Home / microsoft / SkillOpt
Repository Analysis

microsoft/SkillOpt

SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.

25.6 Moderate AI signal View on GitHub

Analysis Overview

This report presents the forensic synthetic code analysis of microsoft/SkillOpt, a Python project with 16,445 GitHub stars. SynthScan v2.0 examined 108,194 lines of code across 397 source files, recording 2256 pattern matches distributed across 16 syntactic categories. The overall adjusted score of 25.6 places this repository in the Moderate AI signal band.

The scanner applied 160+ deterministic lexical heuristics, multi-line block detectors, abstract syntax tree depth profilers, and a cross-file Jaccard similarity matrix to construct a statistically normalised synthetic code estimate. All matches are individually weighted by severity coefficient and contextual multiplier before summation, and the resulting headline score is temporally discounted to account for the repository's development history relative to the commercial emergence of large language model coding tooling (November 2022 onward).

25.6
Adjusted Score
25.6
Raw Score
100%
Time Factor
2026-08-26
Last Push
16.4K
Stars
Python
Language
108.2K
Lines of Code
397
Files
2.3K
Pattern Hits
2026-08-29
Scan Date
0.02
HC Hit Rate

What These Metrics Mean

Adjusted Score
Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score
The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor
The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits
Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate
High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files
Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.

Severity Breakdown

Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.

CRITICAL 0HIGH 9MEDIUM 255LOW 1992

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 2256 distinct pattern matches across 16 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 Identifiers1329 hits · 1294 pts
SeverityFileLineSnippetContext
LOWplugins/devin/harvest_devin.py67def _devin_transcript_candidates() -> List[str]:CODE
LOWplugins/devin/harvest_devin.py106def _workspaces_from_registry(storage_root: str) -> List[tuple]:CODE
LOWplugins/devin/harvest_devin.py333def harvest_devin_transcripts(CODE
LOW⚡tests/test_json_utils.py47 def test_no_json_returns_none(self) -> None:CODE
LOW⚡tests/test_json_utils.py50 def test_empty_string_returns_none(self) -> None:CODE
LOW⚡tests/test_json_utils.py53 def test_malformed_json_returns_none(self) -> None:CODE
LOW⚡tests/test_json_utils.py59 def test_json_with_escaped_chars(self) -> None:CODE
LOW⚡tests/test_json_utils.py63 def test_only_fence_with_no_json_syntax(self) -> None:CODE
LOW⚡tests/test_json_utils.py75 def test_two_top_level_objects(self) -> None:CODE
LOW⚡tests/test_json_utils.py78 def test_brace_inside_quoted_prose_is_ignored(self) -> None:CODE
LOW⚡tests/test_json_utils.py83 def test_real_object_after_quoted_brace(self) -> None:CODE
LOW⚡tests/test_json_utils.py95 def test_two_top_level_arrays(self) -> None:CODE
LOW⚡tests/test_json_utils.py98 def test_array_inside_object_is_ignored(self) -> None:CODE
LOW⚡tests/test_json_utils.py101 def test_bracket_inside_quoted_prose_is_ignored(self) -> None:CODE
LOW⚡tests/test_json_utils.py104 def test_unmatched_prose_brace_does_not_hide_later_array(self) -> None:CODE
LOW⚡tests/test_json_utils.py108 def test_many_unmatched_braces_scan_linearly(self) -> None:CODE
LOW⚡tests/test_json_utils.py145 def test_json_string_values_with_quotes_still_repair(self) -> None:CODE
LOW⚡tests/test_json_utils.py152 def test_no_warning_on_quoted_prose(self, recwarn: pytest.WarningsRecorder) -> None:CODE
LOW⚡tests/test_json_utils.py159 def test_no_warning_on_plain_text(self, recwarn: pytest.WarningsRecorder) -> None:CODE
LOW⚡tests/test_json_utils.py165 def test_trailing_comma_repaired_when_available(self) -> None:CODE
LOW⚡tests/test_json_utils.py170 def test_two_malformed_objects_too_ambiguous(self) -> None:CODE
LOW⚡tests/test_json_utils.py205 def test_no_array_returns_none(self) -> None:CODE
LOW⚡tests/test_json_utils.py208 def test_empty_string_returns_none(self) -> None:CODE
LOW⚡tests/test_json_utils.py211 def test_malformed_array_returns_none(self) -> None:CODE
LOW⚡tests/test_json_utils.py221 def test_object_not_confused_with_array(self) -> None:CODE
LOW⚡tests/test_json_utils.py226 def test_ignores_prose_brackets_before_valid_array(self) -> None:CODE
