Repository Analysis

NousResearch/hermes-agent

The agent that grows with you

34.4 Strong AI signal View on GitHub

Analysis Overview

This report presents the forensic synthetic code analysis of NousResearch/hermes-agent, a Python project with 237,643 GitHub stars. SynthScan v2.0 examined 3,043,853 lines of code across 9318 source files, recording 73153 pattern matches distributed across 28 syntactic categories. The overall adjusted score of 34.4 places this repository in the Strong 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).

34.4
Adjusted Score
34.4
Raw Score
100%
Time Factor
2026-08-28
Last Push
237.6K
Stars
Python
Language
3.0M
Lines of Code
9.3K
Files
73.2K
Pattern Hits
2026-08-29
Scan Date
0.07
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 8HIGH 629MEDIUM 14768LOW 57748

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 73153 distinct pattern matches across 28 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.

Decorative Section Separators13847 hits · 48926 pts
SeverityFileLineSnippetContext
MEDIUMhermes_constants.py21# ── TUI busy-indicator styles ─────────────────────────────────────────COMMENT
MEDIUMhermes_constants.py1646# ─── Well-Known Paths ─────────────────────────────────────────────────────────COMMENT
MEDIUMhermes_constants.py1669# ─── Network Preferences ─────────────────────────────────────────────────────COMMENT
MEDIUMhermes_constants.py1714# ─── Streaming Response Constants ────────────────────────────────────────────COMMENT
MEDIUMhermes_constants.py1728# ─── Venv layout ─────────────────────────────────────────────────────────────COMMENT
MEDIUMhermes_constants.py1781# ─── Partial-update diagnostics ──────────────────────────────────────────────COMMENT
MEDIUMsetup-hermes.sh2# ============================================================================COMMENT
MEDIUMsetup-hermes.sh4# ============================================================================COMMENT
MEDIUMsetup-hermes.sh18# ============================================================================COMMENT
MEDIUMsetup-hermes.sh269# ============================================================================COMMENT
MEDIUMsetup-hermes.sh270# ============================================================================COMMENT
MEDIUMsetup-hermes.sh272# ============================================================================COMMENT
MEDIUMsetup-hermes.sh62# ============================================================================COMMENT
MEDIUMsetup-hermes.sh64# ============================================================================COMMENT
MEDIUMsetup-hermes.sh129# ============================================================================COMMENT
MEDIUMsetup-hermes.sh131# ============================================================================COMMENT
MEDIUMsetup-hermes.sh165# ============================================================================COMMENT
MEDIUMsetup-hermes.sh167# ============================================================================COMMENT
MEDIUMsetup-hermes.sh187# ============================================================================COMMENT
MEDIUMsetup-hermes.sh189# ============================================================================COMMENT
MEDIUMsetup-hermes.sh325# ============================================================================COMMENT
MEDIUMsetup-hermes.sh327# ============================================================================COMMENT
MEDIUMsetup-hermes.sh344# ============================================================================COMMENT
MEDIUMsetup-hermes.sh346# ============================================================================COMMENT
MEDIUMsetup-hermes.sh398# ============================================================================COMMENT
MEDIUMsetup-hermes.sh400# ============================================================================COMMENT
MEDIUMsetup-hermes.sh417# ============================================================================COMMENT
MEDIUMsetup-hermes.sh419# ============================================================================COMMENT
MEDIUMrun_agent.py298# =========================================================================COMMENT
MEDIUMrun_agent.py300# =========================================================================COMMENT
MEDIUMrun_agent.py303# =========================================================================COMMENT
MEDIUMrun_agent.py307# =========================================================================COMMENT
MEDIUMrun_agent.py1861 # ------------------------------------------------------------------COMMENT
MEDIUMrun_agent.py1863 # ------------------------------------------------------------------COMMENT
MEDIUMrun_agent.py1144 # ── Buffered retry/fallback status ────────────────────────────────────COMMENT
MEDIUMrun_agent.py6575 # ── Unified streaming API call ─────────────────────────────────────────COMMENT
MEDIUMrun_agent.py7084 # ── Per-turn primary restoration ─────────────────────────────────────COMMENT
MEDIUMtoolsets.py409 # ==========================================================================COMMENT
MEDIUMtoolsets.py416 # ==========================================================================COMMENT
MEDIUMpyproject.toml526# ---------------------------------------------------------------------------COMMENT
MEDIUMpyproject.toml530# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py161# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py163# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py179# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py181# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py215# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py217# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py255# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py257# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py411# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py413# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py550# ---------------------------------------------------------------------------COMMENT
MEDIUMhermes_logging.py560# ---------------------------------------------------------------------------COMMENT
MEDIUMmini_swe_runner.py64# ============================================================================COMMENT
MEDIUMmini_swe_runner.py66# ============================================================================COMMENT
MEDIUMmini_swe_runner.py626# ============================================================================COMMENT
MEDIUMmini_swe_runner.py628# ============================================================================COMMENT
MEDIUMmini_swe_runner.py113# ============================================================================STRING
MEDIUMmini_swe_runner.py115# ============================================================================STRING
MEDIUMmini_swe_runner.py153# ============================================================================STRING
13787 more matches not shown…
Hyper-Verbose Identifiers37768 hits · 34421 pts
SeverityFileLineSnippetContext
LOWbatch_runner.py101def _normalize_tool_error_counts(tool_error_counts: Dict[str, int]) -> Dict[str, int]:CODE
LOWbatch_runner.py774 def _scan_completed_prompts_by_content(self) -> set:CODE
LOWbatch_runner.py816 def _filter_dataset_by_completed(self, completed_prompts: set) -> Tuple[List[Dict], List[int]]:CODE
LOWhermes_constants.py305def mark_named_profile_deleted(profile_home: str | Path) -> None:CODE
LOWhermes_constants.py311def clear_named_profile_deleted(profile_home: str | Path) -> None:CODE
LOWhermes_constants.py315def assert_named_profile_home_live(path: str | Path) -> None:CODE
LOWhermes_constants.py40def reset_hermes_home_override(token: Token) -> None:CODE
LOWhermes_constants.py53def _get_platform_default_hermes_home() -> Path:CODE
LOWhermes_constants.py77def _warn_profile_fallback_once() -> None:CODE
LOWhermes_constants.py442def _candidate_node_command_names(command: str) -> list[str]:CODE
LOWhermes_constants.py506def hermes_managed_node_tree_present(home: Path | None = None) -> bool:CODE
LOWhermes_constants.py594def _print_managed_node_in_use_notice() -> None:CODE
LOWhermes_constants.py607def _heal_managed_node_windows(home: Path | None = None) -> bool | None:CODE
LOWhermes_constants.py757def _bootstrap_managed_node_posix() -> bool:CODE
LOWhermes_constants.py795def bootstrap_hermes_managed_node() -> str | None:CODE
LOWhermes_constants.py880def _managed_node_tree_outdated(home: Path | None = None) -> bool:CODE
LOWhermes_constants.py915def find_hermes_node_executable(command: str) -> str | None:CODE
LOWhermes_constants.py951def find_node_executable_on_path(command: str) -> str | None:CODE
LOWhermes_constants.py1201def _iter_real_home_candidates(env: dict[str, str] | None = None) -> list[str]:CODE
LOWhermes_constants.py1288def apply_subprocess_home_env(env: dict[str, str]) -> None:CODE
LOWhermes_constants.py1330def _canonical_model_variants(model: str) -> list[str]:CODE
LOWhermes_constants.py1420def resolve_per_model_reasoning_effort(model: str, overrides: dict | None) -> dict | None:CODE
LOWhermes_constants.py1572def translate_cwd_for_wsl_backend(cwd: str) -> str:CODE
LOWtoolset_distributions.py241def sample_toolsets_from_distribution(distribution_name: str) -> List[str]:CODE
LOWhermes_bootstrap.py125def suppress_platform_ver_console() -> None:CODE
LOWhermes_bootstrap.py204def activate_durable_lazy_target() -> None:CODE
LOWrun_agent.py1030 def _warn_context_overflow_blocked(CODE
LOWrun_agent.py1195 def _emit_pending_fallback_notice(self) -> None:CODE
LOWrun_agent.py1255 def _disable_codex_reasoning_replay(CODE
LOWrun_agent.py1871 def _summarize_background_review_actions(CODE
LOWrun_agent.py3035 def _api_request_payload_for_hook(self, api_kwargs: Optional[Dict[str, Any]]) -> Dict[str, Any]:CODE
LOWrun_agent.py3884 def _format_turn_completion_explanation(CODE
LOWrun_agent.py5560 def _request_anthropic_client_cache_ref(self) -> dict:CODE
LOWrun_agent.py5568 def _request_anthropic_client_key(self) -> tuple:CODE
LOWrun_agent.py6729 def _interim_text_was_delivered(self, text: str) -> bool:CODE
LOWrun_agent.py6735 def _record_delivered_interim_text(self, text: str) -> None:CODE
LOWrun_agent.py6744 def _fire_streamed_codex_commentary(self, text: str) -> None:CODE
LOWrun_agent.py7060 def _interruptible_streaming_api_call(CODE
LOWrun_agent.py7874 def _read_reasoning_echo_from_config() -> bool:CODE
LOWrun_agent.py7884 def _needs_kimi_tool_reasoning(self) -> bool:CODE
LOWrun_agent.py7932 def _copy_reasoning_content_for_api(self, source_msg: dict, api_msg: dict) -> None:CODE
LOWrun_agent.py7937 def _reapply_reasoning_echo_for_provider(self, api_messages: list) -> int:CODE
LOWrun_agent.py7943 def _sanitize_tool_calls_for_strict_api(api_msg: dict, model: "str | None" = None) -> dict:CODE
LOWrun_agent.py8353 def _toolguard_controlled_halt_response(self, decision: ToolGuardrailDecision) -> str:CODE
LOWrun_agent.py8362 def _append_guardrail_observation(CODE
LOWrun_agent.py93def _session_source_for_agent(platform: Optional[str]) -> str:CODE
LOWrun_agent.py258def _is_ephemeral_scaffolding(msg: Any) -> bool:CODE
LOWrun_agent.py320def _pool_may_recover_from_rate_limit(pool) -> bool:CODE
LOWrun_agent.py358def _safe_session_filename_component(session_id: str) -> str:CODE
LOWrun_agent.py618 def _get_session_db_for_recall(self):CODE
LOWrun_agent.py693 def _transition_context_engine_session(CODE
LOWrun_agent.py842 def _effective_lmstudio_context_length(CODE
LOWrun_agent.py872 def _lmstudio_load_was_unverified(load_result: Any) -> bool:CODE
LOWrun_agent.py879 def _ensure_lmstudio_runtime_loaded(CODE
LOWrun_agent.py952 def _should_start_quiet_spinner(self) -> bool:CODE
LOWrun_agent.py971 def _should_emit_quiet_tool_messages(self) -> bool:CODE
LOWrun_agent.py1070 def _warn_uncompressed_context_overflow(CODE
