MCPcopy Create free account

hub / github.com/NVIDIA/SkillEvaluator / functions

Functions24,741 in github.com/NVIDIA/SkillEvaluator

↓ 7 callersFunction_action_text
(tool_call: dict[str, Any])
src/skillevaluator/tier3/eval_core/checks.py:883
↓ 7 callersFunction_activation
Reward details recording whether the with-skill agent activated the skill.
tests/tier3/test_baseline_timeouts.py:102
↓ 7 callersFunction_agent_eval_truth_state
Return conservative Tier 3 truth plus whether duplicated claims agree.
src/skillevaluator/reporting/base.py:659
↓ 7 callersFunction_agents
()
tests/evaluation/test_tier3_benchmark_contract.py:39
↓ 7 callersFunction_assert_cursor_shown_and_one_stop_notice
(output: str, notice: str)
tests/test_tier3_progress.py:1197
↓ 7 callersFunction_bounded_json_text
(value: Any, limit: int)
src/skillevaluator/tier3/eval_core/plugin_signals.py:846
↓ 7 callersFunction_build_agent
( name: str, info: dict[str, Any], with_metrics: list[str], baseline_metrics: list[str], m
src/skillevaluator/evaluation/tier3_report.py:2354
↓ 7 callersFunction_by_name
(results: list[ValidationResult])
tests/test_artifact_applicability.py:62
↓ 7 callersFunction_cases_run
(tmp_path: Path, agents: int, cases: int)
tests/reporting/test_tier3_token_usage_reports.py:520
↓ 7 callersFunction_claude_tool_result
(*, name: str, arguments: dict, is_error: bool, content: str)
tests/tier3/test_failed_skill_reads.py:27
↓ 7 callersFunction_codex_items
(item_id: str, server: str, tool: str, args: dict[str, Any], text: str, status: str)
tests/tier3/test_f90_mcp_execution.py:163
↓ 7 callersFunction_collect
(root: Path, servers: dict[str, Any], files: dict[str, str] | None = None)
tests/validators/test_openclaw_mcp.py:28
↓ 7 callersFunction_collect
(root: Path, mcp_servers: Any, files: dict[str, Any] | None = None)
tests/validators/test_plugin_mcp.py:20
↓ 7 callersFunction_collect_all_skill_deps
Collect pip and apt deps from target skill + staged workspace skills.
src/skillevaluator/tier3/harbor/adapter.py:1442
↓ 7 callersFunction_comparison
(**updates)
tests/reporting/test_native_overview_comparison.py:21
↓ 7 callersMethod_compose_model_interpolation_names
(self)
src/skillevaluator/tier3/harbor/secure_docker_environment.py:2213
↓ 7 callersFunction_condition_status
(data: Mapping[str, Any], variant: str)
src/skillevaluator/tier3/result_display.py:117
↓ 7 callersFunction_copy_trial_artifacts
( trial_dir: Path, trial_out: Path, *, include_root_trajectory: bool = True, )
src/skillevaluator/tier3/harbor/collector.py:1369
↓ 7 callersFunction_credential_uri_environment_values
Return operational URI values containing authority user information.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:263
↓ 7 callersFunction_dataset
(tmp_path: Path)
tests/test_harbor_runtime_preflight.py:68
↓ 7 callersFunction_destination_metadata
(parent_descriptor: int, name: str)
src/skillevaluator/tier3/harbor/secure_copy.py:838
↓ 7 callersFunction_details_truncated
(agent: Mapping[str, Any])
src/skillevaluator/tier3/case_summary.py:88
↓ 7 callersFunction_detect_from_path_parts
Detect content type from folder path patterns.
src/skillevaluator/cli_core.py:70
↓ 7 callersFunction_diagnostic
(case, reward, baseline=None)
tests/reporting/test_tier3_expectations.py:31
↓ 7 callersFunction_dimension_row
(html: str, label: str)
tests/reporting/test_dimension_explanations.py:81
↓ 7 callersFunction_engine
(by_server: dict[str, dict[str, int]], census: dict[str, Any] | None = None)
tests/tier3/test_f90_mcp_proof_status.py:21
↓ 7 callersFunction_entry
A legacy entry (question, ground_truth, expected_behavior).
tests/tier3/test_dataset_baseline_fairness.py:35
↓ 7 callersFunction_entry_errors
(count: int, message: str = _MESSAGE)
tests/reporting/test_tier3_source_validation_reporting.py:80
↓ 7 callersFunction_extract_json
Extract JSON from LLM response, handling code fences and truncation.
