MCPcopy Create free account

hub / github.com/NVIDIA/SkillEvaluator / functions

Functions24,741 in github.com/NVIDIA/SkillEvaluator

↓ 3 callersFunction_dataset_result
A Tier 3 result with a dataset, so the report renders the dataset header too.
tests/reporting/test_execution_backend_reporting.py:93
↓ 3 callersMethod_declared_replaces
Whether a declared field suppresses default-folder discovery for its type. Codex drops a path it does not accept and loads the default locati
src/skillevaluator/plugin_discovery.py:310
↓ 3 callersMethod_declared_skill_dirs
Root-relative skill directories declared by the projected Claude Code manifest.
src/skillevaluator/tier3/plugin_native.py:1444
↓ 3 callersFunction_decoy_workspace_skills
Resolve one frozen selection without changing manual workspace policy.
src/skillevaluator/tier3/harbor/runner.py:3452
↓ 3 callersMethod_deduplicate_skillspector_findings
Deduplicate findings repeated by static and enriched reports.
src/skillevaluator/validators/security.py:3552
↓ 3 callersFunction_default_attempt_policy
()
src/skillevaluator/evaluation/tier3_report.py:5060
↓ 3 callersFunction_delta_cell
Lift cell. With no delta it reads N/A for a row that is N/A in either arm, then ``skipped`` when the with-skill arm never ran, then NO SCORE, like
src/skillevaluator/tier3/result_display.py:99
↓ 3 callersFunction_dimension_cell
Render one Results at a Glance dimension cell, explaining a missing score.
src/skillevaluator/reporting/benchmark.py:1832
↓ 3 callersFunction_dimension_definition
(html: str, label: str)
tests/reporting/test_dimension_explanations.py:88
↓ 3 callersFunction_dimension_score
(agent: Mapping[str, Any], key: str, name: str)
src/skillevaluator/tier3/case_summary.py:592
↓ 3 callersFunction_display_metadata_id
Render a bounded preview of a ``metadata.id`` value for a finding. A non-string value is described by its type alone and never rendered. ``me
src/skillevaluator/validators/schema.py:174
↓ 3 callersFunction_dockerfile_final_explicit_user
Return the statically known USER inherited by the final stage.
src/skillevaluator/tier3/harbor/adapter.py:2609
↓ 3 callersFunction_dockerfile_logical_lines_from_content
Return Dockerfile instructions after applying its escape directive.
src/skillevaluator/tier3/harbor/adapter.py:2564
↓ 3 callersMethod_drop_refused_writes
Discard the output this thread's Rich console holds from refused writes, and return it. Rich keeps a refused write for its thread and sends i
src/skillevaluator/tier3/harbor/progress.py:630
↓ 3 callersFunction_edge_result
Synthetic retained shapes shared with missing-arm/telemetry unit coverage.
tests/reporting/test_report_browser.py:64
↓ 3 callersFunction_effective_reasoning_effort
(provider, model, effort)
src/skillevaluator/tier3/harbor/templates/eval.py:1296
↓ 3 callersFunction_eligibility_outputs
Run the actual eligibility script for an eligible same-repository PR.
tests/ci/test_github_admission_workflow.py:146
↓ 3 callersFunction_embedded_ipv4
Return an IPv4 address embedded in an IPv6 literal (mapped/6to4/Teredo/NAT64/compat).
src/skillevaluator/validators/mcp_static.py:2861
↓ 3 callersMethod_empty
(self, status: int)
tests/tier3/test_mcp_proof.py:127
↓ 3 callersFunction_engine
(monkeypatch, root: Path, *, arm_count: int, route: str | None, keep: bool, ordinary: bool = False)
tests/tier3/test_private_comparison_engine_acceptance.py:166
↓ 3 callersFunction_ensure_empty_custom_docker_input_compatibility
Keep directory-level ``COPY input/`` valid for the managed main build.
src/skillevaluator/tier3/harbor/adapter.py:3070
↓ 3 callersFunction_entry
(status: str, detail: str, tools: Iterable[str] = ())
src/skillevaluator/tier3/mcp_proof.py:219
↓ 3 callersFunction_enum
(value: Any, allowed: set[str], config_path: Path, field: str)
src/skillevaluator/tier3/evals_config.py:384
↓ 3 callersFunction_environment_kwarg_policy_error
(env_mode: str, environment_kwargs: Mapping[str, Any])
src/skillevaluator/tier3/harbor/runner.py:1742
↓ 3 callersFunction_escaped_twice
(text: str)
tests/tier3/test_judge_contracts.py:1493
↓ 3 callersFunction_evaluate_tier2_with_unselected_game
Run Tier 2 with only alpha selected and an unselected ``skills/game`` carrying *frontmatter*.
tests/ci/test_run_tiers.py:1010
↓ 3 callersFunction_evaluated_source_from_options
Build the evaluated-source identity from the orchestration input. The identity is supplied rather than inferred, because the tree running the
src/skillevaluator/cli.py:889
↓ 3 callersFunction_evaluation_failed
(reward: Any)
src/skillevaluator/tier3/case_summary.py:261
↓ 3 callersFunction_evaluator_node_fingerprint
(metadata: os.stat_result)
src/skillevaluator/tier3/harbor/adapter.py:1769
↓ 3 callersFunction_evaluator_tests_dir
(task_dir: Path)
src/skillevaluator/tier3/harbor/adapter.py:2409
↓ 3 callersFunction_exec
(cmd: str)
tests/test_checks_routing.py:56
↓ 3 callersMethod_executable_files