LOWrun_agent.py1090 def _clear_context_overflow_warn(self) -> None:CODE
LOWrun_agent.py1298 def _stream_diag_capture_response(CODE
LOWrun_agent.py1311 def _is_provider_stream_parse_error(self, error: BaseException) -> bool:CODE
37708 more matches not shown…
Excessive Try-Catch Wrapping9598 hits · 7299 pts
SeverityFileLineSnippetContext
LOWbatch_runner.py299 except Exception as img_err:CODE
LOWbatch_runner.py381 except Exception as e:CODE
LOWbatch_runner.py748 except Exception as e:CODE
LOWbatch_runner.py811 except Exception as e:CODE
LOWbatch_runner.py1018 except Exception as ckpt_err:CODE
LOWbatch_runner.py1026 except Exception as e:CODE
LOWbatch_runner.py1059 except Exception as ckpt_err:CODE
LOWbatch_runner.py1340 except Exception as e:CODE
LOWbatch_runner.py1371 except Exception as e:CODE
LOWhermes_constants.py110 except Exception:CODE
LOWhermes_constants.py558 except Exception:CODE
LOWhermes_constants.py570 except Exception:CODE
LOWhermes_constants.py575 except Exception:CODE
LOWhermes_constants.py1182 except Exception:CODE
LOWhermes_constants.py1217 except Exception:CODE
LOWhermes_constants.py1542 except Exception:CODE
LOWhermes_bootstrap.py163 except Exception:CODE
LOWhermes_bootstrap.py223 except Exception:CODE
MEDIUMrun_agent.py6914def _emit_stream_start(self) -> None:CODE
LOWrun_agent.py1022 except Exception:CODE
LOWrun_agent.py1027 except Exception:CODE
LOWrun_agent.py1183 except Exception:CODE
LOWrun_agent.py1192 except Exception:CODE
LOWrun_agent.py1250 except Exception:CODE
LOWrun_agent.py1252 except Exception:CODE
LOWrun_agent.py3008 except Exception:CODE
LOWrun_agent.py3018 except Exception:CODE
LOWrun_agent.py3025 except Exception:CODE
LOWrun_agent.py3870 except Exception:CODE
LOWrun_agent.py3879 except Exception:CODE
LOWrun_agent.py4438 except Exception as e:CODE
LOWrun_agent.py4442 except Exception:CODE
LOWrun_agent.py4451 except Exception:CODE
LOWrun_agent.py4575 except Exception:CODE
LOWrun_agent.py4579 except Exception:CODE
LOWrun_agent.py4581 except Exception:CODE
LOWrun_agent.py4595 except Exception:CODE
LOWrun_agent.py4602 except Exception:CODE
LOWrun_agent.py4606 except Exception:CODE
LOWrun_agent.py4631 except Exception:CODE
LOWrun_agent.py4640 except Exception:CODE
LOWrun_agent.py4646 except Exception:CODE
LOWrun_agent.py4652 except Exception:CODE
LOWrun_agent.py4675 except Exception:CODE
LOWrun_agent.py4677 except Exception:CODE
LOWrun_agent.py4686 except Exception:CODE
LOWrun_agent.py4693 except Exception:CODE
LOWrun_agent.py4697 except Exception:CODE
LOWrun_agent.py5552 except Exception as exc:CODE
LOWrun_agent.py7043 except Exception:CODE
LOWrun_agent.py7053 except Exception:CODE
LOWrun_agent.py7881 except Exception:CODE
LOWrun_agent.py98 except Exception:CODE
LOWrun_agent.py642 except Exception:CODE
LOWrun_agent.py659 except Exception:CODE
LOWrun_agent.py672 except Exception:CODE
LOWrun_agent.py686 except Exception as e:CODE
LOWrun_agent.py719 except Exception as exc:CODE
LOWrun_agent.py725 except Exception as exc:CODE
LOWrun_agent.py748 except Exception as exc:CODE
9538 more matches not shown…
Unused Imports4056 hits · 3520 pts
SeverityFileLineSnippetContext
LOWbatch_runner.py26CODE
LOWhermes_bootstrap.py50CODE
LOWrun_agent.py112CODE
LOWrun_agent.py112CODE
LOWrun_agent.py112CODE
LOWrun_agent.py117CODE
LOWrun_agent.py138CODE
LOWrun_agent.py138CODE
LOWrun_agent.py138CODE
LOWrun_agent.py156CODE
LOWrun_agent.py167CODE
LOWrun_agent.py168CODE
LOWrun_agent.py168CODE
LOWrun_agent.py168CODE
LOWrun_agent.py168CODE
LOWrun_agent.py168CODE
LOWrun_agent.py175CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py176CODE
LOWrun_agent.py211CODE
LOWrun_agent.py211CODE
LOWrun_agent.py211CODE
LOWrun_agent.py211CODE
LOWrun_agent.py211CODE
LOWrun_agent.py211CODE
LOWrun_agent.py26CODE
LOWrun_agent.py1290CODE
LOWrun_agent.py1864CODE
LOWrun_agent.py1864CODE
LOWrun_agent.py1864CODE
LOWregistration_lifecycle.py8CODE
LOWhermes_state.py49CODE
LOWhermes_state.py49CODE
LOWhermes_state.py49CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
LOWhermes_state.py60CODE
3996 more matches not shown…
Over-Commented Block2559 hits · 1690 pts
SeverityFileLineSnippetContext
LOWdocker-compose.windows.yml1#COMMENT
LOWsetup-hermes.sh1#!/bin/bashCOMMENT
LOWsetup-hermes.sh241 # (the direct deps in pyproject.toml are exact-pinned, butCOMMENT
LOW.hadolint.yaml1# hadolint configuration for the Hermes Agent Dockerfile.COMMENT
LOW.hadolint.yaml21 - DL3003COMMENT
LOWhermes_bootstrap.py81 os.environ.setdefault("PYTHONIOENCODING", "utf-8")COMMENT
LOWrun_agent.py41import osCOMMENT
LOWrun_agent.py261 return isinstance(msg, dict) and any(COMMENT
LOWrun_agent.py281# CONTRACT (#92231): the marker asserts "this dict's CONTENT is durable asCOMMENT
LOWrun_agent.py1901 skill". The automatic post-turn triggers never set it.COMMENT
LOWrun_agent.py2161 if not self._session_db:COMMENT
LOWrun_agent.py2181 # assumes the live `messages` list is the original history plus aCOMMENT
LOWrun_agent.py2241 if not isinstance(msg, dict):COMMENT
LOWrun_agent.py2261 continueCOMMENT
LOWrun_agent.py2421 )COMMENT
LOWrun_agent.py2441 from hermes_state import (COMMENT
LOWrun_agent.py4701 # it on turn crash / retirement (agent/codex_runtime.py), but hardCOMMENT
LOWrun_agent.py4741 # 'cron_complete' / 'cli_close' reason set by an earlier terminal path.COMMENT
LOWrun_agent.py5401 # Per-request OpenAI-wire clients (used by both the non-streamingCOMMENT
LOWrun_agent.py6941 self._note_dropped_stream_writer("_fire_stream_delta")COMMENT
LOWrun_agent.py8041 # conversation tag and no Portal sticky key, so it routes independentlyCOMMENT
LOWrun_agent.py8101 # allowed to mutate their input list in place; after a hostCOMMENT
LOWrun_agent.py8221 on_commit_overrun=_on_commit_overrun,COMMENT
LOWrun_agent.py8941 # LLM call made inside this turn — main loop, compression, vision,COMMENT
LOWtoolsets.py61 # deliberately NOT here. They only make sense where a GUI can follow theCOMMENT
LOWtoolsets.py401 ],COMMENT
LOWpyproject.toml21 # Rationale: ranges allow PyPI to ship a fresh version of a transitiveCOMMENT
LOWpyproject.toml61 # official Snowball project package, pure Python, zero transitive deps.COMMENT
LOWpyproject.toml81 # (~108KB, no compiled extensions, no platform constraints), so unlike theCOMMENT
LOWpyproject.toml101 "cryptography==50.0.0", # CVE-2026-69247, GHSA-m2h6-j472-rp4c, GHSA-jwv3-5hgf-82ww, CVE-2026-39892, CVE-2026-34073, GCOMMENT
LOWpyproject.toml121 "uvicorn[standard]>=0.31.0,<1",COMMENT
LOWpyproject.toml141 # Windows log rotation. Stdlib ``RotatingFileHandler.doRollover()`` usesCOMMENT
LOWpyproject.toml161 # instead of falling back to the no-op Relay host (#76469, Termux).COMMENT
LOWpyproject.toml221 "numpy==2.4.3",COMMENT
LOWpyproject.toml241vision = []COMMENT
LOWpyproject.toml261# module names.COMMENT
LOWpyproject.toml321 "google-auth-oauthlib==1.3.1",COMMENT
LOWpyproject.toml341all = [COMMENT
LOWpyproject.toml381hermes-acp = "acp_adapter.entry:main"COMMENT
LOWpyproject.toml401# aiohttp, cryptography: same shape — the advisory fixes are newer than theCOMMENT
LOWpyproject.toml501# ASYNC210/220/221/251: blocking calls inside `async def` freeze the wholeCOMMENT
LOWpyproject.toml521"skills/**" = ["PLW1514"]COMMENT
LOWhermes_logging.py41COMMENT
LOWhermes_logging.py541COMMENT
LOWutils.py101# 33 ERROR_LOCK_VIOLATION — byte-range lock on the targetCOMMENT
LOWhermes_state.py361COMMENT
LOWhermes_state.py421# FTS write corruption"). The hermetic conftest redirects HERMES_HOME perCOMMENT
LOWhermes_state.py641 f"child process, export {_STATE_DB_GUARD_BYPASS_ENV}=1 in "COMMENT
LOWhermes_state.py661#COMMENT
LOWhermes_state.py681# which lets state.db-wal keep the high-water mark of the largest-everCOMMENT
LOWhermes_state.py1761# ---------------------------------------------------------------------------COMMENT
LOWhermes_state.py1961 return TrueCOMMENT
LOWhermes_state.py2081 try:COMMENT
LOWhermes_state.py2101# successful repair or replacement changes it and resets the count);COMMENT
LOWhermes_state.py2121COMMENT
LOWhermes_state.py2141 for start, end in _FINGERPRINT_VOLATILE_HEADER_RANGES:COMMENT
LOWhermes_state.py2581 # sorts NEWEST and survives prune forever.COMMENT
LOWhermes_state.py2601 #COMMENT
LOWhermes_state.py2661 )COMMENT
LOWhermes_state.py3041 # index corruption — bad shadow-table segments where reads stillCOMMENT
2499 more matches not shown…
Cross-Language Confusion285 hits · 1570 pts
SeverityFileLineSnippetContext
HIGHhermes_state.py5835 undefined, so every attribute access remains guarded.STRING
HIGHtools/schema_sanitizer.py29 {"$ref": "#/$defs/Foo", "default": null}STRING
HIGHtools/schema_sanitizer.py249 {"anyOf": [{"type": "string"}, {"type": "null"}], "default": null}STRING
HIGHtools/voice_mode.py126 return "pkg install python-numpy portaudio && python -m pip install sounddevice"CODE
HIGHtools/voice_mode.py350 " 3. Verify with: arecord -d 3 /tmp/test.wav && aplay /tmp/test.wav"CODE
HIGHtools/voice_mode.py357 " Then verify: arecord -d 3 /tmp/test.wav && aplay /tmp/test.wav"CODE
HIGHtools/process_registry.py1538 f"kill -0 \"$(cat {quoted_pid_path} 2>/dev/null)\" 2>/dev/null; echo $?",CODE
HIGHtools/tts_tool.py1958 # tts.deepinfra block in DEFAULT_CONFIG to deep-merge over the null).COMMENT
HIGHtools/transcription_tools.py2734 # deep-merge over the null).COMMENT
HIGHtools/transcription_tools.py3111 di_config = stt_config.get("deepinfra") # may be None (YAML null)CODE
HIGHtools/code_kernel_remote.py174 f"kill -0 {shlex.quote(kernel.pid)} 2>/dev/null && echo ALIVE",STRING
HIGHtools/code_kernel_remote.py188 f"pkill -TERM -P {shlex.quote(kernel.pid)} 2>/dev/null; "STRING
HIGHtools/code_kernel_remote.py189 f"kill {shlex.quote(kernel.pid)} 2>/dev/null; true",STRING
HIGHtools/browser_tool.py946 return "npm install -g agent-browser && agent-browser install"CODE
HIGHtools/browser_tool.py947 return "npm install -g agent-browser && agent-browser install --with-deps"CODE
HIGHtools/terminal_tool.py861 """Wrap `A && B &` (or `A || B &`) to `A && { B & }` at depth 0.STRING
HIGHtools/terminal_tool.py861 """Wrap `A && B &` (or `A || B &`) to `A && { B & }` at depth 0.STRING
HIGHtools/terminal_tool.py863 Bash parses ``A && B &`` with `&&` tighter than `&`, so it forks aSTRING
HIGHtools/terminal_tool.py864 subshell for the whole `A && B` compound and backgrounds it. InsideSTRING
HIGHtools/browser_supervisor.py118 xhr.open("GET", ENDPOINT + "?" + params.toString(), false); // syncCODE
HIGHtools/browser_supervisor.py119 xhr.send(null);CODE
HIGHtools/browser_supervisor.py120 if (xhr.status !== 200) return null;CODE
HIGHtools/browser_supervisor.py123 try { parsed = JSON.parse(body); } catch (e) { return null; }CODE
HIGHtools/browser_supervisor.py124 if (kind === "alert") return undefined;CODE
HIGHtools/browser_supervisor.py125 if (kind === "confirm") return Boolean(parsed && parsed.accept);CODE
HIGHtools/browser_supervisor.py127 if (!parsed || !parsed.accept) return null;CODE
HIGHtools/browser_supervisor.py130 return null;CODE
HIGHtools/browser_supervisor.py134 return null;CODE
HIGHtools/browser_camofox.py14 git clone https://github.com/jo-inc/camofox-browser && cd camofox-browserSTRING
HIGHtools/browser_camofox.py15 npm install && npm start # downloads Camoufox (~300MB) on first runSTRING
HIGHtools/file_operations.py1298 'm="$(stat -c%a "$t" 2>/dev/null || stat -f%Lp "$t" 2>/dev/null || true)"; 'CODE
HIGHtools/file_operations.py1299 '[ -n "$m" ] && chmod "$m" "$tmp" 2>/dev/null || true; 'CODE
HIGHtools/file_operations.py1305 'if [ ! -e "$t" ]; then chmod "=rw" "$tmp" 2>/dev/null || true; fi; 'CODE
HIGHtools/file_operations.py1011 result = self._exec(f"command -v {cmd} >/dev/null 2>&1 && echo 'yes'")CODE
HIGHtools/file_operations.py1283 'rt="$(readlink -f "$t" 2>/dev/null || realpath "$t" 2>/dev/null || true)"; 'CODE
HIGHtools/file_operations.py1386 f"if [ -f {arg} ]; then wc -c < {arg} 2>/dev/null; "CODE