LOW⚡tests/test_json_utils.py230 def test_multiple_valid_arrays_are_ambiguous(self) -> None:CODE
LOW⚡tests/test_json_utils.py234 def test_nested_object_array_not_confused_with_array_answer(self) -> None:CODE
LOW⚡tests/test_json_utils.py238 def test_unmatched_prose_brace_before_valid_array(self) -> None:CODE
LOWtests/test_json_utils.py25 def test_code_fence_takes_precedence(self) -> None:CODE
LOWtests/test_json_utils.py33 def test_broken_fence_falls_back_to_bare(self) -> None:CODE
LOWtests/test_json_utils.py127 def test_prose_pseudo_json_returns_none(self) -> None:CODE
LOWtests/test_json_utils.py133 def test_single_quoted_and_backticked_prose_returns_none(self) -> None:CODE
LOWtests/test_json_utils.py187 def test_code_fence_takes_precedence(self) -> None:CODE
LOWtests/test_json_utils.py194 def test_broken_fence_falls_back_to_bare(self) -> None:CODE
LOWtests/test_hierarchical_merge.py29 def test_batch_size_one_terminates_and_merges(self) -> None:CODE
LOWtests/test_hierarchical_merge.py55 def test_batch_size_zero_terminates(self) -> None:CODE
LOWtests/test_hierarchical_merge.py75 def test_normal_batch_size_still_merges(self) -> None:CODE
LOW⚡tests/test_split_hardening_2x3.py46 def test_legacy_holdout_archived_task_not_recalled(self):CODE
LOW⚡tests/test_split_hardening_2x3.py51 def test_archived_val_not_recalled(self):CODE
LOW⚡tests/test_split_hardening_2x3.py56 def test_unrelated_archived_train_can_still_be_recalled(self):CODE
LOWtests/test_split_hardening_2x3.py68 def test_zero_similarity_returns_empty(self):CODE
LOWtests/test_split_hardening_2x3.py77 def test_dream_tasks_always_train_even_with_high_test_fraction(self):CODE
LOWtests/test_split_hardening_2x3.py90 def test_real_tasks_have_exactly_one_split_label(self):CODE
LOWtests/test_split_hardening_2x3.py101 def test_hash_assigned_test_not_demoted_for_val_top_up(self):CODE
LOW⚡tests/test_split_hardening_2x3.py124 def test_assign_splits_rejects_negative_test_fraction(self):CODE
LOW⚡tests/test_split_hardening_2x3.py128 def test_assign_splits_rejects_fraction_sum_ge_one(self):CODE
LOW⚡tests/test_split_hardening_2x3.py132 def test_mine_forwards_invalid_fractions_to_assign_splits(self):CODE
LOW⚡tests/test_split_hardening_2x3.py142 def test_recall_rows_are_train_split_only(self):CODE
LOWtests/test_split_hardening_2x3.py187 def test_archived_test_never_recalled_through_cycle(self):CODE
LOWtests/test_split_hardening_2x3.py240 def test_on_disk_zero_val_beats_holdout_alias(self):CODE
LOW⚡tests/test_backend_opencode.py105def _successful_plain_results(*mcp_names: str, answer: str = "answer") -> list[_FakeProc]:CODE
LOW⚡tests/test_backend_opencode.py113def _assert_replay_permissions(config, env, agent_name: str, expected_permission: dict[str, str]) -> None:CODE
LOW⚡tests/test_backend_opencode.py119def _assert_controlled_tool_environment(env, work: str) -> None:CODE
LOW⚡tests/test_backend_opencode.py167def test_resolve_opencode_path_precedence(monkeypatch):CODE
LOW⚡tests/test_backend_opencode.py174def test_resolve_opencode_path_falls_back_to_command(monkeypatch):CODE
LOW⚡tests/test_backend_opencode.py181def test_resolve_opencode_path_preserves_windows_cmd_shim(monkeypatch):CODE
LOW⚡tests/test_backend_opencode.py191def test_resolve_opencode_path_anchors_relative_paths(monkeypatch, tmp_path, which_result, source):CODE
LOW⚡tests/test_backend_opencode.py696def test_failed_call_is_not_cached():CODE
LOW⚡tests/test_backend_opencode.py704def test_cached_success_clears_stale_call_error():CODE
1269 more matches not shown…
Decorative Section Separators234 hits · 730 pts
SeverityFileLineSnippetContext
MEDIUM⚡requirements.txt1# ── Core ──────────────────────────────────────────COMMENT
MEDIUM⚡requirements.txt10# ── Optional: ALFWorld benchmark ──────────────────COMMENT
MEDIUM⚡requirements.txt14# ── Optional: Claude model backend ────────────────COMMENT
MEDIUM⚡requirements.txt17# ── Optional: Qwen local model (via vLLM) ────────COMMENT
MEDIUM⚡requirements.txt26# ── Optional: WebUI dashboard ────────────────────COMMENT
MEDIUM⚡requirements.txt29# ── Optional: Documentation site ─────────────────COMMENT
MEDIUMskillopt_webui/app.py28# ─── Config helpers ──────────────────────────────────────────────────────────COMMENT
MEDIUMskillopt_webui/app.py174# ─── Training process management ────────────────────────────────────────────COMMENT
MEDIUMskillopt_webui/app.py435# ─── Pipeline Stage HTML ────────────────────────────────────────────────────COMMENT
