MCPcopy Create free account

hub / github.com/NVIDIA/SkillEvaluator / functions

Functions24,741 in github.com/NVIDIA/SkillEvaluator

Methodis_available
(self)
tests/validators/test_private_claude_parity.py:26
Methodis_complete
Check if workflow has all required components.
src/skillevaluator/models/workflows.py:318
Functionis_container_revision
Return whether the value names the evaluator build that produced a card. The path length is checked in Python rather than in the pattern because
src/skillevaluator/source_identity.py:81
Methodis_content_read
A skill load or a plain file read: its result is file or skill text, not a status message.
src/skillevaluator/tier3/eval_core/plugin_signals.py:1689
Methodis_directory
(self)
src/skillevaluator/utils/secure_fs.py:100
Methodis_incomplete
Whether required validation evidence is missing or unusable.
src/skillevaluator/models/result.py:442
Functionis_nonpublic
Treat missing, inconsistent, and non-boolean privacy data conservatively.
src/skillevaluator/_ci_onboarding_pages.py:19
Methodis_open
(self)
tests/test_harbor_gke_environment.py:929
Methodis_plain_directory
A directory that is not itself a reparse point (a junction or directory symlink is one).
src/skillevaluator/utils/secure_fs.py:104
Methodis_reparse
(self)
src/skillevaluator/utils/secure_fs.py:96
Methodis_shell
A local shell call (an MCP tool named like a shell is still an MCP call).
src/skillevaluator/tier3/eval_core/plugin_signals.py:1684
Functionis_tier2_validator_name
Compatibility wrapper for callers that imported this helper here.
src/skillevaluator/reporting/html.py:181
Methodis_trusted_path
(path: Path)
src/skillevaluator/tier3/harbor/secure_docker_environment.py:2078
Functionis_within
Return whether ``path`` resolves inside ``root``.
src/skillevaluator/plugin_dependencies.py:341
Methodisatty
(self)
tests/test_tier3_progress.py:36
Methodisatty
(self)
tests/test_tier3_progress.py:1732
Methodisatty
(self)
tests/test_tier3_progress.py:2100
Methodisatty
(self)
tests/reporting/test_tier3_validate_cases.py:23
Methodisatty
(self)
tests/reporting/test_terminal_polish.py:178
Functionisolate_scanner_preparation
(monkeypatch)
tests/cli/test_tier_workflows.py:46
Functionisolated_provider_environment
Use synthetic credentials and refuse any network address outside loopback.
tests/test_retry_provider_transport.py:27
Methodissues
(name: str)
tests/validators/test_skillspector_hook_surface.py:452
Methodissues_count
(self)
src/skillevaluator/models/quality.py:47
Methoditems
(self)
tests/reporting/test_unified_tier3_report.py:1947
Functioniter_normalized_tool_calls
Yield each trajectory tool call after safe native-Codex normalization.
src/skillevaluator/tier3/eval_core/codex_tool_call_normalizer.py:70
Methodjoinpath
(self, _path: str)
tests/reporting/test_unified_tier3_report.py:710
Methodjpeg_of
(*parts: bytes)
tests/validators/test_security.py:9227
Functionjson_safe_text_spy
(value: str)
tests/reporting/test_advisory_skip_verdict.py:1657
Functionjudge
(*_args, **_kwargs)
tests/tier3/test_security_workdir_provenance.py:121
Functionjudge
()
tests/tier3/test_judge_failure_artifacts.py:929
Functionjudge_accuracy
Judge accuracy from five outcome verdicts; the score is computed here. *skill_identification_applicable* is deprecated and ignored. Accuracy no
src/skillevaluator/tier3/eval_core/llm_judge.py:763
Functionjudge_accuracy
Judge accuracy from five outcome verdicts; the score is computed here. *skill_identification_applicable* is deprecated and ignored. Accuracy no
src/skillevaluator/tier3/harbor/templates/eval.py:3336
Functionjudge_behavior_check
Run the behavior check LLM judge; the score is the share of behaviors observed. *evidence_omitted* is the behavior_check bundle's omission accoun
src/skillevaluator/tier3/eval_core/llm_judge.py:1049
Functionjudge_behavior_check
Judge expected behaviors, or refuse when required evidence did not reach the judge. *evidence_omitted* is the behavior_check bundle's omission ac
src/skillevaluator/tier3/harbor/templates/eval.py:3741
Functionjudge_env
(monkeypatch)
tests/tier3/test_judge_parse_robustness.py:767
Functionjudge_goal_accuracy
Run the goal accuracy judge (two-step: infer goal, compare outcome); the score is binary.
src/skillevaluator/tier3/eval_core/llm_judge.py:847
Functionjudge_goal_accuracy
(question, ground_truth, agent_text, tool_summary="")
src/skillevaluator/tier3/harbor/templates/eval.py:3414
Functionjudge_module
(request: pytest.FixtureRequest)
tests/tier3/test_judge_contracts.py:56
Functionjudge_module
(request: pytest.FixtureRequest)
tests/tier3/test_judge_result_semantics.py:211
Functionjudge_server
(monkeypatch)
tests/tier3/test_grader_validation.py:498
Functionkeep_payload
(payload: dict[str, Any], *args: Any, **kwargs: Any)
tests/reporting/test_tier3_token_usage_reports.py:633
Functionkeep_staged_tasks
Keep the staged task folders, which a Gym run removes at the end, for inspection.
tests/test_nemo_gym_backend.py:29
Functionkey
(path: Path)
src/skillevaluator/tier3/history.py:49
Methodkill
(self)