src/skillevaluator/validators/rubric_eval.py:424
↓ 7 callersFunction_finding
(severity, message)
tests/ci/test_summary_metrics.py:13
↓ 7 callersFunction_finding_types
(result: dict)
tests/tier3/test_issue192_security_false_positives.py:69
↓ 7 callersFunction_fingerprint
(metadata: os.stat_result)
src/skillevaluator/publication_identity.py:94
↓ 7 callersFunction_git
(workspace: Path, *arguments: str)
ci/select_changed_skills.py:101
↓ 7 callersFunction_http_hook
(url: str, **extra: object)
tests/validators/test_plugin_component_risk.py:565
↓ 7 callersFunction_in_skill_output
(skill: Path, name: str)
tests/test_harbor_output_provenance.py:57
↓ 7 callersFunction_interpreter_operands
The interpreter's own arguments, with every redirection removed. ``python3 < /dev/null run.py`` runs run.py: the redirection belongs to the s
src/skillevaluator/tier3/eval_core/script_invocation.py:2474
↓ 7 callersFunction_is_heredoc_redirect
(token: str)
src/skillevaluator/tier3/eval_core/script_invocation.py:55
↓ 7 callersFunction_is_skipped
True when a validator was recorded without executing (OSS conventions).
src/skillevaluator/reporting/console_ui.py:955
↓ 7 callersFunction_is_within
True if ``path`` is ``root`` or a descendant of it (both pre-resolved).
src/skillevaluator/tier3/harbor/local_sandbox.py:356
↓ 7 callersFunction_job
(path: Path, score: float, *, grading_mode: str = "default")
tests/test_baseline_store.py:28
↓ 7 callersFunction_job
()
tests/ci/test_evaluate_workflow.py:20
↓ 7 callersFunction_kinds
(steps: list[dict])
tests/test_tier3_next_steps.py:42
↓ 7 callersFunction_load_bounded_json
( path: Path, diagnostics: list[dict[str, Any]], *, artifact: str, )
src/skillevaluator/tier3/harbor/report_data.py:420
↓ 7 callersFunction_manifest
(selected: list[Path])
tests/tier3/test_decoy_workspace.py:33
↓ 7 callersFunction_mapping
(value: object)
src/skillevaluator/reporting/plugin_static.py:34
↓ 7 callersFunction_mapping
(value: object)
src/skillevaluator/tier3/token_usage.py:146
↓ 7 callersFunction_markdown_tokens
Parse Markdown once so structural checks use rendered block semantics.
scripts/ci/check_public_benchmarks.py:1273
↓ 7 callersFunction_md_cell
(value: object, private_labels: tuple[str, ...] = ())
src/skillevaluator/reporting/benchmark.py:2170
↓ 7 callersFunction_metric_evidence
( metric: str, rewards: list[dict[str, Any]], report_budget: _ReportBudget, sampling: dict[str
src/skillevaluator/evaluation/tier3_report.py:3031
↓ 7 callersFunction_native
(command: str, shell: object = None)
tests/tier3/test_script_invocation_evidence.py:2554
↓ 7 callersFunction_network_findings
(result: dict)
tests/tier3/test_network_client_mentions.py:66
↓ 7 callersFunction_node_identity
(metadata: os.stat_result)
src/skillevaluator/tier3/harbor/secure_copy.py:229
↓ 7 callersFunction_nonnegative_count
(value: object)
src/skillevaluator/reporting/html.py:208
↓ 7 callersFunction_normalized_write_path
Return *target* as a lowercase absolute path, resolved lexically as POSIX does. A home shorthand becomes "/home/~", a Windows drive path keeps it
src/skillevaluator/tier3/eval_core/checks.py:1389
↓ 7 callersFunction_numbered_sections
Return Markdown that chunks into exactly ``count`` sections.
tests/deduplication/intra_skill/test_intra_skill_validator.py:30
↓ 7 callersFunction_overall_score_for_display
Honor persisted score truth before inferring a policy from metric names.
src/skillevaluator/tier3/commands.py:1823
↓ 7 callersFunction_pass_to_reference
What is done to a ``-n`` name is done to the name it refers to. That name's value is unsettled, and it is marked exported where the change ex
src/skillevaluator/tier3/eval_core/script_invocation.py:953
↓ 7 callersMethod_path_is_within_allowed_local_roots
(self, path: Path)
src/skillevaluator/tier3/harbor/local_environment.py:1812
↓ 7 callersFunction_payload
(*, checks=True)
tests/reporting/test_declarative_coverage.py:114
↓ 7 callersMethod_place
( self, piece_id: str, kind: str, source: str | None, *, full:
src/skillevaluator/tier3/loose_parts.py:807
↓ 7 callersFunction_positive
(value)
ci/feedback.py:92
↓ 7 callersFunction_process
(report: dict, source_root: Path | None = None)
tests/validators/test_skillspector_reference_diagnostics.py:84
↓ 7 callersFunction_provider
(name: str, *, api_key: str = "k", base_url: str | None = None)
tests/test_harbor_local_mode.py:216
↓ 7 callersFunction_ratio
(numerator: float, denominator: float)
src/skillevaluator/tier3/eval_core/plugin_signals.py:3570
↓ 7 callersFunction_read_regular_evals_file
Read one bounded evaluator file without following links or accepting swaps.
src/skillevaluator/tier3/harbor/adapter.py:1823
↓ 7 callersFunction_record_truncation
Record one bounded, content-free diagnostic and emit it once per scope.