MEDIUMskillopt_webui/app.py466# ─── Gradio UI ──────────────────────────────────────────────────────────────COMMENT
MEDIUMskillopt_webui/app.py478 # ── Tab 1: Configure & Launch ────────────────────────────COMMENT
MEDIUMskillopt_webui/app.py554 # ── Tab 2: Monitor ───────────────────────────────────────COMMENT
MEDIUMskillopt_webui/app.py585 # ── Tab 3: Results ───────────────────────────────────────COMMENT
MEDIUMplugins/devin/harvest_devin.py44# ── cross-platform path resolution (Linux + Windows + macOS) ──────────────────COMMENT
MEDIUMplugins/devin/harvest_devin.py104# ── workspace auto-detection ─────────────────────────────────────────────────COMMENT
MEDIUMplugins/devin/harvest_devin.py149# ── helpers ───────────────────────────────────────────────────────────────────COMMENT
MEDIUMplugins/devin/harvest_devin.py213# ── task identity + outcome extraction (fuel for the validation gate) ─────────COMMENT
MEDIUMplugins/devin/harvest_devin.py331# ── source 1: Devin ATIF-v1.7 transcripts ────────────────────────────────────COMMENT
MEDIUMplugins/devin/harvest_devin.py412# ── source 2: agentmemory ─────────────────────────────────────────────────────COMMENT
MEDIUMplugins/devin/harvest_devin.py438# ── source 3: skill files (.devin/skills) ─────────────────────────────────────COMMENT
MEDIUMplugins/devin/harvest_devin.py473# ── main ─────────────────────────────────────────────────────────────────────COMMENT
MEDIUMplugins/claude-code/scripts/install-cron.sh17# ── SkillOpt-Sleep nightly cycle ────────────────────────────────────────────COMMENT
MEDIUMplugins/claude-code/scripts/install-cron.sh29# ────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMplugins/openclaw/skillopt_sleep_openclaw.py84# ── Backend implementation ────────────────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_scheduler.py34# ── ConstantScheduler ────────────────────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_scheduler.py79# ── LinearScheduler ──────────────────────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_scheduler.py165# ── CosineScheduler ──────────────────────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_scheduler.py267# ── AutonomousScheduler ──────────────────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_scheduler.py298# ── build_scheduler factory ──────────────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_scheduler.py332# ── Abstract base class ──────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_types.py8# ── Edit ────────────────────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_types.py140# ── Patch ───────────────────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_types.py251# ── RolloutResult ────────────────────────────────────────────────────────────COMMENT
MEDIUMdocs/guide/new-benchmark.md239 # ── Lifecycle ───────────────────────────────────────────────────────COMMENT
MEDIUMdocs/guide/new-benchmark.md248 # ── Env construction ────────────────────────────────────────────────COMMENT
MEDIUMdocs/guide/new-benchmark.md266 # ── The rollout method (reflect is inherited) ───────────────────────COMMENT
MEDIUMdocs/sleep/examples/runner.py32# ----------------------------------------------------------------------------COMMENT
MEDIUMdocs/reference/api.md30 # ── Lifecycle hooks (have defaults; override only if needed) ────────COMMENT
MEDIUMdocs/reference/api.md40 # ── Abstract methods (subclasses MUST implement) ────────────────────COMMENT
MEDIUMconfigs/features/soft_gate.yaml1# ─────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMconfigs/features/soft_gate.yaml3# ─────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMconfigs/features/soft_gate.yaml33# ─────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUM⚡scripts/run_spreadsheetbench.sh2# ──────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUM⚡scripts/run_spreadsheetbench.sh8# ──────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUM⚡scripts/run_searchqa.sh2# ──────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUM⚡scripts/run_searchqa.sh9# ──────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMscripts/eval_only.py67# ── Reuse registry from train.py ───────────────────────────────────────────COMMENT
MEDIUMscripts/eval_only.py147# ── CLI ────────────────────────────────────────────────────────────────────COMMENT
MEDIUMscripts/train.py41# ── Environment registry ────────────────────────────────────────────────────COMMENT