HIGHtools/file_operations.py2835 check = self._exec(f"test -e {self._escape_shell_arg(path)} && echo exists || echo not_found")CODE
HIGHtools/file_operations.py2852 f"test -d {self._escape_shell_arg(parent)} && echo yes || echo no"CODE
HIGHtools/file_operations.py2930 f"test -e {self._escape_shell_arg(expanded)} && echo exists || echo not_found"CODE
HIGHtools/code_execution_tool.py1030 f"ls -1 {quoted_rpc_dir}/req_* 2>/dev/null || true",STRING
HIGHtools/code_execution_tool.py1253 "command -v python3 >/dev/null 2>&1 && echo OK",STRING
HIGHtools/environments/local.py1834 prelude_parts.append(f"[ -r '{safe}' ] && . '{safe}' 2>/dev/null || true")CODE
HIGHtools/environments/docker.py1657 cmd_string = f"unset {quoted_names} 2>/dev/null || true\n{cmd_string}"CODE
HIGHtools/environments/modal.py113 "RUN rm -rf /usr/local/lib/python*/site-packages/pip* 2>/dev/null; "CODE
HIGHtools/environments/modal.py114 "python -m ensurepip --upgrade --default-pip 2>/dev/null || true",CODE
HIGHtools/environments/modal.py118 "RUN apt-get update -qq && apt-get install -y -qq python3 python3-venv > /dev/null 2>&1 || true"CODE
HIGHtools/environments/modal.py118 "RUN apt-get update -qq && apt-get install -y -qq python3 python3-venv > /dev/null 2>&1 || true"CODE
HIGHtools/environments/base.py638 f"HERMES_UI_SESSION_ID{extra_unset} 2>/dev/null; "CODE
HIGHtools/environments/base.py829 f">> {_snap_tmp} 2>/dev/null || true\n"CODE
HIGHtools/environments/base.py837 f"builtin cd -- {_quoted_cwd} 2>/dev/null || true\n"CODE
HIGHtools/environments/base.py955 f"source {_quoted_snap} >/dev/null 2>&1 || true"CODE
HIGHtools/environments/base.py1018 f"2>/dev/null || rm -f {_snap_tmp} 2>/dev/null || true"CODE
HIGHhermes_cli/cli_billing_mixin.py740 upgradeResult(null) — the CLI can't persist the key across a command re-run,STRING
HIGHhermes_cli/gateway.py5393 f"launchctl bootout {shlex.quote(target)} 2>/dev/null; "CODE
HIGHhermes_cli/gateway.py5400 f"while kill -0 {gateway_pid} 2>/dev/null; do "CODE
HIGHhermes_cli/gateway.py5411 f" launchctl bootstrap {shlex.quote(domain)} {shlex.quote(str(plist_path))} 2>/dev/null; "CODE
HIGHhermes_cli/skills_config.py30 Mirrors ``agent.skill_utils._normalize_string_set``: ``None`` (YAML null)STRING
HIGHhermes_cli/models.py985# "paidRecommendedCompactionModel": {modelName, ...} | null,COMMENT
HIGHhermes_cli/models.py986# "paidRecommendedVisionModel": {modelName, ...} | null,COMMENT
HIGHhermes_cli/models.py987# "freeRecommendedCompactionModel": {modelName, ...} | null,COMMENT
225 more matches not shown…
Deep Nesting2149 hits · 1364 pts
SeverityFileLineSnippetContext
LOWbatch_runner.py125CODE
LOWbatch_runner.py244CODE
LOWbatch_runner.py400CODE
LOWbatch_runner.py544CODE
LOWbatch_runner.py684CODE
LOWbatch_runner.py774CODE
LOWbatch_runner.py816CODE
LOWbatch_runner.py850CODE
LOWhermes_constants.py795CODE
LOWhermes_constants.py915CODE
LOWhermes_constants.py926CODE
LOWrun_agent.py9099CODE
LOWrun_agent.py1195CODE
LOWrun_agent.py1225CODE
LOWrun_agent.py2130CODE
LOWrun_agent.py2684CODE
LOWrun_agent.py2713CODE
LOWrun_agent.py2901CODE
LOWrun_agent.py3153CODE
LOWrun_agent.py3182CODE
LOWrun_agent.py3552CODE
LOWrun_agent.py3680CODE
LOWrun_agent.py4545CODE
LOWrun_agent.py4930CODE
LOWrun_agent.py6328CODE
LOWrun_agent.py6577CODE
LOWrun_agent.py7437CODE
LOWrun_agent.py7564CODE
LOWrun_agent.py7597CODE
LOWrun_agent.py7783CODE
LOWrun_agent.py8006CODE
LOWrun_agent.py8592CODE
LOWrun_agent.py8245CODE
LOWrun_agent.py8862CODE
LOWtoolsets.py742CODE
LOWregistration_lifecycle.py88CODE
LOWmini_swe_runner.py630CODE
LOWmini_swe_runner.py298CODE
LOWmini_swe_runner.py408CODE
LOWhermes_time.py37CODE
LOWhermes_state_portability.py309CODE
LOWhermes_state_portability.py507CODE
LOWutils.py194CODE
LOWtrajectory_compressor.py1380CODE
LOWtrajectory_compressor.py477CODE
LOWtrajectory_compressor.py605CODE
LOWtrajectory_compressor.py674CODE
LOWtrajectory_compressor.py1076CODE
LOWhermes_state.py537CODE
LOWhermes_state.py1106CODE
LOWhermes_state.py1988CODE
LOWhermes_state.py2261CODE
LOWhermes_state.py2333CODE
LOWhermes_state.py2530CODE
LOWhermes_state.py3010CODE
LOWhermes_state.py3190CODE
LOWhermes_state.py3383CODE
LOWhermes_state.py3550CODE
LOWhermes_state.py3576CODE
LOWhermes_state.py3967CODE
2089 more matches not shown…
Cross-File Repetition248 hits · 1240 pts
SeverityFileLineSnippetContext
HIGHhermes_cli/subcommands/update.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/prompt_size.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/auth.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/logout.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/backup.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/hooks.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/config.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/whatsapp.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/webhook.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/profile.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/security.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/model.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/dump.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/setup.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/logs.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/debug.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/uninstall.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/login.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/import_cmd.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/doctor.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/gui.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/slack.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/status.py0``hermes status`` subcommand parser. extracted verbatim from ``hermes_cli/main.py:main()`` (god-file phase 2). handler iSTRING
HIGHhermes_cli/subcommands/plugins.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/skills.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/memory.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/tools.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/acp.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/mcp.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/insights.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/pairing.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHhermes_cli/subcommands/claw.py0``hermes claw`` subcommand parser. extracted from ``hermes_cli/main.py:main()`` (god-file phase 2 follow-up). handler inSTRING
HIGHplugins/platforms/dingtalk/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/homeassistant/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/sms/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/feishu/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/wecom/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/irc/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/ntfy/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/line/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/teams/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/mattermost/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/photon/auth.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/photon/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/buzz/adapter.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHgateway/platforms/api_server.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHgateway/platforms/bluebubbles.py0scope-aware credential read with the default-profile startup fallback. secondary profiles construct their adapters underSTRING
HIGHplugins/platforms/dingtalk/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/discord/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/homeassistant/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/sms/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/feishu/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/a2a/__init__.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/telegram/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/raft/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/teams/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/slack/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/mattermost/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/matrix/adapter.py0plugin entry point — called by the hermes plugin system.STRING
HIGHplugins/platforms/email/adapter.py0plugin entry point — called by the hermes plugin system.STRING
188 more matches not shown…
Modern AI Meta-Vocabulary357 hits · 1012 pts
SeverityFileLineSnippetContext
MEDIUMbatch_runner.py1110 # Check for invalid tool names (model hallucinations)COMMENT
MEDIUMrun_agent.py228# recovery scaffolding: the synthetic assistant "(empty)" turn and user nudgeCOMMENT
MEDIUMrun_agent.py272# list (e.g. by scaffolding rewind or in-place compaction) and garbage-COMMENT
MEDIUMrun_agent.py2070 # Pass 1: strip the flagged scaffolding messages themselves.COMMENT
MEDIUMrun_agent.py2083 # Pass 2: if we stripped scaffolding, rewind through any trailingCOMMENT
MEDIUMrun_agent.py2087 # result. Only runs when scaffolding was actually present — normalCOMMENT
MEDIUMrun_agent.py2220 # skipped as ephemeral scaffolding / non-dict), and no code pathCOMMENT
MEDIUMrun_agent.py2243 # Never write ephemeral recovery scaffolding to the sessionCOMMENT
MEDIUMrun_agent.py3218 # Mirror the SQLite flush: ephemeral recovery scaffolding isCOMMENT
MEDIUMrun_agent.py4228 # each turn for repeatable testing, bypassing real headers. Throwaway scaffolding.COMMENT
MEDIUMtoolsets.py75 # Kanban multi-agent coordination — only in schema when the agent isCOMMENT
MEDIUMpyproject.toml222 # openWakeWord's onnx embedding model scores near-zero on macOS ARM64COMMENT
MEDIUMCONTRIBUTING.md221├── model_tools.py # Tool orchestration (thin layer over tools/registry.py)CODE
MEDIUMCONTRIBUTING.md250│ ├── terminal_tool.py # Terminal orchestration (sudo, env lifecycle, backends)CODE
MEDIUMhermes_state.py12513 # scaffold. Keep the child carrier (and its durable rowCOMMENT
MEDIUMhermes_state.py100try: # Hard dependency, but tolerate scaffold-phase imports before pip install.CODE
MEDIUMhermes_state.py102except ImportError: # pragma: no cover - stripped/scaffold installs onlyCODE
MEDIUMmodel_tools.py515 # descriptions that don't actually exist, and hallucinates calls to them.COMMENT
MEDIUMmodel_tools.py562 # "prefer web_search or web_extract" which causes the model to hallucinateCOMMENT
MEDIUMmodel_tools.py1290 # downstream hook (pre/post, edit approval, guardrails) sees the realCOMMENT
MEDIUMAGENTS.md272├── model_tools.py # Tool orchestration, discover_builtin_tools(), handle_function_call()CODE
MEDIUMAGENTS.md1226## Kanban (multi-agent work queue)COMMENT
MEDIUMtools/wake_word.py556 # embedding model is broken, so the listener would arm and never fire.COMMENT
MEDIUMtools/voice_mode.py1209# Whisper hallucination filterCOMMENT
MEDIUMtools/voice_mode.py1211# Whisper commonly hallucinates these phrases on silent/near-silent audio.COMMENT
MEDIUMtools/voice_mode.py1230 # Non-English hallucinations (common on silence)COMMENT
MEDIUMtools/voice_mode.py1242# Regex patterns for repetitive hallucinations (e.g. "Thank you. Thank you. Thank you.")COMMENT
MEDIUMtools/voice_mode.py1425 # Filter out Whisper hallucinations (common on silent/near-silent audio).COMMENT
MEDIUMtools/voice_mode.py1427 # phrases like "bye" or "okay" overlap the hallucination blocklist/repeatCOMMENT
MEDIUMtools/tool_search.py94 # Listing budget as a percentage of the model's context window. UnderCOMMENT
MEDIUMtools/lazy_deps.py158 # openWakeWord's ONNX embedding model returns near-zero scores on macOSCOMMENT
MEDIUMtools/tts_tool.py274# context window. Users can override any of these viaCOMMENT
MEDIUMtools/tts_tool.py283 "gemini": 32000, # Gemini TTS has a 32k-token context window; char cap is conservativeCODE
MEDIUMtools/transcription_tools.py2774# Whisper hallucinates junk tokens on silent stretches exactly like localCOMMENT
