CODE
LOWrdagent/utils/agent/workflow.py2CODE
LOWrdagent/utils/workflow/__init__.py1CODE
LOWrdagent/utils/workflow/__init__.py1CODE
LOWrdagent/utils/workflow/__init__.py2CODE
LOWrdagent/utils/workflow/__init__.py3CODE
LOWrdagent/utils/workflow/loop.py20CODE
LOWrdagent/utils/workflow/tracking.py18CODE
LOWrdagent/utils/repo/repo_utils.py2CODE
LOWrdagent/oai/llm_utils.py1CODE
LOWrdagent/oai/llm_utils.py10CODE
LOWrdagent/oai/llm_conf.py1CODE
LOWrdagent/oai/backend/deprec.py2CODE
LOWrdagent/oai/backend/deprec.py4CODE
LOWrdagent/oai/backend/deprec.py7CODE
244 more matches not shown…
Hyper-Verbose Identifiers256 hits · 263 pts
SeverityFileLineSnippetContext
LOWtest/qlib/test_model_factor_proposal.py42def test_model_proposal_import():CODE
LOWtest/qlib/test_model_factor_proposal.py46def test_factor_proposal_import():CODE
LOW⚡test/core/test_serialization.py26def test_signed_pickle_round_trip(signing_key: Path) -> None:CODE
LOW⚡test/core/test_serialization.py34def test_configured_signing_key_does_not_create_key_file(signing_key: Path, monkeypatch: pytest.MonkeyPatch) -> None:CODE
LOW⚡test/core/test_serialization.py41def test_short_configured_signing_key_is_rejected(signing_key: Path, monkeypatch: pytest.MonkeyPatch) -> None:CODE
LOW⚡test/core/test_serialization.py49def test_signed_dill_round_trip(signing_key: Path) -> None:CODE
LOWtest/core/test_serialization.py60def test_modified_signed_pickle_is_rejected(signing_key: Path) -> None:CODE
LOWtest/core/test_serialization.py69def test_unsigned_pickle_is_rejected_before_deserialization(signing_key: Path, tmp_path: Path) -> None:CODE
LOWtest/core/test_serialization.py84def test_legacy_pickle_requires_explicit_opt_in(signing_key: Path) -> None:CODE
LOWtest/core/test_serialization.py91def test_unsigned_cache_is_treated_as_cache_miss(CODE
LOWtest/core/test_serialization.py116def test_file_storage_writes_and_reads_signed_messages(signing_key: Path, tmp_path: Path) -> None:CODE
LOWtest/utils/test_env.py146 def test_cleanup_container_import(self):CODE
LOWtest/utils/test_import.py17 def import_all_modules_from_directory(directory):CODE
LOWtest/utils/test_conda_security.py10def test_prepare_conda_env_rejects_shell_metacharacters(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:CODE
LOWtest/utils/test_conda_security.py22def test_prepare_conda_env_uses_argument_lists(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:CODE
LOWtest/utils/test_ws.py27 def test_checkpoint_roundtrip(self) -> None:CODE
LOWtest/utils/test_ws.py78 def test_workspace_file_operations_reject_path_escape(self) -> None:CODE
LOWtest/utils/test_ws.py103 def test_workspace_file_operations_reject_symlink_escape(self) -> None:CODE
LOWtest/utils/test_ws.py123 def test_workspace_delete_unlinks_leaf_symlink_without_deleting_target(self) -> None:CODE
LOWtest/utils/test_artifact_transport.py17def test_safe_artifact_bundle_round_trip(tmp_path: Path) -> None:CODE
LOWtest/utils/test_artifact_transport.py37def test_artifact_bundle_rejects_file_path_escape(tmp_path: Path) -> None:CODE
LOWtest/utils/test_artifact_transport.py48def test_result_loader_rejects_pickle(tmp_path: Path) -> None:CODE
LOWtest/utils/test_archive_security.py13def test_competition_slug_rejects_command_and_path_syntax() -> None:CODE
LOWtest/utils/test_archive_security.py21def test_safe_extract_zip_rejects_path_traversal(tmp_path: Path) -> None:CODE
LOWtest/utils/test_archive_security.py32def test_safe_extract_zip_rejects_windows_absolute_path(tmp_path: Path) -> None:CODE
LOWtest/utils/test_archive_security.py42def test_safe_extract_zip_extracts_regular_files(tmp_path: Path) -> None:CODE
LOWtest/utils/test_archive_security.py53def test_safe_extract_zip_enforces_member_and_size_limits(tmp_path: Path) -> None:CODE
LOWtest/utils/test_archive_security.py66def test_safe_extract_tar_rejects_links(tmp_path: Path) -> None:CODE
LOWtest/utils/test_archive_security.py79def test_safe_extract_tar_extracts_regular_files(tmp_path: Path) -> None:CODE
LOWtest/utils/test_archive_security.py93def test_safe_extract_tar_enforces_member_and_size_limits(tmp_path: Path) -> None:CODE
LOWtest/utils/test_kaggle.py12 def test_competition_template(self):CODE
LOW⚡test/oai/test_embedding_and_similarity.py22 def test_embedding_similarity(self) -> None:CODE
LOW⚡test/oai/test_embedding_and_similarity.py29 def test_embedding_long_text_truncation(self) -> None:CODE
LOWtest/oai/test_advanced.py24 def test_chat_cache_multiprocess(self) -> None:CODE
LOWtest/oai/test_completion.py29 def test_chat_completion_json_mode(self) -> None:CODE
LOWtest/oai/test_completion.py41 def test_build_messages_and_calculate_token(self) -> None:CODE
LOWtest/oai/test_completion.py48 def test_json_mode_with_specific_target_type(self) -> None:CODE
LOWtest/oai/test_completion.py68 def test_response_format_with_basemodel(self) -> None:CODE
LOW⚡test/scenarios/data_science/test_secure_parsing.py17def test_score_parser_accepts_finite_json_number() -> None:CODE
LOW⚡test/scenarios/data_science/test_secure_parsing.py22def test_score_parser_does_not_execute_python(tmp_path: Path) -> None:CODE
LOW⚡test/scenarios/data_science/test_secure_parsing.py32def test_score_parser_rejects_non_finite_or_non_numeric_values(score: str) -> None:CODE
LOWtest/scenarios/data_science/test_secure_parsing.py44def test_reusable_grade_script_requires_strict_json(CODE
LOWtest/scenarios/data_science/test_secure_parsing.py77def test_incompatible_cached_grade_script_is_regenerated(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:CODE
LOWtest/scenarios/data_science/test_secure_parsing.py108def test_missing_cached_submission_is_recreated(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:CODE
LOWtest/scenarios/data_science/test_secure_parsing.py142def test_training_metrics_parser_does_not_execute_python(tmp_path: Path) -> None:CODE
LOWtest/scenarios/data_science/test_debug_data.py7def test_json_reducer_extracts_filename_from_unnamed_key() -> None:CODE
LOW⚡test/log/server/test_security.py26def test_validate_scenario_rejects_path_traversal() -> None:CODE
LOW⚡test/log/server/test_security.py32def test_parse_competition_accepts_only_mle_bench_slug() -> None:CODE
LOW⚡test/log/server/test_security.py40def test_resolve_within_rejects_escape(tmp_path: Path) -> None:CODE
LOW⚡test/log/server/test_security.py47def test_upload_filename_rejects_executable_formats() -> None:CODE
LOW⚡test/log/server/test_security.py55def test_log_server_requires_authentication_when_configured(monkeypatch: pytest.MonkeyPatch) -> None:CODE
LOWtest/log/server/test_security.py69def test_web_storage_authenticates_internal_receive_requests(CODE
LOWtest/log/server/test_security.py97def test_upload_rejects_unknown_scenario_before_writing(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:CODE
LOWtest/log/server/test_security.py113def test_upload_rejects_pickle_before_starting_task(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> None:CODE
LOWtest/log/server/test_security.py156def test_all_apis_fail_closed_without_credentials(CODE
LOWtest/log/server/test_security.py175def test_untrusted_origins_cannot_reach_apis(CODE
LOWtest/log/server/test_security.py193def test_cross_site_simple_post_has_no_side_effects(CODE