MEDIUMscripts/train.py128# ── CLI ──────────────────────────────────────────────────────────────────────COMMENT
MEDIUMscripts/train.py303# ── Retired CLI options ──────────────────────────────────────────────────────COMMENT
MEDIUMscripts/train.py759# ── Main ─────────────────────────────────────────────────────────────────────COMMENT
MEDIUM⚡scripts/run_alfworld.sh2# ──────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUM⚡scripts/run_alfworld.sh13# ──────────────────────────────────────────────────────────────────────────────COMMENT
MEDIUMskillopt_sleep/handoff_backend.py83 # ── prompt/answer plumbing ────────────────────────────────────────────COMMENT
MEDIUMskillopt_sleep/handoff_backend.py103 # ── handoff file emission ─────────────────────────────────────────────COMMENT
MEDIUMskillopt_sleep/cycle.py92# ── Model-swap detection (F16) ───────────────────────────────COMMENT
MEDIUMskillopt_sleep/cycle.py671 # ── evidence log (the night's full evidentiary chain) ────────────────COMMENT
MEDIUMskillopt_sleep/cycle.py708 # ── live skill/memory docs ───────────────────────────────────────────COMMENT
MEDIUMskillopt_sleep/cycle.py734 # ── 1+2. harvest + mine (unless seed_tasks injected) ─────────────────COMMENT
174 more matches not shown…
Unused Imports256 hits · 248 pts
SeverityFileLineSnippetContext
LOWplugins/devin/harvest_devin.py31CODE
LOWplugins/devin/harvest_devin.py40CODE
LOWplugins/devin/judge.py23CODE
LOWplugins/devin/mcp_server.py21CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py10CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py15CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py16CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py16CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py18CODE
LOWplugins/openclaw/slash_sleep.py16CODE
LOWplugins/openclaw/slash_sleep.py24CODE
LOWplugins/openclaw/run_sleep.py17CODE
LOWplugins/copilot/mcp_server.py18CODE
LOWplugins/copilot/skillopt/mcp_server.py23CODE
LOWtests/test_json_utils.py2CODE
LOWtests/test_hierarchical_merge.py6CODE
LOWtests/test_split_hardening_2x3.py13CODE
LOWtests/test_backend_opencode.py3CODE
LOWtests/test_copilot_tool_scope.py23CODE
LOWtests/test_sleep_engine.py7CODE
LOWtests/test_holdout_integrity.py1CODE
LOWtests/test_claude_code_exec_resolution.py9CODE
LOWtests/test_outcome_judges.py1CODE
LOWtests/test_role_backend_resolution.py3CODE
LOWtests/test_env_section_survives_dedup.py15CODE
LOWtests/test_sleep_staging_fanout.py6CODE
LOWtests/test_sleep_skill_resolver.py6CODE
LOWtests/test_claude_backend_tempdir.py2CODE
LOWtests/test_model_change_warning.py2CODE
LOWtests/test_sleep_adopt_skill_subset.py6CODE
LOWtests/test_react_agent_no_shell.py9CODE
LOWtests/test_console_encoding.py3CODE
LOWtests/test_sleep_skill_hint.py6CODE
LOWtests/test_gate_no_regression.py3CODE
LOWtests/test_azure_openai_compat.py14CODE
LOWtests/test_azure_openai_compat.py32CODE
LOWtests/test_consolidation_edit_bookkeeping.py3CODE
LOWtests/test_harvest_opencode_live.py10CODE
LOWtests/test_harvest_opencode.py3CODE
LOWtests/test_backend_opencode_live.py13CODE
LOWtests/test_scheduler.py20CODE
LOWtests/test_scoring.py2CODE
LOWtests/test_sleep_dream_metric.py2CODE
LOWtests/test_consolidate_split.py6CODE
LOWtests/test_cursor_exec_backend.py1CODE
LOWtests/test_staging_redaction_azure.py7CODE
LOWtests/test_openai_compatible_backend.py3CODE
LOWtests/test_minimax_region.py3CODE
LOWtests/test_sleep_multi_skill.py6CODE
LOWtests/test_gate.py8CODE
LOWtests/test_minimax_backend.py3CODE
LOWtests/test_split_wiring.py9CODE
LOWtests/test_codex_cli_prompt_stdin.py8CODE
LOWtests/test_harvest_copilot_cli.py1CODE
LOWtests/test_semantic_density.py2CODE
LOWtests/test_handoff_backend.py3CODE
LOWtests/test_pi_integration.py2CODE
LOWtests/test_sleep_scheduler_safety.py2CODE
LOWtests/test_sleep_evidence.py8CODE
LOWtests/test_copilot_exec_backend.py1CODE
196 more matches not shown…
Excessive Try-Catch Wrapping159 hits · 150 pts
SeverityFileLineSnippetContext
LOWskillopt_webui/app.py112 except Exception as exc:CODE
LOWskillopt_webui/app.py223 except Exception as e:CODE
LOWskillopt_webui/app.py526 except Exception as e:CODE
LOWskillopt_webui/app.py617 except Exception:CODE
LOWskillopt_webui/app.py627 except Exception:CODE
LOWplugins/devin/harvest_devin.py121 except Exception:CODE
LOWplugins/devin/harvest_devin.py346 except Exception:CODE
LOWplugins/devin/harvest_devin.py377 except Exception:CODE