MEDIUMtools/transcription_tools.py1807# Silence-hallucination hardening defaults for local faster-whisper.COMMENT
MEDIUMtools/budget_config.py25# threshold and silently bloat context — in agentic evals this measurablyCOMMENT
MEDIUMtools/budget_config.py125# Fraction of a model's context window we allow a SINGLE tool result to occupyCOMMENT
MEDIUMtools/approval.py3173# Approval prompting + orchestrationCOMMENT
MEDIUMtools/subagent_worktree.py8(https://dev.meta.ai/docs/muse-code/extending#multi-agent); no Muse CodeSTRING
MEDIUMdocker/tini-shim.sh3# /usr/bin/tini — compatibility shim for legacy orchestration templates.COMMENT
MEDIUMdocker/entrypoint.sh9# orchestration script that invokes it inside the container), forward toCOMMENT
MEDIUMhermes_cli/auth.py6043 # Skip Hermes models — they're not reliable for agentic tool-callingCOMMENT
MEDIUMhermes_cli/gateway.py6357 # is diagnostic scaffolding; cheap to keep on, costs nothing duringCOMMENT
MEDIUMhermes_cli/models.py394 # Third-party agentic models hosted on build.nvidia.comCOMMENT
MEDIUMhermes_cli/models.py656 # Curated HF model list — only agentic models that map to OpenRouter defaults.COMMENT
MEDIUMhermes_cli/models.py3764# - "openrouter": curated list is already a hand-picked agentic subset ofCOMMENT
MEDIUMhermes_cli/models.py4037 # with our curated agentic catalog so ``/model`` matches whatCOMMENT
MEDIUMhermes_cli/models.py4141 # agentic picks lead the picker instead of whatever the liveCOMMENT
MEDIUMhermes_cli/goals.py1405# GoalManager — the orchestration surface CLI + gateway talk toCOMMENT
MEDIUMhermes_cli/default_soul.py15# These templates contain no persona text -- they are pure comment scaffolding,COMMENT
MEDIUMhermes_cli/default_soul.py42 # block / trailing newline in some historical revisions; the bare scaffoldCOMMENT
MEDIUMhermes_cli/agent_import.py463 # -- orchestration -----------------------------------------------------COMMENT
MEDIUMhermes_cli/model_switch.py355# Non-agentic model warningCOMMENT
MEDIUMhermes_cli/model_switch.py3065 # `hermes model` picker deliberately shows ONLY the curated agenticCOMMENT
MEDIUMhermes_cli/update_lock.py69# Set by an orchestrating updater (the Tauri `hermes-setup --update` flow) toCOMMENT
MEDIUMhermes_cli/web_server.py5326 # hallucinations and maps provider "empty transcript" errors to aCOMMENT
MEDIUMhermes_cli/model_setup_flows.py453 # shows only agentic models users recognize from OpenRouter.COMMENT
MEDIUMhermes_cli/model_setup_flows.py2947 # 1. models.dev registry (cached, filtered for agentic/tool-capable models)COMMENT
MEDIUMhermes_cli/loops.py524# LoopManager — the orchestration surface CLI + gateway + TUI talk toCOMMENT
MEDIUMhermes_cli/config_defaults.py1061 # See: https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails.htmlCOMMENT
297 more matches not shown…
Structural Annotation Overuse614 hits · 979 pts
SeverityFileLineSnippetContext
LOWhermes_state_search.py1248 # Step 3: Collapse repeated * (e.g. "***") into a single one,COMMENT
LOWhermes_state_search.py1253 # Step 4: Remove dangling boolean operators at start/end that wouldCOMMENT
LOWhermes_state_search.py1258 # Step 5: Wrap unquoted dotted and/or hyphenated terms in doubleCOMMENT
LOWhermes_state_search.py1266 # Step 6: Restore preserved quoted phrasesCOMMENT
LOWhermes_state_search.py1201 # Step 1: Extract balanced double-quoted phrases and protect themCOMMENT
LOWhermes_state_search.py1227 # Step 2: Strip remaining (unmatched) FTS5-special characters. ``:`` isCOMMENT
LOWtools/skills_hub.py1746 # Step 1: fetch the sitemap index → list of skill-sitemap URLs.COMMENT
LOWtools/skills_hub.py1768 # Step 2: fetch each skill sitemap and collect canonical "owner/repo/skill" IDs.COMMENT
LOWtools/skills_sync.py1136 # Step 2: drop the manifest entry so next sync treats it as newCOMMENT
LOWtools/skills_sync.py1141 # Step 3: run sync to re-baseline (or re-copy if we deleted)COMMENT
LOWtools/mcp_oauth_manager.py837 # Step 1: Did disk change? Picks up external refresh.COMMENT
LOWtools/mcp_oauth_manager.py846 # Step 2: No disk change — if the SDK can refreshCOMMENT
LOWtools/mcp_oauth_manager.py340 # Step 1: PRM discovery to learn the authorization_server URL.COMMENT
LOWtools/mcp_oauth_manager.py362 # Step 2: ASM discovery against the auth_server_url (orCOMMENT
LOWtools/yuanbao_tools.py372 # Step 2: Send text DM + mediaCOMMENT
LOWtools/yuanbao_tools.py382 # Step 3: Send media filesCOMMENT
LOWtools/yuanbao_tools.py324 # Step 1: Resolve user_id from group member list if not providedCOMMENT
LOWtools/send_message_tool.py2261 # Step 1: Get access tokenCOMMENT
LOWtools/send_message_tool.py2273 # Step 2: Send message via RESTCOMMENT
LOWtools/computer_use/cua_backend.py3173 # Step 1: enumerate on-screen windows to find target pid/window_id.COMMENT
LOWtools/computer_use/cua_backend.py3188 # Step 0: explicit full-screen capture — a composited grab ofCOMMENT
LOWtools/computer_use/cua_backend.py3311 # Step 2: capture.COMMENT
LOWhermes_cli/auth.py8431 # Step 2: Show user the codeCOMMENT
LOWhermes_cli/auth.py8439 # Step 3: Poll for authorization codeCOMMENT
LOWhermes_cli/auth.py8362 # Step 1: Request device code. OpenAI's auth endpoint rate-limits thisCOMMENT
LOWhermes_cli/auth.py8474 # Step 4: Exchange authorization code for tokensCOMMENT
LOWhermes_cli/skills_config.py170 # Step 1: Select platformCOMMENT
LOWhermes_cli/skills_config.py174 # Step 2: Select mode — individual or by categoryCOMMENT
LOWhermes_cli/dingtalk_auth.py66 # Step 1: init → nonceCOMMENT
LOWhermes_cli/dingtalk_auth.py72 # Step 2: begin → device_code, verification_uri_completeCOMMENT
LOWhermes_cli/send_cmd.py251 # Step 1: dotenvCOMMENT
LOWhermes_cli/send_cmd.py287 # Step 2: bridge top-level config.yaml values into the environment soCOMMENT
LOWhermes_cli/web_server.py12099 # Step 1: request device codeCOMMENT
LOWhermes_cli/web_server.py12130 # Step 2: poll until authorizedCOMMENT
LOWhermes_cli/web_server.py12164 # Step 3: exchange authorization_code for tokensCOMMENT
LOWhermes_cli/model_setup_flows.py2144 # Step 1: Check / prompt for API keyCOMMENT
LOWhermes_cli/model_setup_flows.py2156 # Step 2: Auto-detect endpoint from key prefixCOMMENT
LOWhermes_cli/model_setup_flows.py2169 # Step 3: Model selection — show appropriate models for the endpointCOMMENT
LOWhermes_cli/setup.py3393 # Step 2: Terminal Backend — where commands run is a core decisionCOMMENT
LOWhermes_cli/setup.py3396 # Step 3: Apply defaults for everything elseCOMMENT
LOWhermes_cli/setup.py3401 # Step 4: Offer messaging gateway setupCOMMENT
LOWhermes_cli/setup.py3365 # Step 1: Nous Portal — OAuth login + model selection.COMMENT
LOWhermes_cli/tools_config.py2069 # Step 1: only the local browser provider actually needs Chromium onCOMMENT
LOWhermes_cli/tools_config.py2075 # Step 2: ensure the Chromium / headless-shell build agent-browserCOMMENT
LOWhermes_cli/copilot_auth.py278 # Step 2: Show instructionsCOMMENT
LOWhermes_cli/copilot_auth.py285 # Step 3: Poll for completionCOMMENT
LOWhermes_cli/copilot_auth.py245 # Step 1: Request device codeCOMMENT
LOWhermes_cli/main.py4021 # Step 1: Provider selection.COMMENT
LOWhermes_cli/main.py4143 # Step 2: Provider-specific setup + model selectionCOMMENT
LOWplugins/platforms/dingtalk/adapter.py1245 # Step 1: Create card with STREAM callback typeCOMMENT
LOWplugins/platforms/dingtalk/adapter.py1273 # Step 2: Deliver card to the conversationCOMMENT
LOWplugins/platforms/dingtalk/adapter.py1313 # Step 3: Stream initial content. finalize=True closes theCOMMENT
LOWplugins/platforms/homeassistant/adapter.py178 # Step 1: Receive auth_requiredCOMMENT
LOWplugins/platforms/homeassistant/adapter.py185 # Step 2: Send authCOMMENT
LOWplugins/platforms/homeassistant/adapter.py191 # Step 3: Wait for auth_okCOMMENT
LOWplugins/platforms/homeassistant/adapter.py198 # Step 4: Subscribe to state_changed eventsCOMMENT
LOWplugins/platforms/feishu/feishu_comment.py1197 # Step 2: Parallel fetch -- doc meta + comment detailsCOMMENT
LOWplugins/platforms/feishu/feishu_comment.py1216 # Step 3: Build timeline based on comment typeCOMMENT
LOWplugins/platforms/feishu/feishu_comment.py1349 # Step 4: Run agent in a thread (run_conversation is synchronous)COMMENT
LOWplugins/platforms/feishu/feishu_comment.py1362 # Step 5: Deliver replyCOMMENT
554 more matches not shown…
Modern Structural Boilerplate660 hits · 636 pts
SeverityFileLineSnippetContext
LOWbatch_runner.py46logger = logging.getLogger(__name__)CODE
LOWrun_agent.py8348 def _set_tool_guardrail_halt(self, decision: ToolGuardrailDecision) -> None:CODE
LOWrun_agent.py40logger = logging.getLogger(__name__)CODE
LOWhermes_logging.py165def set_session_context(session_id: str) -> None:CODE
LOWhermes_time.py22logger = logging.getLogger(__name__)CODE
LOWutils.py17logger = logging.getLogger(__name__)CODE
LOWhermes_state.py105logger = logging.getLogger(__name__)CODE
LOWhermes_state.py709def _set_last_init_error(msg: Optional[str]) -> None:CODE
LOWhermes_state.py6245 def set_expiry_finalized(self, session_id: str, finalized: bool = True) -> None:CODE
LOWhermes_state.py7622 def set_compression_fallback_streak(self, session_id: str, streak: int) -> None:CODE
LOWhermes_state.py7703 def set_compression_ineffective_count(self, session_id: str, count: int) -> None:CODE
LOWhermes_state.py8461 def set_session_yolo(self, session_id: str, enabled: bool) -> None:CODE
LOWmodel_tools.py43logger = logging.getLogger(__name__)CODE
LOWtools/wake_word.py42logger = logging.getLogger(__name__)CODE
LOWtools/schema_sanitizer.py44logger = logging.getLogger(__name__)CODE
LOWtools/voice_mode.py29logger = logging.getLogger(__name__)CODE
LOWtools/tool_search.py1294__all__ = [CODE
LOWtools/vision_tools.py71logger = logging.getLogger(__name__)CODE
LOWtools/bot_mode_dm.py57logger = logging.getLogger(__name__)CODE
LOWtools/process_registry.py55logger = logging.getLogger(__name__)CODE
LOWtools/web_tools.py103logger = logging.getLogger(__name__)CODE
LOWtools/lazy_deps.py84logger = logging.getLogger(__name__)CODE
LOWtools/daemon_pool.py34__all__ = ["DaemonThreadPoolExecutor"]CODE
LOWtools/skill_ledger.py41logger = logging.getLogger(__name__)CODE
LOWtools/tts_tool.py63logger = logging.getLogger(__name__)CODE
LOWtools/voice_client_config.py38logger = logging.getLogger(__name__)CODE
LOWtools/tirith_security.py39logger = logging.getLogger(__name__)CODE
LOWtools/browser_dialog_tool.py25logger = logging.getLogger(__name__)CODE
LOWtools/path_security.py12logger = logging.getLogger(__name__)CODE
LOWtools/url_safety.py40logger = logging.getLogger(__name__)CODE
LOWtools/skills_hub.py43logger = logging.getLogger(__name__)CODE
LOWtools/write_approval.py55logger = logging.getLogger(__name__)CODE
LOWtools/browser_use_cli.py19logger = logging.getLogger(__name__)CODE
LOWtools/clarify_gateway.py40logger = logging.getLogger(__name__)CODE
LOWtools/delegation_live_log.py41logger = logging.getLogger(__name__)CODE
LOWtools/registry.py30logger = logging.getLogger(__name__)CODE
LOWtools/mcp_oauth.py65logger = logging.getLogger(__name__)CODE
LOWtools/mcp_oauth.py527 async def set_tokens(self, tokens: "OAuthToken") -> None:CODE
LOWtools/mcp_oauth.py572 async def set_client_info(self, client_info: "OAuthClientInformationFull") -> None:CODE
LOWtools/transcription_tools.py54logger = logging.getLogger(__name__)CODE
LOWtools/env_probe.py42logger = logging.getLogger(__name__)CODE
LOWtools/browser_extension_router.py45logger = logging.getLogger(__name__)CODE
LOWtools/interrupt.py21logger = logging.getLogger(__name__)CODE
LOWtools/async_delegation.py53logger = logging.getLogger(__name__)CODE
LOWtools/plugin_guard.py338__all__ = [CODE
LOWtools/thread_context.py40logger = logging.getLogger(__name__)CODE
LOWtools/spill_safety.py40__all__ = [CODE
LOWtools/osv_check.py22logger = logging.getLogger(__name__)CODE
LOWtools/memory_tool.py50logger = logging.getLogger(__name__)CODE