LOWtest/log/server/test_security.py212def test_cookie_post_requires_trusted_provenance(CODE
LOWtest/log/server/test_security.py235def test_authenticated_report_upload_starts_task(CODE
LOWtest/log/server/test_security.py266def test_report_text_input_is_rejected_before_side_effects(CODE
196 more matches not shown…
Excessive Try-Catch Wrapping193 hits · 242 pts
SeverityFileLineSnippetContext
LOWtest/utils/test_import.py56 except Exception as e:CODE
MEDIUMtest/finetune/test_benchmark_tablebench.py122 print(f"Error: {result.stdout[-2000:] if result.stdout else 'No output'}")CODE
MEDIUMtest/finetune/test_benchmark.py125 print(f"Error: {result.stdout[-2000:] if result.stdout else 'No output'}")CODE
MEDIUMtest/finetune/test_benchmark_api.py296 print(f"Error: {result.stdout[-2000:] if result.stdout else 'No output'}")CODE
LOWtest/oai/test_llm_connectivity.py42 except Exception as e:CODE
LOWtest/notebook/test_util.py1475except Exception as e:CODE
LOWtest/notebook/test_util.py1483except Exception as e:CODE
MEDIUM⚡test/notebook/testfiles/main2.py56 print(f"Error loading train.csv: {e}")CODE
MEDIUM⚡test/notebook/testfiles/main2.py62 print(f"Error listing train dir: {e}")CODE
MEDIUM⚡test/notebook/testfiles/main2.py68 print(f"Error listing test dir: {e}")CODE
MEDIUM⚡test/notebook/testfiles/main2.py79 print(f"Error reading sample_submission.csv: {e}")CODE
MEDIUM⚡test/notebook/testfiles/main2.py127 print(f"Error reading {filepath}: {e}")CODE
MEDIUMtest/notebook/testfiles/main2.py372 print("Error computing log_loss on val:", ex)CODE
MEDIUMtest/notebook/testfiles/main2.py420 print("Error computing log_loss on validation:", ex)CODE
MEDIUMtest/notebook/testfiles/main2.py117def load_img_as_numpy_with_mask(filepath):CODE
LOW⚡test/notebook/testfiles/main2.py55 except Exception as e:CODE
LOW⚡test/notebook/testfiles/main2.py61 except Exception as e:CODE
LOW⚡test/notebook/testfiles/main2.py67 except Exception as e:CODE
LOW⚡test/notebook/testfiles/main2.py78 except Exception as e:CODE
LOW⚡test/notebook/testfiles/main2.py126 except Exception as e:CODE
LOWtest/notebook/testfiles/main2.py198 except Exception as e:CODE
LOWtest/notebook/testfiles/main2.py370 except Exception as ex:CODE
LOWtest/notebook/testfiles/main2.py418 except Exception as ex:CODE
LOWtest/notebook/testfiles/main2.py446 except Exception:CODE
LOWtest/notebook/testfiles/main_missing_sections.py303 except Exception as e:CODE
LOWtest/notebook/testfiles/main_missing_sections.py311 except Exception as e:CODE
MEDIUMtest/notebook/testfiles/main_missing_sections.py300def main():CODE
LOWtest/notebook/testfiles/main_missing_main_fn.py303except Exception as e:CODE
LOWtest/notebook/testfiles/main_missing_main_fn.py311except Exception as e:CODE
LOWtest/notebook/testfiles/main.py305 except Exception as e:CODE
LOWtest/notebook/testfiles/main.py313 except Exception as e:CODE
LOWrdagent/app/kaggle/loop.py86 except Exception as e:CODE
LOWrdagent/app/kaggle/loop.py106 except Exception as e:CODE
LOWrdagent/app/utils/health_check.py73 except Exception as e:CODE
LOWrdagent/app/utils/health_check.py89 except Exception as e:CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py55 except Exception:CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py96 except Exception:CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py139 except Exception:CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py181 except Exception:CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py452 except Exception:CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py485 except Exception:CODE
LOWrdagent/app/finetune/llm/ui/components.py225 except Exception:CODE
LOWrdagent/app/rl/ui/rl_summary.py46 except Exception:CODE
MEDIUMrdagent/utils/env.py972def prepare(self) -> None:CODE
MEDIUMrdagent/utils/env.py1027def prepare(self) -> None:CODE
LOWrdagent/utils/env.py112 except Exception as cleanup_error:CODE
LOWrdagent/utils/env.py348 except Exception as e:CODE
LOWrdagent/utils/env.py879 except Exception as e:CODE
LOWrdagent/utils/env.py904 except Exception as exc: # pragma: no cover - best-effort helperCODE
LOWrdagent/utils/env.py1004 except Exception as e:CODE
LOWrdagent/utils/env.py1038 except Exception as e:CODE
LOWrdagent/utils/env.py1438 except Exception:CODE
LOWrdagent/utils/__init__.py83 except Exception as e:CODE
LOWrdagent/utils/__init__.py172 except Exception as e:CODE
LOWrdagent/utils/__init__.py181 except Exception as e:CODE
LOWrdagent/utils/agent/workflow.py53 except Exception as e:CODE
LOWrdagent/utils/workflow/misc.py40 except Exception as e:CODE
MEDIUMrdagent/utils/workflow/misc.py41 print(f"Error: {e}")CODE
LOWrdagent/utils/workflow/loop.py248 except Exception as e:CODE
LOWrdagent/utils/workflow/loop.py557 except Exception as ex:CODE
133 more matches not shown…
Empty Error Handlers42 hits · 210 pts
SeverityFileLineSnippetContext
HIGHrdagent/core/utils.py64except json.decoder.JSONDecodeError:CODE
HIGHrdagent/app/finetune/llm/ui/ft_summary.py55except Exception:CODE
HIGHrdagent/app/finetune/llm/ui/ft_summary.py96except Exception:CODE
HIGHrdagent/app/finetune/llm/ui/ft_summary.py139except Exception:CODE
HIGHrdagent/app/finetune/llm/ui/ft_summary.py181except Exception:CODE
HIGHrdagent/app/finetune/llm/ui/ft_summary.py452except Exception:CODE
HIGHrdagent/app/finetune/llm/ui/ft_summary.py485except Exception:CODE
HIGHrdagent/app/finetune/llm/ui/ft_summary.py349except ValueError:CODE
HIGHrdagent/app/finetune/llm/ui/app.py52except PermissionError:CODE
HIGHrdagent/app/finetune/llm/ui/components.py225except Exception:CODE
HIGHrdagent/app/rl/ui/rl_summary.py46except Exception:CODE
HIGHrdagent/app/rl/ui/app.py64except PermissionError:CODE
HIGHrdagent/utils/env.py1438except Exception:CODE
HIGHrdagent/utils/env.py651except FileNotFoundError:CODE
HIGHrdagent/utils/agent/tpl.py118except ValueError:CODE
HIGH…ent/components/coder/factor_coder/evolving_strategy.py161except (json.decoder.JSONDecodeError, KeyError):CODE
HIGHrdagent/components/coder/finetune/unified_validator.py239except Exception:CODE
HIGHrdagent/scenarios/data_science/scen/utils.py359except MaxLinesExceededError:CODE
HIGHrdagent/scenarios/data_science/scen/utils.py487except MaxLinesExceededError:CODE
HIGHrdagent/scenarios/data_science/scen/utils.py509except MaxLinesExceededError:CODE
HIGHrdagent/scenarios/data_science/scen/utils.py529except MaxLinesExceededError:CODE
HIGHrdagent/scenarios/data_science/scen/utils.py674except ValueError:CODE
HIGH…enarios/data_science/proposal/exp_gen/select/submit.py631except (KeyError, TypeError, ValueError):CODE
HIGHrdagent/scenarios/data_science/debug/data.py15except:CODE
HIGHrdagent/scenarios/finetune/benchmark/merge/merge.py24except Exception:CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/server.py404except requests.ConnectionError:CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/utils.py75except (ProcessLookupError, OSError):CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/utils.py94except FileExistsError:CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/utils.py64except (ProcessLookupError, OSError):CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/utils.py54except (OSError, subprocess.SubprocessError, ValueError):CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/utils.py59except (ProcessLookupError, OSError):CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/utils.py52except (ProcessLookupError, OSError):CODE