LOWplugins/devin/install.sh68except Exception:CODE
LOWplugins/devin/judge.py102 except Exception as exc: # network/auth/parse — degrade gracefullyCODE
LOWplugins/devin/mcp_server.py218 except Exception as exc:CODE
LOWplugins/devin/mcp_server.py287 except Exception as e:CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py62 except Exception as e:CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py80 except Exception:CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py174 except Exception:STRING
LOWplugins/openclaw/skillopt_sleep_openclaw.py278 except Exception as e:STRING
LOWplugins/copilot/mcp_server.py250 except Exception as e:CODE
LOWplugins/copilot/skillopt/mcp_server.py95 except Exception as e: # noqa: BLE001CODE
LOWplugins/copilot/skillopt/mcp_server.py219 except Exception:CODE
LOWplugins/dsh/scripts/sleep.py65 except Exception:CODE
LOWtests/test_model_change_warning.py167 except Exception:CODE
LOWtests/test_model_change_warning.py197 except Exception:CODE
LOWtests/test_model_change_warning.py218 except Exception:CODE
LOWtests/test_model_change_warning.py243 except Exception:CODE
LOWdocs/sleep/examples/watchdog.py39 except Exception as e:CODE
LOWdocs/sleep/examples/watchdog.py49 except Exception as e:CODE
LOWskillopt_sleep/mine.py381 except Exception:CODE
LOWskillopt_sleep/cycle.py117 except Exception:CODE
LOWskillopt_sleep/cycle.py201 except Exception:CODE
LOWskillopt_sleep/cycle.py790 except Exception:CODE
LOWskillopt_sleep/cycle.py1064 except Exception:CODE
LOW⚡skillopt_sleep/backend.py1062 except Exception:CODE
LOW⚡skillopt_sleep/backend.py1069 except Exception:CODE
LOWskillopt_sleep/backend.py276 except Exception:CODE
LOWskillopt_sleep/backend.py675 except Exception as exc:CODE
LOWskillopt_sleep/backend.py681 except Exception:CODE
LOWskillopt_sleep/backend.py783 except Exception as exc:CODE
LOWskillopt_sleep/backend.py852 except Exception as exc:CODE
LOWskillopt_sleep/backend.py1120 except Exception:CODE
LOWskillopt_sleep/backend.py1259 except Exception:CODE
LOWskillopt_sleep/backend.py1479 except Exception as exc:CODE
LOWskillopt_sleep/backend.py1487 except Exception as exc:CODE
LOWskillopt_sleep/backend.py1625 except Exception as exc: # noqa: BLE001CODE
LOWskillopt_sleep/backend.py1761 except Exception:CODE
LOWskillopt_sleep/backend.py1904 except Exception:CODE
LOWskillopt_sleep/backend.py1993 except Exception:CODE
LOWskillopt_sleep/backend.py2086 except Exception as exc:CODE
LOWskillopt_sleep/backend.py2137 except Exception:CODE
LOWskillopt_sleep/backend.py2384 except Exception:CODE
LOWskillopt_sleep/backend.py2394 except Exception as e: # noqa: BLE001CODE
LOWskillopt_sleep/backend.py2493 except Exception:CODE
LOWskillopt_sleep/backend.py2498 except Exception as e: # noqa: BLE001CODE
LOWskillopt_sleep/config.py209 except Exception:CODE
LOWskillopt_sleep/config.py224 except Exception:CODE
LOWskillopt_sleep/prompts.py186 except Exception:CODE
LOWskillopt_sleep/harvest.py60 except Exception:CODE
LOWskillopt_sleep/scheduler.py28 except Exception:CODE
LOWskillopt_sleep/scheduler.py37 except Exception:CODE
LOWskillopt_sleep/scheduler.py74 except Exception:CODE
LOWskillopt_sleep/scheduler.py83 except Exception:CODE
99 more matches not shown…
Deep Nesting124 hits · 106 pts
SeverityFileLineSnippetContext
LOWskillopt_webui/app.py468CODE
LOWskillopt_webui/app.py253CODE
LOWskillopt_webui/app.py306CODE
LOWskillopt_webui/app.py599CODE
LOWplugins/devin/harvest_devin.py333CODE
LOWplugins/openclaw/skillopt_sleep_openclaw.py25CODE
LOWtests/test_split_hardening_2x3.py187CODE
LOWtests/test_sleep_engine.py1993CODE
LOWtests/test_sleep_adopt_skill_subset.py92CODE
LOWtests/test_sleep_adopt_skill_subset.py1553CODE
LOWtests/test_sleep_adopt_skill_subset.py1592CODE
LOWtests/test_sleep_adopt_skill_subset.py1826CODE
LOWtests/test_superpowers_scenarios.py752CODE
LOWscripts/eval_only.py247CODE
LOWscripts/train.py484CODE
LOWskillopt_sleep/mine.py151CODE
LOWskillopt_sleep/harvest_copilot.py106CODE
LOWskillopt_sleep/harvest_copilot.py150CODE
LOWskillopt_sleep/harvest_copilot.py231CODE
LOWskillopt_sleep/harvest_pi.py52CODE
LOWskillopt_sleep/harvest_pi.py173CODE
LOWskillopt_sleep/harvest_pi.py270CODE
LOWskillopt_sleep/cycle.py300CODE
LOWskillopt_sleep/cycle.py625CODE
LOWskillopt_sleep/slow_update.py75CODE
LOWskillopt_sleep/harvest_cursor.py116CODE
LOWskillopt_sleep/harvest_cursor.py269CODE
LOWskillopt_sleep/skill_resolver.py120CODE