src/skillevaluator/tier3/harbor/report_data.py:266
↓ 7 callersFunction_referenced_image_report
Load a captured SkillSpector ``--no-llm`` report for a skill that references images. The scanned skill is ``SKILL.md`` plus the files written by
tests/validators/test_security.py:8894
↓ 7 callersFunction_refs
(values: Sequence[str])
src/skillevaluator/tier3/eval_core/plugin_signals.py:3492
↓ 7 callersFunction_remote_state
(monkeypatch, info, *, actions=None, pages=None)
tests/test_ci_onboarding.py:424
↓ 7 callersFunction_replay
(monkeypatch, replies)
tests/tier3/test_judge_parse_robustness.py:793
↓ 7 callersFunction_require
(condition: bool, message: str)
src/skillevaluator/reporting/retained.py:32
↓ 7 callersFunction_resolve_entry_file_ref
( ref: str, *, skill_path: Path, evals_dir: Path, input_files_dir: Path | None, )
src/skillevaluator/tier3/harbor/adapter.py:4074
↓ 7 callersFunction_resolve_runtime_env
( templates: dict[str, str] | None, *, env_mode: str, )
src/skillevaluator/tier3/harbor/runner.py:2860
↓ 7 callersFunction_result
(finding: Finding)
tests/reporting/test_finding_verification.py:56
↓ 7 callersFunction_reward
(attempt: int, *, goal_accuracy: float | None)
tests/test_tier3_attempt_verdict.py:20
↓ 7 callersFunction_reward
(*, baseline=False)
tests/reporting/test_declarative_coverage.py:59
↓ 7 callersFunction_rollout
(tmp_path: Path, task: str, attempt: int, overall: float | None)
tests/test_nemo_gym_backend.py:133
↓ 7 callersMethod_runtime_ro_binds
Read-only mounts the sandbox needs: managed agent CLIs + evaluator python. Strict mode publishes the visible and canonical interpreter files
src/skillevaluator/tier3/harbor/local_environment.py:1588
↓ 7 callersFunction_safe_harbor_viewer_url
(value: object)
src/skillevaluator/evaluation/tier3_report.py:3935
↓ 7 callersFunction_save_trials
Save per-trial reward.json and trajectory.json into the results directory.
src/skillevaluator/tier3/harbor/collector.py:6521
↓ 7 callersFunction_score
(module, command: str, *, workdir: str | None = None, agent_workdir: str | None = "/workspace")
tests/tier3/test_issue192_security_false_positives.py:65
↓ 7 callersFunction_security_reward
(score: float)
tests/test_tier3_attempt_verdict.py:43
↓ 7 callersFunction_shared_metric_pass_lift
Compare two arms' pass@k on the metrics both arms scored. An arm without the skill under test records ``skill_execution`` and ``skill_efficie
src/skillevaluator/tier3/harbor/collector.py:5409
↓ 7 callersFunction_short_name
(validator_name: str)
src/skillevaluator/reporting/console_ui.py:1133
↓ 7 callersFunction_split_words
Shell words of ``text`` (quotes removed); whitespace split when the quoting is unbalanced.
src/skillevaluator/validators/mcp_static.py:1807
↓ 7 callersFunction_string_list
(value: object)
src/skillevaluator/reporting/html.py:432
↓ 7 callersFunction_summary
(rewards: list[dict])
tests/test_tier3_attempt_verdict.py:33
↓ 7 callersFunction_symlink_or_skip
(link: Path, target: Path, *, target_is_directory: bool = False)
tests/test_harbor_adapter_safety.py:229
↓ 7 callersMethod_terminate_process_tree
( proc: asyncio.subprocess.Process, communication: asyncio.Task[tuple[bytes, bytes]] | None =
src/skillevaluator/tier3/harbor/local_environment.py:1401
↓ 7 callersFunction_tier3_html
(*, agent_count: int = 1)
tests/reporting/test_chart_dependency.py:50
↓ 7 callersFunction_traj
(command: str)
tests/tier3/test_network_client_mentions.py:43
↓ 7 callersFunction_tree
(root: Path, files: dict[str, Any])
tests/test_openclaw_plugin_discovery.py:32
↓ 7 callersFunction_trial_records
(trials: Path, name: str, records: dict[str, dict])
tests/test_tier3_case_summary.py:1088
↓ 7 callersFunction_valid_step_result_names
Return structurally authoritative Harbor step names, or ``None``.
src/skillevaluator/tier3/harbor/collector.py:2820
↓ 7 callersFunction_validate
(*args: str)
tests/test_artifact_applicability.py:81
↓ 7 callersFunction_validate_environment
Validate environment names and values without rendering secret values.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:539
↓ 7 callersMethod_validate_skillspector_report
Return whether JSON is a trustworthy SkillSpector findings report. Suppressed findings are accepted only when SkillEvaluator passed an
src/skillevaluator/validators/security.py:3609
↓ 7 callersFunction_validated_trajectory_dict
(data: Any)
src/skillevaluator/tier3/harbor/collector.py:1983
↓ 7 callersFunction_windows_close_handle
(handle: int)
src/skillevaluator/utils/secure_fs.py:753
↓ 7 callersFunction_windows_handle_metadata
(handle: int)
src/skillevaluator/utils/secure_fs.py:764
← previousnext →1,201–1,300 of 24,741, ranked by callers