HIGHrdagent/scenarios/rl/autorl_bench/core/opencompass.py193except (ValueError, TypeError):CODE
HIGH…agent/scenarios/rl/autorl_bench/test/test_benchmark.py69except:CODE
HIGH…cenarios/rl/autorl_bench/agents/example_agent/train.py23except:CODE
HIGH…cenarios/rl/autorl_bench/agents/example_agent/train.py29except:CODE
HIGH…cenarios/rl/autorl_bench/benchmarks/alpacaeval/eval.py128except Exception:CODE
HIGH…os/rl/autorl_bench/benchmarks/smith/per_sample_eval.py167except Exception:CODE
HIGH…os/rl/autorl_bench/benchmarks/smith/per_sample_eval.py175except Exception:CODE
HIGH…os/rl/autorl_bench/benchmarks/smith/per_sample_eval.py162except Exception:CODE
HIGHrdagent/log/server/app.py185except Exception:CODE
HIGHrdagent/log/server/app.py189except Exception:CODE
Decorative Section Separators66 hits · 197 pts
SeverityFileLineSnippetContext
MEDIUMrdagent/utils/__init__.py108 \d+/\d+\s+[━]+\s+\d+s?\s+\d+ms/step.*?\u0008+ | # e.g. "10/100 ━━━━━━ 3s 50ms/step"CODE
MEDIUMrdagent/utils/__init__.py109 \d+/\d+\s+[━]+\s+\d+s?\s+\d+ms/step | # e.g. "10/100 ━━━━━━ 3s 50ms/step" (no backspaces)CODE
MEDIUM…arios/data_science/proposal/exp_gen/trace_scheduler.py134# ======================================================================================COMMENT
MEDIUM…arios/data_science/proposal/exp_gen/trace_scheduler.py136# ======================================================================================COMMENT
MEDIUM…enarios/data_science/proposal/exp_gen/select/submit.py39# ==============================================================================COMMENT
MEDIUM…enarios/data_science/proposal/exp_gen/select/submit.py41# ==============================================================================COMMENT
MEDIUM…enarios/data_science/proposal/exp_gen/select/submit.py555# ==============================================================================COMMENT
MEDIUM…enarios/data_science/proposal/exp_gen/select/submit.py557# ==============================================================================COMMENT
MEDIUM…enarios/data_science/proposal/exp_gen/select/submit.py667# ==============================================================================COMMENT
MEDIUM…enarios/data_science/proposal/exp_gen/select/submit.py669# ==============================================================================COMMENT
MEDIUMrdagent/scenarios/finetune/benchmark/data/default.py17# ============================================================================COMMENT
MEDIUMrdagent/scenarios/finetune/benchmark/data/default.py19# ============================================================================COMMENT
MEDIUMrdagent/scenarios/finetune/benchmark/data/default.py203# ============================================================================COMMENT
MEDIUMrdagent/scenarios/finetune/benchmark/data/default.py205# ============================================================================COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml1# =============================================================================COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml3# =============================================================================COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml18 # ═══════════════════════════════════════════════════════════════════════════COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml20 # ═══════════════════════════════════════════════════════════════════════════COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml163 # ═══════════════════════════════════════════════════════════════════════════COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml165 # ═══════════════════════════════════════════════════════════════════════════COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml223 # ═══════════════════════════════════════════════════════════════════════════COMMENT
MEDIUMrdagent/scenarios/finetune/proposal/prompts.yaml225 # ═══════════════════════════════════════════════════════════════════════════COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/server.py356# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/server.py358# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py83# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py85# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py98# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py100# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py173# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py175# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py238# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py240# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py272# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py274# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py346# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py348# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py386# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/utils.py388# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/evaluator.py16# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/evaluator.py18# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/evaluator.py49# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/core/evaluator.py51# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/test/test_fixes.py35# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/test/test_fixes.py37# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/test/test_fixes.py228# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/test/test_fixes.py230# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/test/test_fixes.py313# ============================================================COMMENT
MEDIUMrdagent/scenarios/rl/autorl_bench/test/test_fixes.py315# ============================================================COMMENT
MEDIUM⚡…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py67# ============================================================COMMENT
MEDIUM⚡…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py69# ============================================================COMMENT
MEDIUM…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py121# ============================================================COMMENT
MEDIUM…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py123# ============================================================COMMENT
MEDIUM…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py212# ============================================================COMMENT
MEDIUM…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py214# ============================================================COMMENT
MEDIUM…t/scenarios/rl/autorl_bench/benchmarks/webshop/eval.py56# ============================================================COMMENT
MEDIUM…t/scenarios/rl/autorl_bench/benchmarks/webshop/eval.py58# ============================================================COMMENT
MEDIUM⚡…t/scenarios/rl/autorl_bench/benchmarks/webshop/eval.py148# ============================================================COMMENT