tests/test_harbor_secure_docker_environment.py:142
Methodkill
(self)
tests/test_harbor_secure_docker_environment.py:240
Methodkill
(self)
tests/test_harbor_secure_docker_environment.py:266
Methodkill
(self)
tests/test_harbor_runner_status.py:846
Functionkillpg
(pid: int, value: signal.Signals)
tests/test_harbor_secure_docker_environment.py:6918
Functionkillpg
(_pid: int, value: signal.Signals)
tests/test_harbor_local_mode.py:4088
Functionks
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
Methodl
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
Functionlabel
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
FunctionlabelColor
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
FunctionlabelPointStyle
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
FunctionlabelTextColor
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
Methodlast
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
Functionlate_agent_update
()
tests/test_tier3_progress.py:2461
Methodlatest_attempt_coverage
Last directory selection for diagnostics, including a failed build.
src/skillevaluator/embedding/registry.py:226
Functionlaunch
(**kwargs)
tests/test_private_runtime_review.py:35
Functionlaunch
(**kwargs)
tests/test_version_comparison.py:64
Functionlaunch
(**kwargs)
tests/test_harbor_timeout_recovery.py:275
Functionlaunch
(**kwargs)
tests/test_baseline_store.py:436
Functionlaunch
(*_args: object, **_kwargs: object)
tests/test_harbor_runner_status.py:617
Functionlaunch
(**kwargs)
tests/test_harbor_runner_environment.py:1203
Functionlaunch
(tmp_path, monkeypatch)
tests/ci/test_run_tier1_image.py:41
Functionlaunch
(**kwargs)
tests/tier3/test_private_comparison_engine.py:45
Functionlaunch
(**kwargs)
tests/tier3/test_private_runtime_report_acceptance.py:142
Functionlaunch
(**kwargs)
tests/tier3/test_private_comparison_engine_acceptance.py:197
Methodlaunch_env
(self)
src/skillevaluator/tier3/plugin_native.py:1995
Functionleader_only_kill
(pgid: int, sig: int)
tests/test_nemo_gym_backend.py:2659
Methodlegacy_filters_must_be_empty
Tolerate empty ``include``/``exclude`` as no-ops; reject populated ones.
src/skillevaluator/models/plugin.py:143
Methodlegacy_score
(candidate_issues: list[dict])
src/skillevaluator/validators/security.py:4746
Functionlegacy_set_cdata_mode
(_parser, element: str)
tests/validators/test_markdown.py:100
Methodlighten
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
MethodlinkScales
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
Functionlint
(_path: Path)
tests/reporting/test_junit_reporter.py:630
Functionlint_scripts
Run advisory lint checks on skill scripts.
src/skillevaluator/cli.py:3350
Methodlist_foundation_models
(self)
tests/test_harbor_runtime_preflight.py:1335
Methodlist_tasks
List all tasks grouped by status.
src/skillevaluator/tier3/reference_skills/task-list/scripts/tasks.py:243
Functionlive_on_a_refusing_terminal
(*args: Any, **kwargs: Any)
tests/test_tier3_progress.py:1507
Functionllm_calls
Record every judge and what-to-fix LLM call; each one answers with fresh text.
tests/tier3/test_view_results_feedback.py:98
Functionllm_prompts
Stub the public provider and capture every dataset prompt it is sent.
tests/tier3/test_dataset_baseline_fairness.py:507
Functionllm_reply
Stub the public provider; the test appends the reply it returns.
tests/tier3/test_dataset_baseline_fairness.py:1293
Functionload
(_target, path)
tests/tier3/test_decoy_workspace.py:68
Functionload_dataset
Load the first supported Tier 3 dataset from a skill directory.
src/skillevaluator/tier3/harbor/report_data.py:1265
Functionload_directory
(*_args, **kwargs)
tests/evaluation/test_tier3_benchmark_contract.py:485
Functionload_incluster_config
()
tests/test_harbor_local_mode.py:5816
Functionload_trajectory_with_fallback
(trajectory_path, logs_dir=None)
src/skillevaluator/tier3/harbor/templates/eval.py:62
Functionlocal_connect
(sock, address)
tests/test_retry_provider_transport.py:62
Functionlocal_connect
(sock: socket.socket, address: Any)
tests/tier3/test_ragas_judge_evidence.py:99
Functionlocal_getaddrinfo
(host, *args, **kwargs)
tests/test_retry_provider_transport.py:58
Functionlocal_getaddrinfo
(host: Any, *args: Any, **kwargs: Any)
tests/tier3/test_ragas_judge_evidence.py:93
Functionlocal_output_environment
(tmp_path: Path)
tests/test_harbor_local_environment.py:31
Methodlocation
Return formatted location like 'file.py:42'.
src/skillevaluator/models/result.py:234
Methodlog_message
Prevent the base handler from emitting unreviewed request diagnostics.
src/skillevaluator/tier3/harbor/nvidia_build_bridge.py:314
Methodlog_message
(self, _format: str, *_args: object)
tests/test_harbor_gke_environment.py:64
Methodlog_message
(self, *_args)
tests/test_nemo_gym_backend.py:1740
Methodlog_message
(self, _format, *_args)
tests/test_harbor_runtime_preflight.py:1630
Methodlog_message
(self, _format, *_args)
tests/test_harbor_runtime_preflight.py:1873
Methodlog_message
(self, _format: str, *_args: object)
tests/test_harbor_runtime_preflight.py:3424
← previousnext →11,201–11,300 of 24,741, ranked by callers