LOWskillopt_sleep/backend.py913CODE
LOWskillopt_sleep/backend.py433CODE
LOWskillopt_sleep/backend.py1284CODE
LOWskillopt_sleep/backend.py1547CODE
LOWskillopt_sleep/backend.py1811CODE
LOWskillopt_sleep/backend.py2121CODE
LOWskillopt_sleep/consolidate.py121CODE
LOWskillopt_sleep/consolidate.py180CODE
LOWskillopt_sleep/llm_miner.py46CODE
LOWskillopt_sleep/llm_miner.py125CODE
LOWskillopt_sleep/llm_miner.py133CODE
LOWskillopt_sleep/memory.py87CODE
LOWskillopt_sleep/judges.py194CODE
LOWskillopt_sleep/harvest.py51CODE
LOWskillopt_sleep/harvest.py66CODE
LOWskillopt_sleep/harvest.py231CODE
LOWskillopt_sleep/harvest_opencode.py208CODE
LOWskillopt_sleep/scheduler.py87CODE
LOWskillopt_sleep/scheduler.py238CODE
LOWskillopt_sleep/staging.py215CODE
LOWskillopt_sleep/staging.py436CODE
LOWskillopt_sleep/staging.py483CODE
LOWskillopt_sleep/staging.py562CODE
LOWskillopt_sleep/staging.py1175CODE
LOWskillopt_sleep/staging.py1701CODE
LOWskillopt_sleep/staging.py1944CODE
LOWskillopt_sleep/staging.py2475CODE
LOWskillopt_sleep/staging.py2658CODE
LOWskillopt_sleep/staging.py2700CODE
LOWskillopt_sleep/staging.py2969CODE
LOWskillopt_sleep/evidence.py127CODE
LOWskillopt_sleep/__main__.py273CODE
64 more matches not shown…
AI Structural Patterns63 hits · 56 pts
SeverityFileLineSnippetContext
LOWplugins/openclaw/run_sleep.py38CODE
LOWskillopt_sleep/mine.py355CODE
LOWskillopt_sleep/backend.py2505CODE
LOWskillopt_sleep/backend.py2550CODE
LOWskillopt_sleep/consolidate.py180CODE
LOWskillopt_sleep/dream.py119CODE
LOWskillopt_sleep/experiments/run_experiment.py124CODE
LOWskillopt_sleep/experiments/run_experiment.py52CODE
LOWskillopt/gradient/reflect.py269CODE
LOWskillopt/gradient/reflect.py379CODE
LOWskillopt/gradient/reflect.py485CODE
LOWskillopt/datasets/base.py236CODE
LOWskillopt/model/minimax_backend.py345CODE
LOWskillopt/model/qwen_backend.py346CODE
LOWskillopt/model/qwen_backend.py426CODE
LOWskillopt/model/qwen_backend.py542CODE
LOWskillopt/model/qwen_backend.py568CODE
LOWskillopt/model/openai_compatible_backend.py276CODE
LOWskillopt/model/openai_compatible_backend.py302CODE
LOWskillopt/model/openai_compatible_backend.py364CODE
LOWskillopt/model/__init__.py282CODE
LOWskillopt/model/__init__.py388CODE
LOWskillopt/model/__init__.py476CODE
LOWskillopt/model/__init__.py632CODE
LOWskillopt/model/__init__.py675CODE
LOWskillopt/model/__init__.py756CODE
LOWskillopt/model/azure_openai.py600CODE
LOWskillopt/model/azure_openai.py788CODE
LOWskillopt/model/azure_openai.py816CODE
LOWskillopt/model/azure_openai.py845CODE
LOWskillopt/model/claude_code_backend.py121CODE
LOWskillopt/model/claude_code_backend.py212CODE
LOWskillopt/model/claude_code_backend.py238CODE
LOWskillopt/model/claude_code_backend.py264CODE
LOWskillopt/model/backend_config.py201CODE
LOWskillopt/model/copilot_backend.py206CODE
LOWskillopt/model/copilot_backend.py231CODE
LOWskillopt/model/router.py196CODE
LOWskillopt/envs/alfworld/adapter.py66CODE
LOWskillopt/envs/alfworld/rollout.py148CODE
LOWskillopt/envs/alfworld/dataloader.py14CODE
LOWskillopt/envs/livemathematicianbench/adapter.py42CODE
LOWskillopt/envs/livemathematicianbench/rollout.py111CODE
LOWskillopt/envs/livemathematicianbench/rollout.py290CODE
LOWskillopt/envs/livemathematicianbench/dataloader.py151CODE
LOWskillopt/envs/_template/env_template.py32CODE
LOWskillopt/envs/spreadsheetbench/react_agent.py392CODE
LOWskillopt/envs/spreadsheetbench/adapter.py29CODE
LOWskillopt/envs/spreadsheetbench/rollout.py587CODE
LOWskillopt/envs/spreadsheetbench/rollout.py841CODE
LOWskillopt/envs/spreadsheetbench/codegen_agent.py443CODE
LOWskillopt/envs/spreadsheetbench/codegen_agent.py544CODE
LOWskillopt/envs/spreadsheetbench/dataloader.py14CODE
LOWskillopt/envs/officeqa/adapter.py12CODE
LOWskillopt/envs/officeqa/tool_runtime.py441CODE
LOWskillopt/envs/officeqa/rollout.py140CODE
LOWskillopt/envs/officeqa/rollout.py717CODE
LOWskillopt/envs/officeqa/rollout.py510CODE
LOWskillopt/envs/officeqa/rollout.py718CODE
LOWskillopt/envs/docvqa/adapter.py10CODE
3 more matches not shown…
Modern Structural Boilerplate41 hits · 52 pts
SeverityFileLineSnippetContext
LOWscripts/eval_only.py372 def _set_role(key: str, value: str) -> None:CODE
LOWscripts/train.py645 def _set_role(key: str, value: str) -> None:CODE
LOWskillopt_sleep/backend.py594 def _set_call_error(self, message: object) -> None:CODE
LOWskillopt_sleep/__init__.py19__all__ = ["__version__"]CODE