MEDIUM⚡…t/scenarios/rl/autorl_bench/benchmarks/webshop/eval.py150# ============================================================COMMENT
MEDIUM…t/scenarios/rl/autorl_bench/benchmarks/webshop/eval.py261# ============================================================STRING
MEDIUM…t/scenarios/rl/autorl_bench/benchmarks/webshop/eval.py263# ============================================================STRING
6 more matches not shown…
Deep Nesting177 hits · 164 pts
SeverityFileLineSnippetContext
LOWtest/notebook/testfiles/main2.py37CODE
LOWrdagent/core/evolving_agent.py140CODE
LOWrdagent/core/experiment.py346CODE
LOWrdagent/core/experiment.py372CODE
LOWrdagent/app/CI/run.py182CODE
LOWrdagent/app/CI/run.py432CODE
LOWrdagent/app/CI/run.py185CODE
LOWrdagent/app/finetune/llm/ui/data_loader.py98CODE
LOWrdagent/app/finetune/llm/ui/data_loader.py365CODE
LOWrdagent/app/finetune/llm/ui/data_loader.py423CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py27CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py73CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py101CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py224CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py429CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py457CODE
LOWrdagent/app/finetune/llm/ui/ft_summary.py490CODE
LOWrdagent/app/finetune/llm/ui/app.py28CODE
LOWrdagent/app/finetune/llm/ui/app.py66CODE
LOWrdagent/app/finetune/llm/ui/components.py42CODE
LOWrdagent/app/finetune/llm/ui/components.py58CODE
LOWrdagent/app/finetune/llm/ui/components.py208CODE
LOWrdagent/app/finetune/llm/ui/components.py313CODE
LOWrdagent/app/finetune/llm/ui/components.py461CODE
LOWrdagent/app/finetune/llm/ui/components.py602CODE
LOWrdagent/app/finetune/llm/ui/components.py665CODE
LOWrdagent/app/finetune/llm/ui/benchmarks/bioprobench.py21CODE
LOWrdagent/app/finetune/llm/ui/benchmarks/__init__.py22CODE
LOWrdagent/app/finetune/llm/ui/benchmarks/chemcotbench.py47CODE
LOWrdagent/app/qlib_rd_loop/factor_from_report.py112CODE
LOWrdagent/app/rl/ui/data_loader.py238CODE
LOWrdagent/app/rl/ui/data_loader.py286CODE
LOWrdagent/app/rl/ui/rl_summary.py75CODE
LOWrdagent/app/rl/ui/app.py46CODE
LOWrdagent/app/rl/ui/app.py75CODE
LOWrdagent/app/rl/ui/components.py24CODE
LOWrdagent/app/rl/ui/components.py38CODE
LOWrdagent/app/rl/ui/components.py211CODE
LOWrdagent/utils/env.py141CODE
LOWrdagent/utils/env.py277CODE
LOWrdagent/utils/env.py609CODE
LOWrdagent/utils/env.py1195CODE
LOWrdagent/utils/env.py1341CODE
LOWrdagent/utils/env.py633CODE
LOWrdagent/utils/__init__.py100CODE
LOWrdagent/utils/agent/apply_patch.py275CODE
LOWrdagent/utils/agent/apply_patch.py385CODE
LOWrdagent/utils/agent/apply_patch.py457CODE
LOWrdagent/utils/agent/apply_patch.py173CODE
LOWrdagent/utils/agent/tpl.py33CODE
LOWrdagent/utils/workflow/loop.py195CODE
LOWrdagent/utils/workflow/loop.py314CODE
LOWrdagent/oai/backend/deprec.py109CODE
LOWrdagent/oai/backend/deprec.py294CODE
LOWrdagent/oai/backend/litellm.py95CODE
LOWrdagent/oai/backend/litellm.py127CODE
LOWrdagent/oai/backend/base.py520CODE
LOWrdagent/components/benchmark/eval_method.py200CODE
LOWrdagent/components/knowledge_management/graph.py196CODE
LOWrdagent/components/workflow/rd_loop.py75CODE
117 more matches not shown…
Verbosity Indicators41 hits · 85 pts
SeverityFileLineSnippetContext
LOW⚡rdagent/utils/env.py979 # Step 1: Install base dependencies (torch, llamafactory, etc.)COMMENT
LOW⚡rdagent/utils/env.py983 # Step 2: Install flash-attn (requires torch first, uses --no-build-isolation)COMMENT
LOW⚡rdagent/utils/agent/ret.py97 # Step 1: extract patch by patternCOMMENT
LOW⚡rdagent/utils/agent/ret.py103 # Step 2: apply the patch, this will modify the file in placeCOMMENT
LOW…nt/components/coder/data_science/pipeline/prompts.yaml302 ### Step 6: Similar Successful Implementations to help Code ImprovementCOMMENT
LOW⚡rdagent/components/coder/finetune/unified_validator.py71 # Step 1: Parameter filteringCOMMENT
LOW⚡rdagent/components/coder/finetune/unified_validator.py74 # Step 2: Inject required parameters for multi-task environmentsCOMMENT
LOW⚡rdagent/components/coder/finetune/unified_validator.py77 # Step 3: Micro-batch testing (validates everything at runtime)COMMENT
LOW⚡rdagent/components/coder/finetune/prompts.yaml87 # Step 1: Run complete sampling/filtering (fast, no LLM) - runs in BOTH modesCOMMENT
LOW⚡rdagent/components/coder/finetune/prompts.yaml90 # Step 2: Limit LLM processing in debug mode onlyCOMMENT
LOW⚡rdagent/components/coder/finetune/prompts.yaml96 # Step 3: Show the actual number of sampled items (Do not estimate; count the exact number of samples that will be pCOMMENT
LOWrdagent/components/coder/finetune/eval.py68 # Step 1: Check script existsCOMMENT
LOWrdagent/components/coder/finetune/eval.py82 # Step 3: Execute script in DEBUG mode (generates ~10 samples for fast validation)COMMENT
LOW⚡rdagent/components/coder/finetune/eval.py101 # Step 4: Validate outputCOMMENT
LOW⚡rdagent/components/coder/finetune/eval.py111 # Step 5: Load data if validCOMMENT
LOW⚡rdagent/components/coder/finetune/eval.py120 # Step 6: Generate LLM feedbackCOMMENT
LOW…cenarios/data_science/proposal/exp_gen/prompts_v2.yaml803 ### Step 2: **Workflow Update** :COMMENT
LOW⚡…nt/scenarios/data_science/proposal/exp_gen/proposal.py401 # Step 2: Generate the rest of the hypothesis & taskCOMMENT
LOW⚡…nt/scenarios/data_science/proposal/exp_gen/proposal.py1458 # Step 3: Select the best hypothesisCOMMENT
LOW…nt/scenarios/data_science/proposal/exp_gen/proposal.py347 # Step 1: Generate componentCOMMENT
LOW…nt/scenarios/data_science/proposal/exp_gen/proposal.py1367 # Step 1: Identify problemsCOMMENT
LOW…nt/scenarios/data_science/proposal/exp_gen/proposal.py1391 # Step 2: Propose hypothesis based on the identified problems (and sampled ideas)COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py205 # Step 1: If we have fewer traces than our target, start a new one.COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py210 # Step 2: Probabilistically select a leaf to expand.COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py359 # Step 1: keep same policy to reach target number of parallel tracesCOMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py364 # Step 2: consider only available leaves (not being expanded)COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py369 # Step 3: compute priors (P) from potentials via softmaxCOMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py377 # Step 4: score each leaf using PUCT-like rule: Q + UCOMMENT
LOW…arios/data_science/proposal/exp_gen/trace_scheduler.py391 # # Step 5: optimistic visit update on selection; value update deferred to observe_feedbackCOMMENT
LOW…data_science/proposal/exp_gen/draft/prompts_draft.yaml220 # Step 2: Workflow UpdateCOMMENT
LOW…scenarios/data_science/proposal/exp_gen/draft/draft.py246 # Step 0: PrepareCOMMENT
LOW⚡…scenarios/data_science/proposal/exp_gen/draft/draft.py271 # Step 1: Retrieve KnowledgeCOMMENT
LOW⚡…scenarios/data_science/proposal/exp_gen/draft/draft.py274 # Step 2: Generate Hypothesis based on General KnowledgeCOMMENT