LOWtools/__init__.py25__all__ = ["check_file_requirements"]CODE
LOWtools/code_kernel_remote.py44logger = logging.getLogger(__name__)CODE
LOWtools/skill_usage.py39logger = logging.getLogger(__name__)CODE
LOWtools/skill_usage.py993def set_state(skill_name: str, state: str) -> None:CODE
LOWtools/skill_usage.py1037def set_sync(skill_name: str, sync: bool) -> None:CODE
LOWtools/code_kernel.py66logger = logging.getLogger(__name__)CODE
LOWtools/credential_files.py36logger = logging.getLogger(__name__)CODE
LOWtools/cronjob_tools.py19logger = logging.getLogger(__name__)CODE
LOWtools/file_tools.py28logger = logging.getLogger(__name__)CODE
LOWtools/file_tools.py2087def _update_read_timestamp(filepath: str, task_id: str) -> None:CODE
LOWtools/blueprints.py39logger = logging.getLogger(__name__)CODE
600 more matches not shown…
Self-Referential Comments98 hits · 312 pts
SeverityFileLineSnippetContext
MEDIUMtrajectory_compressor.py1493 # Create a temporary directory for processingCOMMENT
MEDIUMtrajectory_compressor.py1540 # Create a temp directory with sampled filesCOMMENT
MEDIUMhermes_state_search.py730 # Create the empty v23 schema outside the write transaction —COMMENT
MEDIUMtools/skill_manager_tool.py969 # Create the skill directoryCOMMENT
MEDIUMtools/file_operations.py1286 # Create the parent dir in the SAME subprocess that writes theCOMMENT
MEDIUMtools/discord_tool.py601 # Create a standalone threadCOMMENT
MEDIUMdocker/stage2-hook.sh77# Create the directory (and any missing parents) while we still have rootCOMMENT
MEDIUMdocker/stage2-hook.sh488 # Create an empty, owner-only .env so the append below (and anyCOMMENT
MEDIUMhermes_cli/backup.py890 # Create the zipCOMMENT
MEDIUMhermes_cli/service_manager.py794 # Create the leaf and clear a stale s6-log lock as hermes whenCOMMENT
MEDIUMhermes_cli/config_defaults.py1060 # Create a guardrail in the Bedrock console, then set the ID and version here.COMMENT
MEDIUMhermes_cli/cli_commands_mixin.py1439 # Create the new session with parent link.COMMENT
MEDIUMhermes_cli/main.py7915 # Create a self-signed code-signing cert (valid 10 years) and import itCOMMENT
MEDIUMplugins/platforms/matrix/adapter.py1734 # Create the HTTP API layer.COMMENT
MEDIUMplugins/platforms/matrix/adapter.py1742 # Create the client.COMMENT
MEDIUMplugins/platforms/whatsapp/adapter.py817 # Create a persistent HTTP session for all bridge communicationCOMMENT
MEDIUMplugins/memory/honcho/__init__.py520 # Create the remote session before running startup-only migration andCOMMENT
MEDIUMtests/test_tui_gateway_server.py11213 # Create a minimal PNG file with a space in its nameCOMMENT
MEDIUMtests/test_toolsets.py73 # Create a cycle: A includes B, B includes ACOMMENT
MEDIUMtests/test_mcp_serve.py186 # Create a mock SessionDB that reads from our test DBCOMMENT
MEDIUMtests/test_mcp_serve.py1133 # Create a mock SessionDB that reads our test DBCOMMENT
MEDIUMtests/test_hermes_state.py1113 # Create a recent sessionCOMMENT
MEDIUMtests/test_timezone.py210 # Create a naive datetime — will be interpreted as system-local timeCOMMENT
MEDIUMtests/test_hermes_logging.py629 # Create a stderr-like stream with ASCII encodingCOMMENT
MEDIUMtests/test_hermes_constants.py627 # Create a symlink with fewer path componentsCOMMENT
MEDIUMtests/honcho_plugin/test_session.py465 # Create a cached session so dialectic_query doesn't bail earlyCOMMENT
MEDIUMtests/tools/test_transcription_tools.py486 # Create a directory with a valid audio extensionCOMMENT
MEDIUMtests/tools/test_docker_find.py26 # Create a fake docker binary at a known pathCOMMENT
MEDIUMtests/tools/test_browser_hardening.py205 # Create a snapshot that exceeds the summarize thresholdCOMMENT
MEDIUMtests/tools/test_send_message_tool.py871 # Create a test fileCOMMENT
MEDIUMtests/tools/test_skill_view_traversal.py24 # Create a legitimate file inside the skillCOMMENT
MEDIUMtests/tools/test_skill_view_traversal.py29 # Create a sensitive file outside skills dir (simulating .env)COMMENT
MEDIUMtests/tools/test_credential_files.py96 # Create a symlink pointing outside the skills treeCOMMENT
MEDIUMtests/tools/test_credential_files.py174 # Create a sensitive file one level above hermes_homeCOMMENT
MEDIUMtests/tools/test_credential_files.py189 # Create a fake sensitive file outside hermes_homeCOMMENT
MEDIUMtests/tools/test_credential_files.py205 # Create a file at an absolute pathCOMMENT
MEDIUMtests/tools/test_credential_files.py234 # Create a sensitive file outside hermes_homeCOMMENT
MEDIUMtests/tools/test_credential_files.py238 # Create a symlink inside hermes_home pointing outsideCOMMENT
MEDIUMtests/tools/test_docker_environment.py1398 # Create a directory that looks like a corrupted credential file pathCOMMENT
MEDIUMtests/tools/test_mcp_stability.py445 # Create a mock HermesCLI-like object with the needed attributesCOMMENT
MEDIUMtests/tools/test_skill_size_limits.py77 # Create a small skill firstCOMMENT
MEDIUMtests/tools/test_skill_size_limits.py94 # Create a skill near the limitCOMMENT
MEDIUMtests/tools/test_skill_size_limits.py113 # Create a supporting fileCOMMENT
MEDIUMtests/tools/test_vision_tools.py549 # Create a fake image file under a fake home directoryCOMMENT
MEDIUMtests/tools/test_vision_tools.py731 # Create a small 10x10 red PNGCOMMENT
MEDIUMtests/tools/test_vision_tools.py747 # Create a dummy fileCOMMENT
MEDIUMtests/tools/test_sync_back_backends.py192 # Create a fake control socket so cleanup tries the SSH exitCOMMENT
MEDIUMtests/tools/test_interrupt.py84 # Create a minimal mock agent with _interrupt_requested = TrueCOMMENT
MEDIUMtests/docker/test_toplevel_chown.py78 # Create a non-allowlisted file as rootCOMMENT
MEDIUMtests/hermes_cli/test_copilot_token_exchange.py103 # Create a minimal pconfig mockCOMMENT
MEDIUMtests/hermes_cli/test_mcp_config.py160 # Create a fake token fileCOMMENT
MEDIUMtests/hermes_cli/test_setup_openclaw_migration.py38 # Create a fake hermes home with configCOMMENT
MEDIUMtests/hermes_cli/test_web_server_files.py271 # Create a regular file and .env variants including shorthand suffixes.COMMENT
MEDIUMtests/hermes_cli/test_profile_export_credentials.py39 # Create a profile with credentialsCOMMENT
MEDIUMtests/hermes_cli/test_backup.py624 # Create a file with epoch timestamp (1970-01-01)COMMENT
MEDIUMtests/hermes_cli/test_doctor_command_install.py22 # Create a fake venv entry pointCOMMENT
MEDIUMtests/hermes_cli/test_doctor_command_install.py80 # Create a symlink pointing to wrong targetCOMMENT
MEDIUMtests/hermes_cli/test_web_server.py1590 # Create a client WITHOUT the dashboard session headerCOMMENT
MEDIUMtests/hermes_cli/test_skin_engine.py95 # Create a user skin YAMLCOMMENT
MEDIUMtests/hermes_cli/test_update_check.py20 # Create a fake git repo and fresh cacheCOMMENT
38 more matches not shown…
AI Slop Vocabulary99 hits · 263 pts
SeverityFileLineSnippetContext
MEDIUMrun_agent.py628 # to write the fork's harness turn into the user's real session. RecallCOMMENT
MEDIUMrun_agent.py2155 # here would land its harness turn ("Review the conversation above andCOMMENT
MEDIUMpackage-lock.json16789 "resolved": "https://registry.npmjs.org/robust-predicates/-/robust-predicates-3.0.3.tgz",CODE
MEDIUMtrajectory_compressor.py1272 """Print comprehensive compression summary statistics."""STRING
MEDIUMhermes_state.py12523 # fix) shared the parent's session_id wrote its harness turn into thisCOMMENT
MEDIUMhermes_state.py12524 # real session. The harness is a user/system message instructing theCOMMENT
MEDIUMhermes_state.py12528 # actual task. Strip any such harness message AND the curator-modeCOMMENT
MEDIUMhermes_state.py737# Distinctive opening shared by both background-review harness promptsCOMMENT
MEDIUMhermes_state.py786 # The curator-mode reply to the harness prompt — drop it.COMMENT
MEDIUMui-tui/src/lib/resizeCoalescer.test.ts83 // ~300ms / 30ms ≈ 10 reflows, not 30. Bound it loosely to stay robust.COMMENT
MEDIUMtools/browser_use_cli.py34# browsers (local Chrome / CDP override). The harness daemon attaches to theCOMMENT
MEDIUMtools/browser_use_cli.py38# once per daemon (marker file keyed by BU_NAME under the harness runtimeCOMMENT
MEDIUMtools/browser_use_cli.py731 # with the backend: BU_NAME namespaces the harness daemon (its IPCCOMMENT
LOWtools/browser_use_cli.py561 # provider here would just create a second, redundant session. TheCOMMENT
MEDIUMtools/browser_supervisor.py1118 # Parse query string for dialog metadata. Use urllib to be robust.COMMENT
LOWtools/kanban_tools.py788 # the write txn), so the worker can simply callCOMMENT
LOWtools/kanban_tools.py847 # this, a worker that learns kanban_complete is gated can just callCOMMENT
MEDIUMtools/environments/base.py632 # (re-exported by every _wrap_command with outer-harness-preservingCOMMENT
MEDIUMtools/environments/base.py635 # harness value arriving via the process env, exactly like theCOMMENT
MEDIUMtools/environments/base.py969 # agent-harness registry (``hermes-agent`` — see huggingface.jsCOMMENT
MEDIUMtools/environments/base.py975 # never clobber an outer harness value that arrived via the inheritedCOMMENT
MEDIUMdocker/hermes-exec-shim.sh69# Reference it by absolute path so the drop is robust against PATHCOMMENT
MEDIUMhermes_cli/backup.py1173 # so the two agree: neither step can re-elevate the restored file.COMMENT
MEDIUMhermes_cli/gateway.py6502 # Absolutely everything else: Exception, asyncio.CancelledError,COMMENT
LOWhermes_cli/models.py3547 # Already on openrouter, just return the resolved slugCOMMENT
LOWhermes_cli/web_server.py8071 pass # can't read disk config — just use the string formCODE
LOWhermes_cli/kanban.py1482 # dispatcher doesn't need to enumerate — we just pass the nameCOMMENT
MEDIUMhermes_cli/main.py13110 # under an AI agent harness.COMMENT
MEDIUMhermes_cli/main.py7376 # robust even if the hook is somehow skipped.)COMMENT
MEDIUMplugins/platforms/discord/voice_mixer.py107 # Pad to a whole number of frames so looping is seamless and the finalCOMMENT
MEDIUMplugins/platforms/telegram/adapter.py10107 # Normalize mime_type for robust comparisons (some clients sendCOMMENT
MEDIUMplugins/kanban/dashboard/plugin_api.py2612# the simplest and most robust approach; it adds a fraction of a percentCOMMENT
MEDIUMoptional-skills/security/unbroker/scripts/autopilot.py235 # broker (the whole CA Data Broker Registry) -- the highest-leverage removal there is.COMMENT
MEDIUMoptional-skills/security/unbroker/scripts/registry.py74# Field label -> substring to locate its column on the header row (robust toCOMMENT
MEDIUMweb/src/components/ChatSessionList.tsx131 // "New chat" prefers ChatPage's robust handler (clears resume + forces aCOMMENT
MEDIUMweb/src/pages/SkillsPage.tsx149 // call signature supports it (clearer, and robust if a caller bypassesCOMMENT
MEDIUMwebsite/package-lock.json17200 "resolved": "https://registry.npmjs.org/robust-predicates/-/robust-predicates-3.0.3.tgz",CODE
MEDIUMtests/test_background_review_session_isolation.py36 # An assistant message that quotes the harness text is not itself a harness prompt.COMMENT
MEDIUMtests/tools/test_delegate.py1230 # Use tiny thresholds so the assertion is scheduler-robust in CI:COMMENT
LOWtests/tools/test_env_probe.py142 # Should not raise, should just return ""COMMENT
MEDIUMtests/docker/test_gateway_run_supervised.py141 # harness profile has a configured model, but the want-intentCOMMENT
LOWtests/hermes_cli/test_codex_runtime_plugin_migration.py402 # We can't easily create one in the test, so just use a stable pathCOMMENT
MEDIUMtests/hermes_cli/test_agent_env_advertisement.py3Port of earendil-works/pi#7493: entry points advertise the agent harness toSTRING
MEDIUMtests/hermes_cli/test_model_switch_parsing.py92 # api_server._clean_request_string equivalent for the parity harness.COMMENT
MEDIUMtests/hermes_cli/test_kanban_notify.py618 # dispatcher owner, which this focused watcher harness represents.COMMENT
MEDIUM…sts/hermes_cli/test_model_picker_excluded_providers.py81 # alias from _PROVIDER_ALIASES to stay robust to renames.COMMENT