LOWskillopt_sleep/evidence.py109def set_phase(backend, phase: str) -> None:CODE
LOW⚡skillopt_sleep/state.py75 def set_last_harvest(self, project: str, iso_ts: str) -> None:CODE
LOW⚡skillopt_sleep/state.py82 def set_slow_memory(self, content: str) -> None:CODE
LOWskillopt_sleep/state.py109 def set_last_model_key(self, key: str) -> None:CODE
LOWskillopt/optimizer/select.py4__all__ = ["rank_and_select"]CODE
LOWskillopt/datasets/base.py83 def set_out_root(self, out_root: str) -> None:CODE
LOWskillopt/model/minimax_backend.py378def set_reasoning_effort(effort: str | None) -> None:CODE
LOWskillopt/model/minimax_backend.py382def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/qwen_backend.py602def set_reasoning_effort(effort: str | None) -> None:CODE
LOW⚡skillopt/model/qwen_backend.py606def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/qwen_backend.py611def set_optimizer_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/openai_compatible_backend.py432def set_reasoning_effort(effort: str | None) -> None:CODE
LOW⚡skillopt/model/openai_compatible_backend.py438def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/openai_compatible_backend.py444def set_optimizer_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/__init__.py787def set_reasoning_effort(effort: str | None) -> None:CODE
LOW⚡skillopt/model/__init__.py797def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/__init__.py807def set_optimizer_deployment(deployment: str) -> None:CODE
LOWskillopt/model/azure_openai.py882def set_target_deployment(deployment: str) -> None:CODE
LOWskillopt/model/azure_openai.py898def set_reasoning_effort(effort: str | None) -> None:CODE
LOWskillopt/model/azure_openai.py909def set_optimizer_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/claude_backend.py362def set_reasoning_effort(effort: str | None) -> None:CODE
LOW⚡skillopt/model/claude_backend.py367def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/claude_backend.py373def set_optimizer_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/claude_code_backend.py299def set_reasoning_effort(effort: str | None) -> None:CODE
LOW⚡skillopt/model/claude_code_backend.py304def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/claude_code_backend.py310def set_optimizer_deployment(deployment: str) -> None:CODE
LOWskillopt/model/backend_config.py125def set_optimizer_backend(backend: str) -> None:CODE
LOWskillopt/model/backend_config.py150def set_target_backend(backend: str) -> None:CODE
LOW⚡skillopt/model/router.py181def set_reasoning_effort(effort: str | None) -> None:CODE
LOW⚡skillopt/model/router.py186def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/router.py191def set_optimizer_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/codex_backend.py663def set_target_deployment(deployment: str) -> None:CODE
LOW⚡skillopt/model/codex_backend.py669def set_reasoning_effort(effort: str | None) -> None:CODE
LOW⚡skillopt/model/codex_backend.py674def set_optimizer_deployment(deployment: str) -> None:CODE
LOWskillopt/envs/alfworld/__init__.py5__all__ = ["ALFWorldAdapter"]CODE
LOWskillopt/envs/spreadsheetbench/__init__.py5__all__ = ["SpreadsheetBenchAdapter"]CODE
LOWskillopt/engine/__init__.py9__all__ = ["ReflACTTrainer"]CODE
AI Slop Vocabulary15 hits · 44 pts
SeverityFileLineSnippetContext
LOWtests/test_sleep_engine.py964 # presence of a heading an optimizer can simply add.STRING
MEDIUMtests/test_sleep_multi_skill.py315 # Guards the guard: if the harness above ever stops watching the placeCOMMENT
LOWtests/test_aggregate_fallback.py76 # We patch _hierarchical_merge to just return the single patches to skip the batch merges,COMMENT
MEDIUMscripts/smoke_superpowers.sh3# Run this manually (not in CI) to verify the adapter works with the real harness.COMMENT
MEDIUMskillopt_sleep/cycle.py899 # experiment harness reports its per-night test score. With the defaultCOMMENT
MEDIUMskillopt_sleep/backend.py502 # user for the range" rule scores 0 because the harness can't run VBA andCOMMENT
MEDIUMskillopt_sleep/consolidate.py84 # Be robust if a split is empty: fall back so a night still does something,COMMENT
MEDIUMskillopt_sleep/rollout.py141 # single-shot reflect uses — prevents harness-violating rules like "return VBA"COMMENT
MEDIUMskillopt_sleep/adapters/superpowers.py85# pytest executed by the harness. Both remain reachable by the same OS user, soCOMMENT
MEDIUM⚡skillopt_sleep/adapters/superpowers.py411 # harness-collected: shim log mtime vs newest project source mtimeCOMMENT