LOW⚡…scenarios/data_science/proposal/exp_gen/draft/draft.py282 # Step 3: Design TaskCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml12 ## Step 0: Pre-definitionCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml17 ## Step 1: Benchmark Metrics Evaluation (HIGHEST PRIORITY)COMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml30 ## Step 2: Code Quality AssessmentCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml34 ## Step 3: Final Decision (Acceptance as SOTA)COMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml190 ## Step 1: Error ClassificationCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml197 ## Step 2: Root Cause AnalysisCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml202 ## Step 3: Actionable SuggestionsCOMMENT
Structural Annotation Overuse40 hits · 83 pts
SeverityFileLineSnippetContext
LOW⚡rdagent/utils/env.py979 # Step 1: Install base dependencies (torch, llamafactory, etc.)COMMENT
LOW⚡rdagent/utils/env.py983 # Step 2: Install flash-attn (requires torch first, uses --no-build-isolation)COMMENT
LOW⚡rdagent/utils/agent/ret.py97 # Step 1: extract patch by patternCOMMENT
LOW⚡rdagent/utils/agent/ret.py103 # Step 2: apply the patch, this will modify the file in placeCOMMENT
LOW…nt/components/coder/data_science/pipeline/prompts.yaml302 ### Step 6: Similar Successful Implementations to help Code ImprovementCOMMENT
LOW⚡rdagent/components/coder/finetune/unified_validator.py71 # Step 1: Parameter filteringCOMMENT
LOW⚡rdagent/components/coder/finetune/unified_validator.py74 # Step 2: Inject required parameters for multi-task environmentsCOMMENT
LOW⚡rdagent/components/coder/finetune/unified_validator.py77 # Step 3: Micro-batch testing (validates everything at runtime)COMMENT
LOW⚡rdagent/components/coder/finetune/prompts.yaml87 # Step 1: Run complete sampling/filtering (fast, no LLM) - runs in BOTH modesCOMMENT
LOW⚡rdagent/components/coder/finetune/prompts.yaml90 # Step 2: Limit LLM processing in debug mode onlyCOMMENT
LOW⚡rdagent/components/coder/finetune/prompts.yaml96 # Step 3: Show the actual number of sampled items (Do not estimate; count the exact number of samples that will be pCOMMENT
LOWrdagent/components/coder/finetune/eval.py68 # Step 1: Check script existsCOMMENT
LOWrdagent/components/coder/finetune/eval.py82 # Step 3: Execute script in DEBUG mode (generates ~10 samples for fast validation)COMMENT
LOW⚡rdagent/components/coder/finetune/eval.py101 # Step 4: Validate outputCOMMENT
LOW⚡rdagent/components/coder/finetune/eval.py111 # Step 5: Load data if validCOMMENT
LOW⚡rdagent/components/coder/finetune/eval.py120 # Step 6: Generate LLM feedbackCOMMENT
LOW⚡…nt/scenarios/data_science/proposal/exp_gen/proposal.py401 # Step 2: Generate the rest of the hypothesis & taskCOMMENT
LOW⚡…nt/scenarios/data_science/proposal/exp_gen/proposal.py1458 # Step 3: Select the best hypothesisCOMMENT
LOW…nt/scenarios/data_science/proposal/exp_gen/proposal.py347 # Step 1: Generate componentCOMMENT
LOW…nt/scenarios/data_science/proposal/exp_gen/proposal.py1367 # Step 1: Identify problemsCOMMENT
LOW…nt/scenarios/data_science/proposal/exp_gen/proposal.py1391 # Step 2: Propose hypothesis based on the identified problems (and sampled ideas)COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py205 # Step 1: If we have fewer traces than our target, start a new one.COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py210 # Step 2: Probabilistically select a leaf to expand.COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py359 # Step 1: keep same policy to reach target number of parallel tracesCOMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py364 # Step 2: consider only available leaves (not being expanded)COMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py369 # Step 3: compute priors (P) from potentials via softmaxCOMMENT
LOW⚡…arios/data_science/proposal/exp_gen/trace_scheduler.py377 # Step 4: score each leaf using PUCT-like rule: Q + UCOMMENT
LOW…arios/data_science/proposal/exp_gen/trace_scheduler.py391 # # Step 5: optimistic visit update on selection; value update deferred to observe_feedbackCOMMENT
LOW…data_science/proposal/exp_gen/draft/prompts_draft.yaml220 # Step 2: Workflow UpdateCOMMENT
LOW…scenarios/data_science/proposal/exp_gen/draft/draft.py246 # Step 0: PrepareCOMMENT
LOW⚡…scenarios/data_science/proposal/exp_gen/draft/draft.py271 # Step 1: Retrieve KnowledgeCOMMENT
LOW⚡…scenarios/data_science/proposal/exp_gen/draft/draft.py274 # Step 2: Generate Hypothesis based on General KnowledgeCOMMENT
LOW⚡…scenarios/data_science/proposal/exp_gen/draft/draft.py282 # Step 3: Design TaskCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml12 ## Step 0: Pre-definitionCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml17 ## Step 1: Benchmark Metrics Evaluation (HIGHEST PRIORITY)COMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml30 ## Step 2: Code Quality AssessmentCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml34 ## Step 3: Final Decision (Acceptance as SOTA)COMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml190 ## Step 1: Error ClassificationCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml197 ## Step 2: Root Cause AnalysisCOMMENT
LOW⚡rdagent/scenarios/finetune/dev/prompts.yaml202 ## Step 3: Actionable SuggestionsCOMMENT
Self-Referential Comments24 hits · 69 pts
SeverityFileLineSnippetContext
MEDIUMtest/utils/coder/test_finetune_coder.py14 # Create the ensemble task with actual data context and specificationCOMMENT
MEDIUM⚡test/oai/test_embedding_and_similarity.py31 # Create a very long text that will definitely exceed embedding token limitsCOMMENT
MEDIUMrdagent/app/CI/run.py773 # Create a table to display the counts and ratiosCOMMENT
MEDIUMrdagent/utils/agent/apply_patch.py2# The following code is modified from https://cookbook.openai.com/examples/gpt4-1_prompting_guideCOMMENT
MEDIUMrdagent/utils/workflow/tracking.py22# Define a placeholder for mlflow if it's not availableCOMMENT
MEDIUMrdagent/components/proposal/__init__.py22 # The following methods are scenario related so they should be implemented in the subclassCOMMENT
MEDIUMrdagent/components/coder/factor_coder/evaluators.py21 """This class is the v1 version of evaluator for a single factor implementation.STRING
MEDIUMrdagent/components/coder/factor_coder/eva_utils.py399 # Initialize result variablesCOMMENT
MEDIUMrdagent/components/coder/CoSTEER/evaluators.py126 """This class is a base class for all code generator feedback to single implementation"""STRING
MEDIUMrdagent/components/coder/data_science/ensemble/test.py39 # Create the ensemble task with actual data context and specificationCOMMENT
MEDIUMrdagent/components/coder/data_science/model/test.py23 # Create the taskCOMMENT
MEDIUMrdagent/components/coder/data_science/feature/test.py22 # Create the experimentCOMMENT
MEDIUM…/components/coder/data_science/raw_data_loader/test.py19 # Create the experimentCOMMENT
MEDIUMrdagent/components/coder/data_science/share/notebook.py89 # Create a markdown cell for the section name and commentsCOMMENT
MEDIUMrdagent/components/coder/data_science/share/notebook.py98 # Create a code cell for the section code and outputCOMMENT
MEDIUM⚡rdagent/scenarios/kaggle/experiment/prompts.yaml241 # Define the forward passCOMMENT