LOWtests/hermes_cli/test_commands.py1057 # Both calls should succeed and just return the built-in set.COMMENT
LOWtests/hermes_cli/test_voice_wrapper.py174 # Should simply return None without raising.COMMENT
MEDIUMtests/plugins/test_hindsight_root_guard.py57 # sys.stderr write under the isolation harness, so assert on the log.COMMENT
LOWtests/plugins/test_langfuse_plugin.py160 # Each hook should just return; no exceptions.COMMENT
MEDIUMtests/plugins/transcription/check_parity_vs_main.py192 # so the harness can distinguish command-provider dispatch from aCOMMENT
MEDIUMtests/plugins/tts/check_parity_vs_main.py198 # This is the ONLY acceptable diff in the harness.STRING
MEDIUMtests/agent/test_moa_cold_start_cache_66793.py221# ─── test harness helpers ──────────────────────────────────────────────COMMENT
MEDIUM…/agent/test_rotation_flush_persisted_boundary_68196.py107 # which is then ended. Read it back verbatim (include_inactive to be robustCOMMENT
MEDIUMtests/agent/test_codex_ttfb_watchdog.py49 # the test is robust to detection-logic changes elsewhere.COMMENT
MEDIUMtests/agent/test_org_skill_namespace.py206 # highest-leverage ones in the system.COMMENT
LOWtests/agent/test_context_engine.py48 # Trivial: just return as-isCOMMENT
MEDIUMtests/agent/test_pre_compress_checkpoint_contract.py413 # this harness genuinely exercises the call site (the finalizer swallowsCOMMENT
MEDIUMtests/agent/test_coding_context.py257# ── edit-format steering (per-model harness tuning) ──────────────────────────COMMENT
MEDIUM…ts/agent/test_context_engine_on_turn_complete_usage.py26# Reuse the minimal agent harness that exercises the real finalize_turn path.COMMENT
39 more matches not shown…
Fake / Example Data268 hits · 254 pts
SeverityFileLineSnippetContext
LOWui-tui/scripts/billing-fixtures.tsx62 org_name: 'Acme Inc',CODE
LOWui-tui/scripts/billing-fixtures.tsx119 org_name: 'Acme Inc',CODE
LOWhermes_cli/auth.py667 "placeholder",CODE
LOWhermes_cli/web_server.py5997 "placeholder": field.placeholder,CODE
LOWhermes_cli/web_server.py6183 entry["placeholder"] = hostCODE
LOWhermes_cli/web_server.py6770 "placeholder": str(raw.get("placeholder") or ""),CODE
LOWhermes_cli/web_server.py6915 "placeholder": field["placeholder"],CODE
LOWhermes_cli/skin_engine.py1031 "placeholder": f"{dim} italic",CODE
LOWplugins/platforms/feishu/adapter.py328 "placeholder",CODE
LOWplugins/platforms/slack/adapter.py786 "placeholder",CODE
LOWplugins/memory/holographic/store.py452 1. Capitalized multi-word phrases e.g. "John Doe"CODE
LOWoptional-skills/mlops/research/dspy/SKILL.md358print(result.person.name) # "John Doe"CODE
LOW…ional-skills/mlops/research/dspy/references/modules.md236print(result.person.name) # "John Doe"CODE
LOWoptional-skills/mlops/inference/outlines/SKILL.md96print(user.name) # "John Doe"CODE
LOWoptional-skills/mlops/instructor/SKILL.md70print(user.name) # "John Doe"CODE
LOWoptional-skills/mlops/instructor/SKILL.md636 "name": "John Doe",STRING
LOWoptional-skills/mlops/instructor/SKILL.md678 "name": "John Doe",CODE
LOWoptional-skills/productivity/here-now/SKILL.md140 -d '{"email": "user@example.com"}'CODE
LOWoptional-skills/productivity/here-now/SKILL.md149 -d '{"email":"user@example.com","code":"ABCD-2345"}'CODE
LOW…/user-guide/skills/bundled/creative/creative-sketch.md105- Realistic fake content — actual sentences, actual names, not "Lorem ipsum"CODE
LOW…user-guide/skills/bundled/creative/creative-pretext.md67- **Real source/text, not lorem ipsum.** The corpus should mean something. Short manifestos, poetry, real source code, aCODE
LOW…user-guide/skills/bundled/creative/creative-pretext.md217- [ ] Corpus is real prose, not lorem ipsum, and matches the demo's conceptCODE
LOW…ide/skills/bundled/productivity/productivity-notion.md332- **Email:** `{"email": "user@example.com"}`CODE
LOW…e/skills/bundled/productivity/productivity-nano-pdf.md62nano-pdf edit contract.pdf 2 "Change the client name from 'Acme Corp' to 'Acme Industries'"CODE
LOW…e/skills/bundled/productivity/productivity-airtable.md122- Exact match: `{Email}='user@example.com'`CODE
LOW…e/skills/bundled/productivity/productivity-airtable.md184 {"fields": {"Email": "user@example.com", "Status": "Active"}}CODE
LOW…s/user-guide/skills/optional/mlops/mlops-instructor.md87print(user.name) # "John Doe"CODE
LOW…s/user-guide/skills/optional/mlops/mlops-instructor.md653 "name": "John Doe",STRING
LOW…s/user-guide/skills/optional/mlops/mlops-instructor.md695 "name": "John Doe",CODE
LOW…ser-guide/skills/optional/mlops/mlops-research-dspy.md375print(result.person.name) # "John Doe"CODE
LOW…uide/skills/optional/mlops/mlops-inference-outlines.md113print(user.name) # "John Doe"CODE
LOW…/skills/optional/productivity/productivity-here-now.md154 -d '{"email": "user@example.com"}'CODE
LOW…/skills/optional/productivity/productivity-here-now.md163 -d '{"email":"user@example.com","code":"ABCD-2345"}'CODE
LOW…/user-guide/skills/bundled/creative/creative-sketch.md103- 真实的虚假内容——实际句子、实际姓名,而非"Lorem ipsum"CODE
LOW…user-guide/skills/bundled/creative/creative-pretext.md67- **使用真实语料,而非 lorem ipsum。** 语料库应有意义。短篇宣言、诗歌、真实源代码、发现的文本、库自身的 README —— 绝不用 `lorem ipsum`。CODE
LOW…user-guide/skills/bundled/creative/creative-pretext.md217- [ ] 语料库为真实散文,非 lorem ipsum,且与演示概念匹配CODE
LOW…ide/skills/bundled/productivity/productivity-notion.md332- **邮箱(Email):** `{"email": "user@example.com"}`CODE
LOW…e/skills/bundled/productivity/productivity-nano-pdf.md61nano-pdf edit contract.pdf 2 "Change the client name from 'Acme Corp' to 'Acme Industries'"CODE
LOW…e/skills/bundled/productivity/productivity-airtable.md122- 精确匹配:`{Email}='user@example.com'`CODE
LOW…e/skills/bundled/productivity/productivity-airtable.md184 {"fields": {"Email": "user@example.com", "Status": "Active"}}CODE
LOW…t/user-guide/skills/optional/mlops/mlops-instructor.md87print(user.name) # "John Doe"CODE
LOW…t/user-guide/skills/optional/mlops/mlops-instructor.md653 "name": "John Doe",STRING
LOW…t/user-guide/skills/optional/mlops/mlops-instructor.md695 "name": "John Doe",CODE
LOW…uide/skills/optional/mlops/mlops-inference-outlines.md95print(user.name) # "John Doe"CODE
LOW…/skills/optional/productivity/productivity-here-now.md154 -d '{"email": "user@example.com"}'CODE
LOW…/skills/optional/productivity/productivity-here-now.md163 -d '{"email":"user@example.com","code":"ABCD-2345"}'CODE
LOWtests/test_iron_proxy.py637 (state / "ca.key").write_text("fake-key")CODE
LOWtests/test_tui_gateway_server.py15550 fake_key = "replacement-" + "value"CODE
LOWtests/test_tui_gateway_server.py15556 "api_key": fake_key,CODE
LOWtests/test_tui_gateway_server.py15563 save_credential.assert_called_once_with(env_var, fake_key)CODE
LOWtests/test_mcp_serve.py1297 db_path.write_text("placeholder")CODE
LOWtests/test_mcp_serve.py1344 db_path.write_text("placeholder")CODE
LOWtests/test_mcp_serve.py1390 db_path.write_text("placeholder")CODE
LOWtests/tools/test_subprocess_utf8_encoding.py49 returncode=0, stdout="user@example.com", stderr=""CODE
LOWtests/tools/test_local_env_blocklist.py79 "OPENAI_API_KEY": "sk-fake-key",CODE
LOWtests/tools/test_local_env_blocklist.py80 "OPENROUTER_API_KEY": "or-fake-key",CODE
LOWtests/tools/test_local_env_blocklist.py81 "ANTHROPIC_API_KEY": "ant-fake-key",CODE
LOWtests/tools/test_browser_secret_exfil.py215 fake_key = "sk-" + "FAKESECRETVALUE1234567890ABCDEF"CODE
LOWtests/tools/test_browser_secret_exfil.py218 f"text: API Key: {fake_key}\n"CODE
LOWtests/tools/test_browser_secret_exfil.py244 fake_token = "ghp_" + "FAKEGITHUBTOKEN12345678901234"CODE
208 more matches not shown…
Cross-Language Confusion (JS/TS)34 hits · 245 pts
SeverityFileLineSnippetContext
HIGHapps/desktop/electron/remote-lifecycle.ts303 print('CLEAR');raise SystemExitCODE
HIGHapps/desktop/electron/remote-lifecycle.ts305 print('UNCERTAIN');raise SystemExitCODE
HIGHapps/desktop/electron/remote-lifecycle.ts307 print('UNCERTAIN');raise SystemExitCODE
HIGHapps/desktop/electron/remote-lifecycle.ts310 print('UNCERTAIN');raise SystemExitCODE
HIGHapps/desktop/electron/remote-lifecycle.ts315 print('UNCERTAIN');raise SystemExitCODE
HIGHapps/desktop/electron/remote-lifecycle.ts319 print('CLEAR')CODE
HIGHapps/desktop/electron/remote-lifecycle.ts321 print('LIVE:'+str(owner))CODE
HIGHapps/desktop/electron/remote-lifecycle.ts324 elif error.errno==errno.EPERM:print('LIVE:'+str(owner))CODE
HIGHapps/desktop/electron/remote-lifecycle.ts327 print('LIVE:'+str(owner))CODE
HIGHapps/desktop/electron/remote-lifecycle.ts851 except (ValueError,IndexError):return FalseCODE
HIGHapps/desktop/electron/remote-lifecycle.ts856 print("UNAVAILABLE");sys.exit(2)CODE
HIGHapps/desktop/electron/remote-lifecycle.ts864 print("REFUSED");sys.exit(3)CODE
HIGHapps/desktop/electron/remote-lifecycle.ts869 print("DARWIN_UNAVAILABLE");sys.exit(2)CODE
HIGHapps/desktop/electron/remote-lifecycle.ts885 print("TERMINATED")CODE
HIGHapps/desktop/electron/managed-ssh-update.ts334 os.kill(pid,0);return TrueCODE
HIGHapps/desktop/electron/managed-ssh-update.ts335 except ProcessLookupError:return FalseCODE
HIGHapps/desktop/electron/managed-ssh-update.ts336 except PermissionError:return TrueCODE
HIGHapps/desktop/electron/managed-ssh-update.ts337 except OSError:return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts348 return False if ctypes.get_last_error()==87 else NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts351 if not kernel.GetExitCodeProcess(handle,ctypes.byref(code)):return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts354 except Exception:return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts362 except (OSError,UnicodeError,IndexError):return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts368 except (OSError,subprocess.CalledProcessError):return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts369 return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts379 if not handle:return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts382 if not kernel.GetProcessTimes(handle,ctypes.byref(created),ctypes.byref(exited),ctypes.byref(kernel_time),ctCODE
HIGHapps/desktop/electron/managed-ssh-update.ts386 except Exception:return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts428 return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts432 except (FileNotFoundError,OSError,UnicodeError,ValueError):return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts433 if not isinstance(payload,dict) or payload.get('correlation_id')!=correlation:return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts404 except FileNotFoundError:return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts413 except OSError:return NoneCODE
HIGHapps/desktop/electron/managed-ssh-update.ts453print(json.dumps({'marker':state['state'],'markerPid':state.get('pid'),'launchIntent':launch_intent(),'exitCode':terminaCODE
HIGHapps/desktop/src/app/settings/helpers.ts216// `if key in BUILTIN_*_PROVIDERS: return None`), so a ``providers.edge`` blockCOMMENT
Verbosity Indicators116 hits · 195 pts
SeverityFileLineSnippetContext
LOWhermes_state_search.py1248 # Step 3: Collapse repeated * (e.g. "***") into a single one,COMMENT
LOWhermes_state_search.py1253 # Step 4: Remove dangling boolean operators at start/end that wouldCOMMENT
LOWhermes_state_search.py1258 # Step 5: Wrap unquoted dotted and/or hyphenated terms in doubleCOMMENT
LOWhermes_state_search.py1266 # Step 6: Restore preserved quoted phrasesCOMMENT
LOWhermes_state_search.py1201 # Step 1: Extract balanced double-quoted phrases and protect themCOMMENT
LOWhermes_state_search.py1227 # Step 2: Strip remaining (unmatched) FTS5-special characters. ``:`` isCOMMENT
LOWtools/skills_hub.py1746 # Step 1: fetch the sitemap index → list of skill-sitemap URLs.COMMENT
LOWtools/skills_hub.py1768 # Step 2: fetch each skill sitemap and collect canonical "owner/repo/skill" IDs.COMMENT
LOWtools/skills_sync.py1136 # Step 2: drop the manifest entry so next sync treats it as newCOMMENT