MEDIUM⚡skillopt_sleep/adapters/superpowers.py414 # harness-collected: counted by the nonce-tagged pytest shimCOMMENT
MEDIUM⚡skillopt_sleep/adapters/superpowers.py421 # harness re-runs the tests itself after the agent exitsCOMMENT
MEDIUMskillopt_sleep/adapters/superpowers.py678 # Isolated project, HOME and (harness-only) audit dir per scenarioCOMMENT
MEDIUMskillopt_sleep/adapters/superpowers.py684 # config; harness-created but reachable by the agent (trusted-candidate scope)COMMENT
LOW⚡skillopt/envs/_template/env_template.py78 # Dataset-backed envs typically just pass items straight through.COMMENT
Cross-Language Confusion5 hits · 32 pts
SeverityFileLineSnippetContext
HIGHtests/test_aggregate_fallback.py123 ['{"edits": null}', '{"edits": ["not an object"]}'],CODE
HIGH⚡skillopt_sleep/backend.py1053 ' if (args === null || typeof args !== "object" || 'CODE
HIGH⚡skillopt_sleep/backend.py1054 "Object.keys(args).length !== 1 || args.query !== query) {\n"CODE
HIGHskillopt_sleep/llm_miner.py51 # A non-string rubric (e.g. a JSON null) must not become the literal "None"COMMENT
HIGHskillopt_sleep/adapters/superpowers.py467 f'count=$(grep -c "^{nonce} run " "$audit_log" 2>/dev/null || true); 'CODE
Cross-File Repetition4 hits · 20 pts
SeverityFileLineSnippetContext
HIGHskillopt/envs/alfworld/reflect.py0searchqa reflect stage. prompts are now loaded from .md files by the base adapter.STRING
HIGHskillopt/envs/livemathematicianbench/reflect.py0searchqa reflect stage. prompts are now loaded from .md files by the base adapter.STRING
HIGHskillopt/envs/spreadsheetbench/reflect.py0searchqa reflect stage. prompts are now loaded from .md files by the base adapter.STRING
HIGHskillopt/envs/searchqa/reflect.py0searchqa reflect stage. prompts are now loaded from .md files by the base adapter.STRING
Over-Commented Block15 hits · 15 pts
SeverityFileLineSnippetContext
LOWrequirements.txt21# backends (Claude/Qwen). Without it extract_json() falls back safely andCOMMENT
LOWplugins/devin/hooks/on-session-end.sh1#!/usr/bin/env bashCOMMENT
LOWplugins/claude-code/scripts/install-cron.sh21# (unless you pass --auto-adopt below).COMMENT
LOWplugins/dsh/cordis.patch.yml1# dsh-skillopt bundle patch layer.COMMENT
LOWplugins/dsh/src/index.js61 .default(false)COMMENT
LOWplugins/dsh/src/index.js161// editBudget, …) that buildArgv would forward to the engine — crossing theCOMMENT
LOWconfigs/features/soft_gate.yaml1# ─────────────────────────────────────────────────────────────────────────────COMMENT
LOWconfigs/features/soft_gate.yaml21# `soft` or `mixed` lets these partial improvements be accepted.COMMENT
LOWscripts/smoke_superpowers.sh1#!/bin/bashCOMMENT
LOWscripts/run_alfworld.sh1#!/usr/bin/env bashCOMMENT
LOWskillopt_sleep/consolidate.py41 rejected_edits: List[EditRecord]COMMENT
LOWskillopt_sleep/consolidate.py421 # edit was dropped during the per-target trials, `all_applied` is emptyCOMMENT
LOWskillopt_sleep/types.py61 # carrying the research repo's exact rollout_system), the backend uses THISCOMMENT
LOWskillopt/utils/json_utils.py201 # JSON strict json.loads rejects — unescaped ASCII quotes inside CJK stringCOMMENT
LOWskillopt/envs/_template/config_template.yaml1# ──────────────────────────────────────────────────COMMENT
Example Usage Blocks4 hits · 8 pts
SeverityFileLineSnippetContext
LOWscripts/smoke_superpowers.sh10# Usage:COMMENT
LOW⚡scripts/run_spreadsheetbench.sh5# Usage:COMMENT
LOW⚡scripts/run_searchqa.sh5# Usage:COMMENT
LOW⚡scripts/run_alfworld.sh9# Usage:COMMENT
Modern AI Meta-Vocabulary2 hits · 6 pts
SeverityFileLineSnippetContext
MEDIUMtests/test_sleep_engine.py2952 # arguments observed at the public orchestration boundary.COMMENT
MEDIUMskillopt_sleep/harvest_pi.py239 # agentic coding and are frequently followed by recovery and aCOMMENT
Structural Annotation Overuse3 hits · 4 pts
SeverityFileLineSnippetContext
LOWdocs/guide/new-backend.md130## Step 1: implement the module contractCOMMENT
LOWdocs/guide/new-backend.md163## Step 2: register and route the backendCOMMENT
LOWdocs/guide/new-backend.md189## Step 3: test the integrationCOMMENT
Slop Phrases1 hit · 3 pts
SeverityFileLineSnippetContext
MEDIUM⚡tests/test_outcome_judges.py42 "I'm unable to do this.",CODE
Fake / Example Data1 hit · 0 pts
SeverityFileLineSnippetContext
LOWtests/test_sleep_engine.py2325 self.assertIn("placeholder", res.rejected_edits[0].content)CODE