MEDIUM…xperiment/templates/digit-recognizer/model/model_nn.py11# Define the neural network model with Batch NormalizationCOMMENT
MEDIUM…nt/scenarios/data_science/proposal/exp_gen/proposal.py918 # Create a random but reproducible integerCOMMENT
MEDIUM…enarios/data_science/proposal/exp_gen/select/submit.py491 # Create a temporary workspace to test the generated scriptCOMMENT
MEDIUMrdagent/scenarios/data_science/debug/data.py366 # Create a sampled subsetCOMMENT
MEDIUMrdagent/scenarios/finetune/benchmark/benchmark.py183 # Create a temporary environment for merging (use FT env as it has peft/transformers)COMMENT
MEDIUMrdagent/log/ui/ds_trace.py917 # Create a color map for different root nodes - using colors that work well in both light and dark modesCOMMENT
MEDIUMrdagent/log/ui/app.py1038 # Create a results tableSTRING
MEDIUMrdagent/log/utils/__init__.py40 # This method is called too frequently, which is not good.COMMENT
Redundant / Tautological Comments37 hits · 57 pts
SeverityFileLineSnippetContext
LOWrdagent/app/finetune/llm/ui/ft_summary.py338 # Check if it's a numeric score (with optional "/" separator)COMMENT
LOWrdagent/app/finetune/llm/ui/app.py44 # Check if standalone task (has __session__ directly)COMMENT
LOWrdagent/app/finetune/llm/ui/app.py47 # Check if job directory (subdirs have __session__)COMMENT
LOWrdagent/app/finetune/llm/ui/components.py322 # Check if markdown rendering is enabledCOMMENT
LOWrdagent/utils/env.py1283 # Check if specific GPUs are requested via CUDA_VISIBLE_DEVICESCOMMENT
LOWrdagent/components/coder/factor_coder/eva_utils.py407 # Check if both dataframe has only one columns Mute this since factor task might generate more than one columns COMMENT
LOWrdagent/components/coder/factor_coder/eva_utils.py422 # Check if the index of the dataframe is ("datetime", "instrument")COMMENT
LOWrdagent/components/coder/data_science/pipeline/eval.py78 # Check if error_message contains Context7 documentation resultsCOMMENT
LOWrdagent/components/coder/data_science/pipeline/eval.py242 # Check if scores contain NaN (values)COMMENT
LOWrdagent/components/coder/data_science/workflow/eval.py113 # Check if scores contain NaN (values)COMMENT
LOWrdagent/components/coder/data_science/share/eval.py126 # Check if scores contain NaN (values)COMMENT
LOWrdagent/components/coder/data_science/share/eval.py167 # Check if the content has changedCOMMENT
LOWrdagent/components/coder/finetune/__init__.py71 # Check if proposal decided to skip data processing (reuse SOTA's data processing script)COMMENT
LOWrdagent/components/coder/model_coder/benchmark/eval.py69 # Check if it is not a good evaluation!!COMMENT
LOW⚡…narios/qlib/experiment/factor_template/read_exp_res.py31 # Check if the recorder has a valid end timeCOMMENT
LOW⚡…narios/qlib/experiment/factor_template/read_exp_res.py40# Check if the latest recorder is foundCOMMENT
LOW⚡…enarios/qlib/experiment/model_template/read_exp_res.py31 # Check if the recorder has a valid end timeCOMMENT
LOW⚡…enarios/qlib/experiment/model_template/read_exp_res.py40# Check if the latest recorder is foundCOMMENT
LOWrdagent/scenarios/kaggle/developer/feedback.py59 # Check if there are any based experimentsCOMMENT
LOWrdagent/scenarios/kaggle/experiment/prompts.yaml164 params = ... # Set parameters to XGBoost modelCODE
LOW…riment/templates/meta_tpl_deprecated/model/model_nn.py8# Check if a GPU is availableCOMMENT
LOW…a_science/example/eval/playground-series-s4e9/valid.py3# Check if our submission file existsCOMMENT
LOW…nce/example/eval/arf-12-hours-prediction-task/valid.py3# Check if our submission file existsCOMMENT
LOW…arios/data_science/proposal/exp_gen/trace_scheduler.py292 # Check if this experiment was successful (decision=True)COMMENT
LOW…enarios/data_science/proposal/exp_gen/select/expand.py83 # Check if we've reached the maximum number of tracesCOMMENT
LOW…enarios/data_science/proposal/exp_gen/select/expand.py131 # Check if we've reached the maximum number of tracesCOMMENT
LOW…enarios/data_science/proposal/exp_gen/select/expand.py191 # Check if we've reached the maximum number of traces before creating a new oneCOMMENT
LOWrdagent/scenarios/data_science/debug/data.py404 # Check if each file is in the "used" listCOMMENT
LOWrdagent/scenarios/finetune/scen/utils.py558 # Check if tokenizer supports <think> token for CoT trainingCOMMENT
LOWrdagent/scenarios/finetune/scen/scenario.py246 # Check if already configuredCOMMENT
LOWrdagent/scenarios/finetune/benchmark/benchmark.py179 # Check if we need to merge the model (e.g. vLLM doesn't support LoRA with modules_to_save)COMMENT
LOWrdagent/scenarios/finetune/benchmark/benchmark.py253 # Check if results already exist (skip re-running if cached)COMMENT
LOWrdagent/scenarios/finetune/benchmark/data/default.py87 # Check if it's conversation formatCOMMENT
LOWrdagent/scenarios/finetune/train/eval.py97 # Check if FT_YAML_FILE_NAME existsCOMMENT
LOWrdagent/log/ui/utils.py913 # Check if G is a path (a single line)COMMENT
LOWrdagent/log/ui/app.py588 # # Check if metric series exists and has the matching roundCOMMENT
LOWrdagent/log/ui/app.py912 # Display resultsSTRING
Modern AI Meta-Vocabulary16 hits · 50 pts
SeverityFileLineSnippetContext
MEDIUMREADME.md56 <a href="https://www.microsoft.com/en-us/research/publication/agent2-rl-bench-can-llm-agents-engineer-agentic-rl-post-CODE
MEDIUMREADME.md245 # EMBEDDING MODEL: Using SiliconFlow for embedding since deepseek has no embedding model.COMMENT
MEDIUMtest/finetune/test_benchmark.py194 # PANORAMA - Patent Analysis (zero-shot)COMMENT
MEDIUMtest/finetune/test_benchmark_api.py145 # Limit few-shot examples to avoid index out of rangeCOMMENT
MEDIUMtest/finetune/test_benchmark_api.py441 # PANORAMA - Patent Analysis (zero-shot)COMMENT
MEDIUMrdagent/core/evolving_agent.py147 # 1. RAGCOMMENT
MEDIUMrdagent/app/kaggle/conf.py54 # Conditionally set the knowledge_base based on the use of graph RAGCOMMENT
MEDIUMrdagent/app/data_science/conf.py146 # constrains the number of SOTA experiments to retrieve, otherwise too many SOTA experiments to retrieve will cause COMMENT
MEDIUMrdagent/oai/utils/embedding.py12# Common embedding model token limitsCOMMENT
MEDIUM⚡…nt/scenarios/data_science/proposal/exp_gen/proposal.py399 # - after we know the selected component, we can use RAG.COMMENT
MEDIUM…nt/scenarios/data_science/proposal/exp_gen/proposal.py334 # - Extra RAGCOMMENT
MEDIUMrdagent/scenarios/finetune/benchmark/data/default.py92 # Multi-turn few-shot: format with ChatML-style markersCOMMENT
MEDIUMrdagent/scenarios/finetune/benchmark/data/adaptor.py56 # PANORAMA - Patent Analysis Benchmarks (zero-shot)COMMENT
MEDIUMrdagent/scenarios/finetune/datasets/financeiq/split.py35 # FinanceIQ specific: 'dev' folder is small and used for few-shot.COMMENT
MEDIUM⚡…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py71# 任务类型 → few-shot prompt key 的映射COMMENT
MEDIUM…/scenarios/rl/autorl_bench/benchmarks/alfworld/eval.py279 # --- 加载 ReAct few-shot prompts ---COMMENT
Slop Phrases25 hits · 48 pts
SeverityFileLineSnippetContext
MEDIUMrdagent/components/coder/factor_coder/factor.py144 # TODO you can change the name of the data folder for a better understandingCOMMENT