LOWtools/skills_sync.py1141 # Step 3: run sync to re-baseline (or re-copy if we deleted)COMMENT
LOWtools/mcp_oauth_manager.py837 # Step 1: Did disk change? Picks up external refresh.COMMENT
LOWtools/mcp_oauth_manager.py846 # Step 2: No disk change — if the SDK can refreshCOMMENT
LOWtools/mcp_oauth_manager.py340 # Step 1: PRM discovery to learn the authorization_server URL.COMMENT
LOWtools/mcp_oauth_manager.py362 # Step 2: ASM discovery against the auth_server_url (orCOMMENT
LOWtools/yuanbao_tools.py372 # Step 2: Send text DM + mediaCOMMENT
LOWtools/yuanbao_tools.py382 # Step 3: Send media filesCOMMENT
LOWtools/yuanbao_tools.py324 # Step 1: Resolve user_id from group member list if not providedCOMMENT
LOWtools/send_message_tool.py2261 # Step 1: Get access tokenCOMMENT
LOWtools/send_message_tool.py2273 # Step 2: Send message via RESTCOMMENT
LOWtools/computer_use/cua_backend.py3173 # Step 1: enumerate on-screen windows to find target pid/window_id.COMMENT
LOWtools/computer_use/cua_backend.py3188 # Step 0: explicit full-screen capture — a composited grab ofCOMMENT
LOWtools/computer_use/cua_backend.py3311 # Step 2: capture.COMMENT
LOWhermes_cli/auth.py8431 # Step 2: Show user the codeCOMMENT
LOWhermes_cli/auth.py8439 # Step 3: Poll for authorization codeCOMMENT
LOWhermes_cli/auth.py8362 # Step 1: Request device code. OpenAI's auth endpoint rate-limits thisCOMMENT
LOWhermes_cli/auth.py8474 # Step 4: Exchange authorization code for tokensCOMMENT
LOWhermes_cli/skills_config.py170 # Step 1: Select platformCOMMENT
LOWhermes_cli/skills_config.py174 # Step 2: Select mode — individual or by categoryCOMMENT
LOWhermes_cli/dingtalk_auth.py66 # Step 1: init → nonceCOMMENT
LOWhermes_cli/dingtalk_auth.py72 # Step 2: begin → device_code, verification_uri_completeCOMMENT
LOWhermes_cli/send_cmd.py251 # Step 1: dotenvCOMMENT
LOWhermes_cli/send_cmd.py287 # Step 2: bridge top-level config.yaml values into the environment soCOMMENT
LOWhermes_cli/web_server.py12099 # Step 1: request device codeCOMMENT
LOWhermes_cli/web_server.py12130 # Step 2: poll until authorizedCOMMENT
LOWhermes_cli/web_server.py12164 # Step 3: exchange authorization_code for tokensCOMMENT
LOWhermes_cli/model_setup_flows.py2144 # Step 1: Check / prompt for API keyCOMMENT
LOWhermes_cli/model_setup_flows.py2156 # Step 2: Auto-detect endpoint from key prefixCOMMENT
LOWhermes_cli/model_setup_flows.py2169 # Step 3: Model selection — show appropriate models for the endpointCOMMENT
LOWhermes_cli/setup.py3393 # Step 2: Terminal Backend — where commands run is a core decisionCOMMENT
LOWhermes_cli/setup.py3396 # Step 3: Apply defaults for everything elseCOMMENT
LOWhermes_cli/setup.py3401 # Step 4: Offer messaging gateway setupCOMMENT
LOWhermes_cli/setup.py3365 # Step 1: Nous Portal — OAuth login + model selection.COMMENT
LOWhermes_cli/tools_config.py2069 # Step 1: only the local browser provider actually needs Chromium onCOMMENT
LOWhermes_cli/tools_config.py2075 # Step 2: ensure the Chromium / headless-shell build agent-browserCOMMENT
LOWhermes_cli/copilot_auth.py278 # Step 2: Show instructionsCOMMENT
LOWhermes_cli/copilot_auth.py285 # Step 3: Poll for completionCOMMENT
LOWhermes_cli/copilot_auth.py245 # Step 1: Request device codeCOMMENT
LOWhermes_cli/main.py4021 # Step 1: Provider selection.COMMENT
LOWhermes_cli/main.py4143 # Step 2: Provider-specific setup + model selectionCOMMENT
LOWplugins/platforms/dingtalk/adapter.py1245 # Step 1: Create card with STREAM callback typeCOMMENT
LOWplugins/platforms/dingtalk/adapter.py1273 # Step 2: Deliver card to the conversationCOMMENT
LOWplugins/platforms/dingtalk/adapter.py1313 # Step 3: Stream initial content. finalize=True closes theCOMMENT
LOWplugins/platforms/homeassistant/adapter.py178 # Step 1: Receive auth_requiredCOMMENT
LOWplugins/platforms/homeassistant/adapter.py185 # Step 2: Send authCOMMENT
LOWplugins/platforms/homeassistant/adapter.py191 # Step 3: Wait for auth_okCOMMENT
LOWplugins/platforms/homeassistant/adapter.py198 # Step 4: Subscribe to state_changed eventsCOMMENT
LOWplugins/platforms/feishu/feishu_comment.py1197 # Step 2: Parallel fetch -- doc meta + comment detailsCOMMENT
LOWplugins/platforms/feishu/feishu_comment.py1216 # Step 3: Build timeline based on comment typeCOMMENT
LOWplugins/platforms/feishu/feishu_comment.py1349 # Step 4: Run agent in a thread (run_conversation is synchronous)COMMENT
LOWplugins/platforms/feishu/feishu_comment.py1362 # Step 5: Deliver replyCOMMENT
56 more matches not shown…
Magic Placeholder Names34 hits · 175 pts
SeverityFileLineSnippetContext
HIGHhermes_cli/auth.py664 "your_api_key",CODE
HIGHhermes_cli/auth.py666 "your-api-key",CODE
HIGHoptional-skills/research/pinecone-research/SKILL.md43export PINECONE_API_KEY="your-api-key"CODE
HIGH…s/devops/inference-sh-cli/references/authentication.md30export INFSH_API_KEY=your-api-keyCODE
HIGHoptional-skills/mlops/research/dspy/SKILL.md296 api_key="your-api-key", # Or set ANTHROPIC_API_KEY env varCODE
HIGHoptional-skills/mlops/research/dspy/SKILL.md308 api_key="your-api-key",CODE
HIGHoptional-skills/mlops/guidance/SKILL.md289 api_key="your-api-key" # Or set OPENAI_API_KEY env varCODE
HIGHoptional-skills/mlops/guidance/references/backends.md28 api_key="your-api-key-here"CODE
HIGHoptional-skills/mlops/guidance/references/backends.md53 api_key="your-api-key",CODE
HIGHoptional-skills/mlops/guidance/references/backends.md95 api_key="your-api-key-here"CODE
HIGHoptional-skills/mlops/guidance/references/backends.md120 api_key="your-api-key",CODE
HIGHoptional-skills/mlops/pinecone/SKILL.md56pc = Pinecone(api_key="your-api-key")CODE
HIGHoptional-skills/mlops/qdrant/SKILL.md430 api_key="your-api-key"CODE
HIGH…onal-skills/mlops/qdrant/references/troubleshooting.md117 api_key="your-api-key"CODE
HIGH…skills/mlops/inference/outlines/references/backends.md381model = outlines.models.openai("gpt-4o-mini", api_key="your-api-key")CODE
HIGH…skills/mlops/inference/outlines/references/backends.md393 api_key="your-api-key", # Or set OPENAI_API_KEY env varCODE
HIGHoptional-skills/mlops/instructor/SKILL.md345 Anthropic(api_key="your-api-key")CODE
HIGHoptional-skills/mlops/instructor/SKILL.md363 OpenAI(api_key="your-api-key")CODE
HIGH…tional-skills/mlops/instructor/references/providers.md16 Anthropic(api_key="your-api-key")CODE
HIGH…skills/optional/research/research-pinecone-research.md60export PINECONE_API_KEY="your-api-key"CODE
HIGH…/docs/user-guide/skills/optional/mlops/mlops-qdrant.md447 api_key="your-api-key"CODE
HIGH…s/user-guide/skills/optional/mlops/mlops-instructor.md362 Anthropic(api_key="your-api-key")CODE
HIGH…s/user-guide/skills/optional/mlops/mlops-instructor.md380 OpenAI(api_key="your-api-key")CODE
HIGH…ocs/user-guide/skills/optional/mlops/mlops-pinecone.md73pc = Pinecone(api_key="your-api-key")CODE
HIGH…ser-guide/skills/optional/mlops/mlops-research-dspy.md313 api_key="your-api-key", # Or set ANTHROPIC_API_KEY env varCODE
HIGH…ser-guide/skills/optional/mlops/mlops-research-dspy.md325 api_key="your-api-key",CODE
HIGH…ocs/user-guide/skills/optional/mlops/mlops-guidance.md306 api_key="your-api-key" # Or set OPENAI_API_KEY env varCODE
HIGH…rrent/user-guide/skills/optional/mlops/mlops-qdrant.md439 api_key="your-api-key"CODE
HIGH…t/user-guide/skills/optional/mlops/mlops-instructor.md362 Anthropic(api_key="your-api-key")CODE
HIGH…t/user-guide/skills/optional/mlops/mlops-instructor.md380 OpenAI(api_key="your-api-key")CODE
HIGH…ent/user-guide/skills/optional/mlops/mlops-pinecone.md71pc = Pinecone(api_key="your-api-key")CODE
HIGH…ent/user-guide/skills/optional/mlops/mlops-guidance.md293 api_key="your-api-key" # 或设置 ANTHROPIC_API_KEY 环境变量CODE
HIGH…ent/user-guide/skills/optional/mlops/mlops-guidance.md302 api_key="your-api-key" # 或设置 OPENAI_API_KEY 环境变量CODE
HIGH…uide/skills/optional/mlops/mlops-inference-outlines.md245 api_key="your-api-key"CODE
Redundant / Tautological Comments141 hits · 158 pts
SeverityFileLineSnippetContext
LOWbatch_runner.py171 # Check if error field exists AND has a non-null valueCOMMENT
LOWhermes_constants.py218 # Check if this is a profile path: <root>/profiles/<name>COMMENT
LOWsetup-hermes.sh288 # Check if sudo is availableCOMMENT
LOWrun_agent.py1757 # Check if there's any non-whitespace content remainingCOMMENT
LOWtrajectory_compressor.py773 # Check if compression neededCOMMENT
LOWtrajectory_compressor.py788 # Check if there's anything to compressCOMMENT
LOWtrajectory_compressor.py813 # Check if we have enough savingsCOMMENT
LOWtrajectory_compressor.py908 # Check if compression neededCOMMENT
LOWtrajectory_compressor.py923 # Check if there's anything to compressCOMMENT
LOWhermes_state.py12254 # Check if the current node has messages.COMMENT
LOWtools/vision_tools.py1758 # Check if vision model is availableCOMMENT
LOWtools/process_registry.py1536 # Check if process is still runningCOMMENT
LOWtools/web_tools.py1577 # Check if API keys are availableCOMMENT
LOWtools/skills_hub.py982 # Check if ANY entry lives under the target pathCOMMENT
LOWtools/fuzzy_match.py983 # Check if this block matchesCOMMENT
LOWtools/browser_tool.py2426 # Check if the daemon is still alive. ``os.kill(pid, 0)`` on WindowsCOMMENT
LOWtools/browser_tool.py2779 # Check if we already have a session for this taskCOMMENT
LOWtools/browser_tool.py2962 # Check if it's in PATH (global install)COMMENT
LOWtools/browser_tool.py5288 # Check if screenshot file was createdCOMMENT
LOWtools/browser_tool.py5575 # Check if session exists (under lock), but don't remove yet -COMMENT
LOWtools/terminal_tool.py3247 # Check if this is an approval_required (gateway ask mode)COMMENT
LOWtools/skills_tool.py1445 # Read the file once — reused for platform check and main content belowCOMMENT
LOWtools/skills_tool.py1503 # Check if the skill is disabled by the userCOMMENT
LOWtools/skills_tool.py1591 # Read the file contentCOMMENT
LOWtools/skills_guard.py1320 # Check if source matches any trusted repo exactly, or a skill path insideCOMMENT
LOWtools/mcp_tool.py4255 # Check if shutdown was requested during the sleepCOMMENT
LOWtools/mcp_tool.py6375 # Check if the MCP tool itself returned an errorCOMMENT
LOWtools/file_operations.py1524 # Check if file exists and get size (POSIX, works on Linux + macOS)COMMENT
LOWtools/file_operations.py1555 # Check if file is too largeCOMMENT
LOWtools/file_operations.py1652 # Check if truncatedCOMMENT
LOWtools/file_operations.py2850 # Check if parent directory exists and list similar entriesCOMMENT
LOWhermes_cli/backup.py1049 # Check if all entries share a common first directoryCOMMENT
LOWhermes_cli/gateway.py7513 # Check if signal-cli is availableCOMMENT
LOWhermes_cli/config.py2172 # Check if dict keys look like they should be list-entry fieldsCOMMENT
LOWhermes_cli/config.py5618 # Check if it's an API key (goes to .env)COMMENT
LOWhermes_cli/models.py2764 # Check if this provider has credentials availableCOMMENT
LOWhermes_cli/skills_hub.py594 # Check if any source hit GitHub API rate limitCOMMENT
LOWhermes_cli/skills_hub.py673 # Check if already installedCOMMENT
LOWhermes_cli/model_switch.py2106 # Check if the requested model matches the entry's modelCOMMENT
LOWhermes_cli/model_switch.py2876 # Check if any env var is setCOMMENT
LOWhermes_cli/model_switch.py2957 # Check if credentials existCOMMENT
LOWhermes_cli/web_server.py6248 # Write the file reads resolve, or a save shadows it with a sparse copy.COMMENT
LOWhermes_cli/config_defaults.py185 # Set False to disable both.COMMENT
LOWhermes_cli/config_defaults.py245 # Set false to keep the evidence nudge terse.COMMENT
LOWhermes_cli/config_defaults.py1390 # Set false to restore the legacy c-j submit fallback on unusual POSIXCOMMENT
LOWhermes_cli/config_defaults.py1500 # Set 0 to disable the background refresh if it fights terminalCOMMENT
LOWhermes_cli/config_defaults.py1646 # Set this to True to re-enable the surfaces with the understandingCOMMENT