MEDIUM⚡rdagent/scenarios/kaggle/experiment/prompts.yaml247 optimizer = torch.optim.Adam(model.parameters(), lr=0.01) # Example optimizer, you can use any optimizerCODE
MEDIUM⚡rdagent/scenarios/kaggle/experiment/prompts.yaml248 criterion = torch.nn.CrossEntropyLoss() # Example loss function, you can use any loss functionCODE
MEDIUM⚡…val/annotators_gpt52_fn/annotations_seed0_configs.json1036 "output_1":"Escargot is a classic French delicacy made from cooked land snails. It is often served as an appetizer iCODE
MEDIUM⚡…val/annotators_gpt52_fn/annotations_seed0_configs.json1037 "output_2":"Making escargot is a delightful culinary experience, and it involves a few key steps. Here\u2019s a basiCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json17 "output_2":"The names of the states in the United States are derived from various historical and cultural factors, bCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json125 "output_2":"As an AI language model, I don't have personal experiences or emotions, but I can provide you with some CODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json148 "output_1":"Certainly! Canjeero, also known as Anjero, Laxoox or Somali pancake, is a traditional Somali dish similaCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json149 "output_2":"Certainly! Canjeero is a traditional Somali dish that is known for its rich, flavorful broth and hearty CODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json197 "output_2":"As an AI language model, I don't have personal experiences or beliefs, but I can provide you with some gCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json221 "output_2":"As an AI language model, I don't have the ability to perceive or feel physical sensations, including theCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json281 "output_2":"As an AI language model, I don't have a physical form or a personal experience, so I don't become an autCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json436 "output_1":"Escargot is a classic French delicacy made from cooked land snails. It is often served as an appetizer iCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json473 "output_2":"As an AI language model, I don't have personal experiences or emotions, but I can provide some general iCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json497 "output_2":"As an AI language model, I don't have the ability to physically use a phone, but I can provide you with CODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json533 "output_2":"As an AI language model, I don't have personal beliefs or emotions, but I can provide some general inforCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json557 "output_2":"As an AI language model, I don't have personal beliefs or opinions, but I can provide some general inforCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json568 "output_1":"Certainly! Mine Frite, which means \"fried noodles\" in English, is a popular street food dish in MauritCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json748 "output_1":"Certainly! Canjeero, also known as Anjero, Laxoox or Somali pancake, is a traditional Somali dish similaCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json1061 "output_2":"Of course! Tahini is a delicious and versatile ingredient that can be used in many dishes. Here's a simpCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json1168 "output_1":"Certainly! Mine Frite, which means \"fried noodles\" in English, is a popular street food dish in MauritCODE
LOW⚡…val/annotators_gpt52_fn/annotations_seed0_configs.json1036 "output_1":"Escargot is a classic French delicacy made from cooked land snails. It is often served as an appetizer iCODE
LOW…val/annotators_gpt52_fn/annotations_seed0_configs.json232 "output_1":"Roasting a whole pig is a traditional and festive way to celebrate for many cultures, and it can be an eCODE
LOW…val/annotators_gpt52_fn/annotations_seed0_configs.json436 "output_1":"Escargot is a classic French delicacy made from cooked land snails. It is often served as an appetizer iCODE
LOW…val/annotators_gpt52_fn/annotations_seed0_configs.json832 "output_1":"Roasting a whole pig is a traditional and festive way to celebrate for many cultures, and it can be an eCODE
Cross-Language Confusion6 hits · 38 pts
SeverityFileLineSnippetContext
HIGHrdagent/components/agent/context7/conf.py10 bun i && bun run buildSTRING
HIGHrdagent/scenarios/qlib/developer/utils.py54 if all(first_values.equals(values) for values in candidate_values[1:]):CODE
HIGH⚡rdagent/scenarios/shared/get_runtime_info.py22 strace_check = implementation.execute(env=env, entry="which strace || echo MISSING").strip()CODE
HIGH⚡rdagent/scenarios/shared/get_runtime_info.py27 coverage_check = implementation.execute(env=env, entry="python -m coverage --version || echo MISSING").strip()CODE
HIGHrdagent/log/mle_summary.py228# "score": null,COMMENT
HIGH⚡rdagent/log/ui/st_fixed_container.py34let lastBackgroundColor = null;CODE
Docstring Block Structure6 hits · 30 pts
SeverityFileLineSnippetContext
HIGHrdagent/utils/env.py72 Extract the first directory component from a relative path string. This is used to get the basename from path STRING
HIGHrdagent/components/coder/CoSTEER/evaluators.py56 Validates and converts the 'final_decision' field in the given data dictionary. Args: dataSTRING
HIGH…/kaggle/tpl_ex/aerial-cactus-identification/feature.py11 Perform feature engineering on the input data. Parameters: - X: np.ndarray The input data to be trSTRING
HIGHrdagent/scenarios/data_science/scen/utils.py339 Generate a tree structure of files in a directory. Args: path: Target directory path STRING
HIGHrdagent/scenarios/finetune/scen/utils.py831Scan datasets directory and return top-level dataset names not yet in existing_config. Only scans first-level direcSTRING
HIGHrdagent/scenarios/rl/autorl_bench/core/server.py114 提交模型评测 Args: model_path: 模型路径 gpu: 指定 GPU（如 "0", "1", "0,1"），必须是 CUDA_VISIBLE_STRING
AI Slop Vocabulary10 hits · 27 pts
SeverityFileLineSnippetContext
MEDIUMtest/notebook/test_util.py1667 # Defensive import; fallback to the most robust method for v1.4.15COMMENT
MEDIUMtest/notebook/testfiles/main_missing_sections.py89 # Defensive import; fallback to the most robust method for v1.4.15COMMENT
MEDIUMtest/notebook/testfiles/main_missing_main_fn.py89 # Defensive import; fallback to the most robust method for v1.4.15COMMENT
MEDIUMtest/notebook/testfiles/main.py89 # Defensive import; fallback to the most robust method for v1.4.15COMMENT
MEDIUMrdagent/scenarios/data_science/loop.py72 # rsync is more robust choice, but it is not installed in some docker images.COMMENT
MEDIUMrdagent/scenarios/finetune/train/eval.py153 # Combine data processing and training stdout for comprehensive feedbackCOMMENT
MEDIUMrdagent/scenarios/finetune/train/eval.py214 # Build comprehensive result with training metrics and benchmark resultsCOMMENT
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json460 "output_1":"Certainly! Tahini is a paste made from sesame seeds and is quite easy to make at home. You just need sesCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json845 "output_2":"Certainly! When choosing an electric saw, it's important to consider factors such as the type of saw youCODE
MEDIUM…val/annotators_gpt52_fn/annotations_seed0_configs.json1060 "output_1":"Certainly! Tahini is a paste made from sesame seeds and is quite easy to make at home. You just need sesCODE
Over-Commented Block25 hits · 25 pts
SeverityFileLineSnippetContext
LOWtest/finetune/test_benchmark.py181 results_summary = {}COMMENT
LOWweb/src/utils/snap.svg-min.js1// Snap.svg 0.5.0COMMENT
LOWweb/src/router/index.ts41 // },COMMENT
LOWweb/src/router/index.ts61// query: { redirect: to.fullPath }COMMENT
LOWrdagent/app/finetune/llm/job/run_ft_job.sh1#!/bin/bashCOMMENT
LOWrdagent/components/loader/task_loader.py21 # self.json_uri = json_uriCOMMENT
LOWrdagent/components/loader/task_loader.py41COMMENT
LOWrdagent/components/coder/CoSTEER/evaluators.py21# 2. If it proves to be useful, relocate it to a more general location.COMMENT
LOWrdagent/components/coder/CoSTEER/evaluators.py301 task_li_feedback_li.append(multi_implementation_feedback)COMMENT
LOWrdagent/components/coder/data_science/ensemble/conf.py1# Configuration file for ensemble componentCOMMENT
LOWrdagent/scenarios/kaggle/kaggle_crawler.py401COMMENT
LOWrdagent/scenarios/data_science/scen/utils.py361 except Exception as e:COMMENT
LOWrdagent/scenarios/data_science/scen/utils.py381 #COMMENT
LOW…rios/data_science/proposal/exp_gen/planner/__init__.py41 # elif DS_RD_SETTING.merge_hours > 0:COMMENT
LOW…/rl/autorl_bench/benchmarks/humaneval/requirements.txt1# HumanEval benchmark 额外依赖COMMENT
LOW…os/rl/autorl_bench/benchmarks/webshop/requirements.txt1# WebShop benchmark 依赖COMMENT
LOW…narios/rl/autorl_bench/benchmarks/deepsearchqa/eval.py201 # for step in range(max_steps):COMMENT
LOW…narios/rl/autorl_bench/benchmarks/deepsearchqa/eval.py221 # observation = search_fn(action_content)COMMENT
LOWrdagent/log/mle_summary.py221 print(f"Old {save_name} removed.")COMMENT
LOWrdagent/log/ui/llm_st.py121COMMENT
LOWrdagent/log/ui/llm_st.py141# st.json(cxt)COMMENT
LOWrdagent/log/ui/llm_st.py161# rdict.pop("spec")COMMENT
LOWrdagent/log/ui/app.py581 if mem := state.msgs[0]["load_experiment"]:COMMENT
LOWrdagent/log/ui/app.py601 # 'Sharpe': float(f"{metric['1day.excess_return_with_cost.annualized_return'] / abs(metric['1day.excCOMMENT
LOWrdagent/log/server/debug_app.py61 # try:COMMENT
Modern Structural Boilerplate22 hits · 22 pts
SeverityFileLineSnippetContext
LOWrdagent/core/proposal.py200 def set_current_selection(self, selection: tuple[int, ...]) -> None:CODE
LOWrdagent/core/experiment.py468 def set_user_instructions(self, user_instructions: UserInstructions | None) -> None:CODE
LOWrdagent/core/utils.py107 def set_seed(self, seed: int) -> None:CODE
LOWrdagent/app/finetune/llm/ui/benchmarks/__init__.py62__all__ = [CODE
LOWrdagent/utils/env.py231 def update_stdout(self, stdout: str) -> None:CODE
LOWrdagent/utils/env.py763 def _update_bin_path(self) -> None:CODE
LOWrdagent/utils/agent/__init__.py3__all__ = ["build_cls_from_json_with_retry"]CODE
LOWrdagent/utils/workflow/__init__.py5__all__ = ["LoopBase", "LoopMeta", "WorkflowTracker", "wait_retry"]CODE
LOWrdagent/components/benchmark/utils.py10logger = logging.getLogger(__name__)CODE
LOWrdagent/scenarios/kaggle/proposal/proposal.py200 def update_reward_estimates(self, trace: Trace) -> None:STRING
LOW…nt/scenarios/data_science/proposal/exp_gen/__init__.py3__all__ = ["DSTrace"]CODE
LOW…t/scenarios/data_science/proposal/exp_gen/idea_pool.py182 def update_pickled_problem(self, problems: Dict, pickled_problem_name: str) -> None:CODE
LOWrdagent/scenarios/data_science/proposal/exp_gen/base.py88 def set_sota_exp_to_submit(self, exp: DSExperiment) -> None:CODE
LOWrdagent/scenarios/data_science/experiment/experiment.py42 def set_local_selection(self, local_selection: tuple[int, ...]) -> None:CODE
LOWrdagent/scenarios/finetune/benchmark/__init__.py3__all__ = ["get_benchmark_ranges", "run_benchmark"]CODE
LOWrdagent/scenarios/rl/autorl_bench/core/__init__.py63__all__ = [CODE
LOWrdagent/scenarios/rl/autorl_bench/agents/__init__.py3__all__ = ["get_agent", "list_agents"]CODE
LOW…scenarios/rl/autorl_bench/benchmarks/smith/__init__.py16logger = logging.getLogger(__name__)CODE
LOW…enarios/rl/autorl_bench/benchmarks/webshop/__init__.py6__all__ = ["WebShopEvaluator", "download_train_data"]CODE
LOWrdagent/log/timer.py63 def update_remain_time(self) -> None:CODE
LOWrdagent/log/conf.py22 def set_ui_server_port(self, port: int | None) -> None:CODE
LOWrdagent/log/logger.py106 def set_storages_path(self, path: str | Path) -> None:CODE
AI Structural Patterns15 hits · 15 pts
SeverityFileLineSnippetContext
LOWtest/finetune/test_benchmark_api.py94CODE
LOWtest/finetune/test_benchmark_api.py180CODE
LOWrdagent/app/cli.py150CODE
LOWrdagent/app/data_science/loop.py13CODE
LOWrdagent/app/finetune/llm/loop.py17CODE
LOWrdagent/oai/backend/base.py625CODE
LOWrdagent/components/coder/CoSTEER/evaluators.py128CODE
LOW…mponents/coder/model_coder/benchmark/gt_code/visnet.py752CODE
LOW…mponents/coder/model_coder/benchmark/gt_code/visnet.py1077CODE
LOW…ponents/coder/model_coder/benchmark/gt_code/gpsconv.py59CODE
LOWrdagent/scenarios/kaggle/kaggle_crawler.py273CODE
LOW…t/scenarios/kaggle/knowledge_management/vector_base.py22CODE
LOW…t/scenarios/kaggle/knowledge_management/vector_base.py281CODE
LOWrdagent/scenarios/data_science/proposal/exp_gen/base.py14CODE
LOWrdagent/scenarios/finetune/train/eval.py330CODE
Dead Code7 hits · 14 pts
SeverityFileLineSnippetContext
MEDIUM…agent/components/coder/CoSTEER/knowledge_management.py187CODE
MEDIUM…agent/components/coder/CoSTEER/knowledge_management.py229CODE
MEDIUM…agent/components/coder/CoSTEER/knowledge_management.py230CODE
MEDIUM…agent/components/coder/CoSTEER/knowledge_management.py231CODE
MEDIUM…agent/components/coder/CoSTEER/knowledge_management.py233CODE
MEDIUM…agent/components/coder/CoSTEER/knowledge_management.py234CODE
MEDIUM…agent/components/coder/CoSTEER/knowledge_management.py280CODE
Magic Placeholder Names3 hits · 12 pts
SeverityFileLineSnippetContext
HIGHrdagent/app/finetune/llm/README.md58OPENAI_API_KEY=<your_api_key>CODE
HIGHrdagent/scenarios/finetune/benchmark/benchmark.py11FT_JUDGE_API_KEY="<your_api_key>"STRING
HIGHrdagent/scenarios/rl/autorl_bench/README.md75OPENAI_API_KEY=your_api_keyCODE
Overly Generic Function Names7 hits · 7 pts
SeverityFileLineSnippetContext
LOWtest/notebook/test_util.py997 "def process_data(data):",CODE
LOWtest/notebook/test_util.py1071 "def process_data(data):",CODE
LOWtest/notebook/test_util.py1155 "def process_data(data):",CODE
LOWtest/notebook/test_util.py1216 "def process_data(data):",CODE
LOWtest/notebook/test_util.py1240 "def process_data(data):",CODE
LOWtest/notebook/test_util.py1326 "def process_data(data):",CODE
LOW…l-iceberg-classifier-challenge/fea_share_preprocess.py20 def process_data(df):CODE
Example Usage Blocks2 hits · 2 pts
SeverityFileLineSnippetContext
LOWrdagent/app/utils/ape.py22# Example usageCOMMENT
LOWrdagent/log/base.py29 # Usage:STRING
TODO Padding1 hit · 2 pts
SeverityFileLineSnippetContext
LOW…enarios/data_science/proposal/exp_gen/select/expand.py244# TODO: implement these selectors and moreCOMMENT