LOWhermes_cli/config_defaults.py1776 # Set explicitly to pin a backend:COMMENT
LOWhermes_cli/config_defaults.py1858 # Set false to keep STT for the agent while suppressing that user-facing echo.COMMENT
LOWhermes_cli/config_defaults.py3688 # Set 0 to disable.COMMENT
LOWhermes_cli/cli_commands_mixin.py2293 # Display result in the CLI (thread-safe via patch_stdout).COMMENT
LOWhermes_cli/cli_commands_mixin.py2493 # Check if a Chromium-family browser is already serving CDP on the debug port.COMMENT
LOWhermes_cli/setup.py1216 # Check if already installedCOMMENT
LOWhermes_cli/setup.py1362 # Check if already installedCOMMENT
LOWhermes_cli/setup.py1479 # Check if Docker is availableCOMMENT
LOWhermes_cli/setup.py1517 # Check if singularity/apptainer is availableCOMMENT
LOWhermes_cli/setup.py1576 # Check if modal SDK is installedCOMMENT
LOWhermes_cli/setup.py1617 # Check if daytona SDK is installedCOMMENT
LOWhermes_cli/setup.py2312 # Check if any home channels are missingCOMMENT
LOWhermes_cli/setup.py3129 # Check if a specific section was requestedCOMMENT
81 more matches not shown…
Docstring Block Structure21 hits · 105 pts
SeverityFileLineSnippetContext
HIGHtoolset_distributions.py242 Sample toolsets based on a distribution's probabilities. Each toolset in the distribution has a % chance oSTRING
HIGHtools/vision_tools.py554 Download an image from a URL to a local destination (async) with retry logic. Args: image_url (strSTRING
HIGHtools/vision_tools.py1401 Analyze an image from a URL or local file path using vision AI. This tool accepts either an HTTP/HTTPS URLSTRING
HIGHtools/web_tools.py839 Search the web for information using available search API backend. This function provides a generic interface STRING
HIGHtools/web_tools.py1053 Extract content from specific web pages using available extraction API backend. Returns clean page content (maSTRING
HIGHtools/terminal_tool.py2832 Execute a command in the configured terminal environment. Args: command: The command to execute STRING
HIGHhermes_cli/plugins.py1896Call a tool on a configured MCP server (#64204, capability-gated). Synchronous; safe to call from plugin hooks STRING
HIGHhermes_cli/plugins.py3087Register a native-client handler factory for a gateway platform. The generic surface for plugins that need to rSTRING
HIGHhermes_cli/plugins.py3164Register a python-telegram-bot handler factory from a plugin. Hermes' Telegram adapter invokes registered factoSTRING
HIGHhermes_cli/model_normalize.py291Detect the vendor slug from a bare model name. Uses the first hyphen-delimited token of the model name to look up STRING
HIGHhermes_cli/model_normalize.py428Translate a model name into the format the target provider's API expects. This is the primary entry point for modelSTRING
HIGHplugins/memory/honcho/session.py874 Query Honcho's dialectic endpoint about a peer. Runs an LLM on Honcho's backend against the target peeSTRING
HIGHplugins/memory/honcho/session.py1472 Search raw messages across every session visible from the target peer's perspective. Results include alSTRING
HIGHagent/auxiliary_client.py9722Centralized synchronous LLM call. Resolves provider + model (from task config, explicit args, or auto-detect), STRING
HIGHgateway/platforms/base.py855 Save raw image bytes to the cache and return the absolute file path. Args: data: Raw image bytes. STRING
HIGHgateway/platforms/base.py884 Download an image from a URL and save it to the local cache. Retries on transient failures (timeouts, 429, 5xxSTRING
HIGHgateway/platforms/base.py1026 Download an audio file from a URL and save it to the local cache. Retries on transient failures (timeouts, 429STRING
HIGHgateway/platforms/base.py2207 Save raw document bytes to the cache and return the absolute file path. The cached filename preserves the origSTRING
HIGHgateway/platforms/yuanbao_media.py206 下载 URL 内容,返回 (bytes, content_type)。 Args: url: HTTP(S) URL max_size_mb: 最大允许大小(MB),超STRING
HIGHgateway/platforms/yuanbao_media.py368 调用 genUploadInfo 接口获取 COS 临时密钥及上传配置。 Args: app_key: 应用 Key(用于 X-ID 头) api_domain: ASTRING
HIGHgateway/platforms/yuanbao_media.py445 通过 httpx PUT 请求将文件上传到 COS。 使用临时凭证(tmpSecretId/tmpSecretKey/sessionToken)构建 HMAC-SHA1 签名。 Args: filSTRING
AI Structural Patterns144 hits · 91 pts
SeverityFileLineSnippetContext
LOWbatch_runner.py1206CODE
LOWbatch_runner.py574CODE
LOWrun_agent.py9099CODE
LOWrun_agent.py445CODE
LOWrun_agent.py8592CODE
LOWmini_swe_runner.py630CODE
LOWmini_swe_runner.py163CODE
LOWhermes_state.py5918CODE
LOWhermes_state.py6115CODE
LOWhermes_state.py7034CODE
LOWhermes_state.py8869CODE
LOWhermes_state.py9151CODE
LOWhermes_state.py10389CODE
LOWhermes_state.py11061CODE
LOWhermes_state.py13193CODE
LOWhermes_state.py13785CODE
LOWmodel_tools.py1184CODE
LOWmodel_tools.py1240CODE
LOWhermes_state_search.py1415CODE
LOWhermes_state_search.py1709CODE
LOWtools/voice_mode.py1807CODE
LOWtools/voice_mode.py2021CODE
LOWtools/web_tools.py690CODE
LOWtools/session_search_tool.py953CODE
LOWtools/session_search_tool.py1082CODE
LOWtools/registry.py763CODE
LOWtools/tour_tool.py35CODE
LOWtools/cronjob_tools.py1460CODE
LOWtools/file_tools.py2312CODE
LOWtools/file_tools.py2532CODE
LOWtools/drive_preview_tool.py54CODE
LOWtools/terminal_tool.py2687CODE
LOWtools/terminal_tool.py2819CODE
LOWtools/skill_manager_tool.py1576CODE
LOWtools/image_generation_tool.py1200CODE
LOWtools/file_operations.py2411CODE
LOWtools/delegate_tool.py1606CODE
LOWtools/delegate_tool.py3626CODE
LOWtools/discord_tool.py986CODE
LOWtools/environments/vercel_sandbox.py248CODE
LOWtools/environments/docker.py901CODE
LOWtools/computer_use/backend.py143CODE
LOWtools/computer_use/backend.py157CODE
LOWtools/computer_use/cua_backend.py3610CODE
LOWtools/computer_use/cua_backend.py3663CODE
LOWhermes_cli/plugins.py6922CODE
LOWhermes_cli/auth.py6233CODE
LOWhermes_cli/auth.py7705CODE
LOWhermes_cli/auth.py9062CODE
LOWhermes_cli/model_switch.py1227CODE
LOWhermes_cli/model_switch.py1293CODE
LOWhermes_cli/model_switch.py2571CODE
LOWhermes_cli/model_switch.py3934CODE
LOWhermes_cli/web_server.py19628CODE
LOWhermes_cli/update_receipt.py92CODE
LOWhermes_cli/kanban_swarm.py128CODE
LOWhermes_cli/kanban_swarm.py199CODE
LOWhermes_cli/kanban_db.py3169CODE
LOWhermes_cli/kanban_db.py3661CODE
LOWhermes_cli/kanban_db.py9819CODE
84 more matches not shown…
Hallucination Indicators8 hits · 85 pts
SeverityFileLineSnippetContext
CRITICAL…tional-skills/mlops/training/axolotl/references/api.md658core.trainers.grpo.sampler.SequenceParallelRepeatRandomSampler.set_epoch(epoch)CODE
CRITICAL…tional-skills/mlops/training/axolotl/references/api.md1795core.trainers.grpo.trainer.AxolotlGRPOSequenceParallelTrainer.get_train_dataloader(CODE
CRITICAL…tional-skills/mlops/training/axolotl/references/api.md4690core.trainers.dpo.trainer.AxolotlDPOTrainer.push_to_hub(*args, **kwargs)CODE
CRITICAL…tional-skills/mlops/training/axolotl/references/api.md4892core.trainers.mixins.scheduler.SchedulerMixin.create_scheduler(CODE
CRITICALtests/cli/test_manual_compress.py195 assert shell.agent._compress_context.call_args.kwargs.get("force") is TrueCODE
CRITICALtests/run_agent/test_stream_stale_breaker_reset.py174 agent.client.chat.completions.create.assert_not_called()CODE
CRITICALtests/run_agent/test_run_agent.py2800 sent_msgs = agent.client.chat.completions.create.call_args.kwargs.get("messages", [])CODE
CRITICALtests/gateway/test_stream_consumer_thread_routing.py152 mock_client.im.v1.message.create.assert_called_once()CODE
Synthetic Comment Markers5 hits · 30 pts
SeverityFileLineSnippetContext
HIGH…ve/kanban-video-orchestrator/references/tool-matrix.md165 # - comfyui (AI-generated visuals)COMMENT
HIGH…er-guide/skills/bundled/creative/creative-humanizer.md659This skill is ported from [blader/humanizer](https://github.com/blader/humanizer) (MIT licensed), which is itself based CODE
HIGHskills/index-cache/lobehub_index.json1{"schemaVersion": 1, "agents": [{"author": "CSY2022", "createdAt": "2025-06-19", "homepage": "https://github.com/CSY2022CODE
HIGH…ls/creative/popular-web-designs/templates/replicate.md266- "Create a manifesto section on Replicate Dark (#202020) with 'Imagine what you can build.' at 128px rb-freigeist-neue CODE
HIGHskills/creative/humanizer/SKILL.md643This skill is ported from [blader/humanizer](https://github.com/blader/humanizer) (MIT licensed), which is itself based CODE
Example Usage Blocks16 hits · 26 pts
SeverityFileLineSnippetContext
LOWdocker-compose.windows.yml9# Usage:COMMENT
LOWsetup-hermes.sh8# Usage:COMMENT
LOWdocker-compose.yml4# Usage:COMMENT
LOWtools/registry.py1299# Usage:COMMENT
LOWtests/install/install-update-e2e.sh22# Usage:COMMENT
LOWscripts/kill_modal.sh4# Usage:COMMENT
LOWscripts/install.sh8# Usage:COMMENT
LOWscripts/run_tests.sh16# Usage:COMMENT
LOWscripts/sandbox/pick-release-tags.sh15# Usage:COMMENT
LOWscripts/lib/node-bootstrap.sh15# Usage:COMMENT
LOWscripts/toolperf_abeval/run_all.sh4# Usage:COMMENT
LOWskills/creative/comfyui/scripts/comfyui_setup.sh14# Usage:COMMENT
LOWskills/creative/p5js/scripts/render.sh5# Usage:COMMENT
LOWskills/creative/p5js/scripts/serve.sh5# Usage:COMMENT
LOWdatagen-config-examples/web_research.yaml6# Usage:COMMENT
LOWdatagen-config-examples/run_browser_tasks.sh17# Usage:COMMENT
Dead Code12 hits · 22 pts
SeverityFileLineSnippetContext
MEDIUMtests/hermes_cli/test_config.py924CODE
MEDIUMtests/hermes_cli/test_config.py928CODE
MEDIUMtests/hermes_cli/test_config.py936CODE
MEDIUM…ts/run_agent/test_70773_shared_client_fd_corruption.py82CODE
MEDIUM…ts/run_agent/test_70773_shared_client_fd_corruption.py132CODE
MEDIUMtests/run_agent/test_stream_stale_circuit_breaker.py133CODE
MEDIUMtests/gateway/test_discord_double_dispatch.py60CODE
MEDIUMtests/gateway/test_discord_double_dispatch.py81CODE
MEDIUMtests/gateway/test_discord_free_response.py70CODE
MEDIUMtests/gateway/test_discord_free_response.py95CODE
MEDIUMtests/gateway/test_discord_slash_commands.py472CODE
MEDIUMtests/gateway/test_discord_slash_commands.py491CODE
AI Response Leakage2 hits · 12 pts
SeverityFileLineSnippetContext
HIGH…kills/mlops/pytorch-fsdp/references/common-patterns.md11**Pattern 2:** Distributed communication package - torch.distributed# Created On: Jul 12, 2017 | Last Updated On: Sep 04CODE
HIGHskills/index-cache/lobehub_index.json1{"schemaVersion": 1, "agents": [{"author": "CSY2022", "createdAt": "2025-06-19", "homepage": "https://github.com/CSY2022CODE
Slop Phrases4 hits · 10 pts
SeverityFileLineSnippetContext
MEDIUMmcp-research-data/ue_hard_haiku_rows.json1352 "final_response": "I'm unable to capture the Unreal Editor window. The computer_use tool isn't finding any windows at CODE
MEDIUMapps/desktop/e2e/mock-server.ts288 { text: 'I cannot provide fresh verification evidence for that edit.' },CODE
MEDIUMapps/desktop/e2e/mock-server.ts293export const VERIFICATION_STOP_TEXT = 'I cannot provide fresh verification evidence for that edit.'CODE
MEDIUMdatagen-config-examples/run_browser_tasks.sh57# Common options you can add:COMMENT
Overly Generic Function Names9 hits · 8 pts
SeverityFileLineSnippetContext
LOWmini_swe_runner.py408 def run_task(self, task: str) -> Dict[str, Any]:CODE
LOWevals/readtool/runner.py74def run_task(task, model: str, provider: str, timeout_mult: float,CODE
LOWtools/url_safety.py745 def handle_request(self, request: Any) -> Any:CODE
LOWtests/tools/test_patch_parser.py244+def helper():CODE
LOWtests/tools/test_patch_parser.py271 assert "def helper():" in file_ops.writtenCODE
LOWtests/hermes_cli/test_gateway_restart_loop.py1487 def helper(self, tmp_path):CODE
LOWtests/hermes_cli/test_gateway_restart_loop.py1543 def helper(self, tmp_path):CODE
LOWtui_gateway/server.py2926def handle_request(req: dict) -> dict | None:CODE
LOWscripts/sandbox/proxy.py200def handle_request(conn):CODE
TODO Padding1 hit · 2 pts
SeverityFileLineSnippetContext
LOW…darwinian-evolver/templates/custom_problem_template.py80 # TODO: implement. For prompt evolution this typically calls _prompt_llmCOMMENT