(skill_path: Path, patterns: tuple[str, ...] = EXECUTABLE_SCRIPT_PATTERNS)
src/skillevaluator/validators/quality_score.py:274
↓ 3 callersFunction_expected_observed_suffix
(detail: dict[str, Any], reward: dict[str, Any] | None = None)
src/skillevaluator/tier3/harbor/report.py:655
↓ 3 callersFunction_expected_summary
(status: str)
tests/reporting/test_dataset_sufficiency_rendering.py:17
↓ 3 callersFunction_extract_eval_hints
Extract structured hints from developer eval guidance for template-based generation. Looks for markdown sections like ## Questions, ## Behaviors,
src/skillevaluator/tier3/generate_dataset.py:401
↓ 3 callersFunction_extract_findings
Extract actionable findings from reward.json details across trials.
src/skillevaluator/tier3/harbor/report.py:288
↓ 3 callersFunction_extract_pod_missing_exec_detail
Return a diagnostic message if an exception chain indicates the target pod no longer exists.
src/skillevaluator/tier3/harbor/gke_environment.py:265
↓ 3 callersFunction_extract_trial_failures
(job_dir: Path)
src/skillevaluator/tier3/harbor/collector.py:1169
↓ 3 callersFunction_extreme_result
(scores)
tests/reporting/test_native_feature_restoration.py:228
↓ 3 callersFunction_fail_tier3_commands_import
(monkeypatch, *, missing: str)
tests/test_cli.py:725
↓ 3 callersFunction_failed_judge_diagnostic
Project failed judge sidecars into one safe, intrinsically unscoreable record.
src/skillevaluator/tier3/harbor/collector.py:1074
↓ 3 callersFunction_failed_single_skill_read
A shell failure only proves the skill read failed for an isolated read. Compound commands, multi-file reads and operands that the shell may expan
src/skillevaluator/tier3/eval_core/checks.py:1731
↓ 3 callersFunction_failed_trial
( job: Path, name: str, case: str, *, exception_type: str = "AgentTimeoutError", outpu
tests/tier3/test_baseline_timeouts.py:120
↓ 3 callersFunction_file_changes_section
(evidence: str)
tests/tier3/test_behavior_evidence.py:1456
↓ 3 callersFunction_file_key
``(normalized path, base name)`` of a file argument; ``None`` for stdout, devices, and empty names. Simple variables the same text assigns are ex
src/skillevaluator/plugin_component_risk.py:2509
↓ 3 callersMethod_filter_command_env
Reject process-control values and drop protected path overrides.
src/skillevaluator/tier3/harbor/local_environment.py:1912
↓ 3 callersFunction_finalize_harbor_artifacts
Finalize transient paths and persist corrected metadata when available.
src/skillevaluator/tier3/harbor/runner.py:7027
↓ 3 callersFunction_findings_eligible
Return whether persisted execution truth permits quality findings.
src/skillevaluator/tier3/harbor/report.py:195
↓ 3 callersFunction_finish_pipeline_view
Render the quiet-mode verdict panel and report footer.
src/skillevaluator/cli.py:1972
↓ 3 callersFunction_finite_number
(value: Any)
src/skillevaluator/evaluation/tier3_feedback_chat.py:233
↓ 3 callersFunction_finite_numeric
(value: object)
src/skillevaluator/tier3/commands.py:1739
↓ 3 callersFunction_finite_reward_number
(value)
src/skillevaluator/tier3/harbor/templates/eval.py:3924
↓ 3 callersFunction_first_publication_safe_label
( values: tuple[object, ...], private_labels: tuple[str, ...] = (), )
src/skillevaluator/reporting/benchmark.py:1496
↓ 3 callersFunction_follows_ansi_code
Whether an ANSI code (``\\u001b[0m``, ``\\x1b[K`` or a raw one) ends right before ``index``.
src/skillevaluator/evaluation/tier3_feedback_chat.py:427
↓ 3 callersMethod_footer_lines
(self)
src/skillevaluator/reporting/console_ui.py:855
↓ 3 callersFunction_format_http_error_with_fallback
(error)
src/skillevaluator/tier3/harbor/templates/eval.py:1230
↓ 3 callersFunction_fresh_python
( site: Path, code: str, *args: str, extra_env: dict[str, str] | None = None )
tests/test_custom_backend_package.py:144
↓ 3 callersMethod_from_payload
(cls, raw, *, client, max_scalar_comparisons)
src/skillevaluator/deduplication/plugin/catalog.py:83
↓ 3 callersFunction_fsync_directory
Durably publish a directory entry where the platform supports it.
src/skillevaluator/tier3/output_provenance.py:73
↓ 3 callersFunction_function_header_end
Return the index after a function header at ``idx``, or ``None``. ``f()``, ``f ()``, ``function f`` and ``function f()`` each start a functio
src/skillevaluator/tier3/eval_core/checks.py:2552
↓ 3 callersFunction_gated
(results: list[ValidationResult], tier: int, *, blocking: bool = True)
tests/reporting/test_report_redesign.py:262
↓ 3 callersFunction_generate_simple
Generate 1 simple test case (default, no LLM needed).
src/skillevaluator/tier3/generate_dataset.py:581
↓ 3 callersMethod_getAnims
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 3 callersMethod_getSortedDatasetMetas
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersMethod_get_core_api
Return the initialized Kubernetes CoreV1Api client without raising on uninitialized property access.
src/skillevaluator/tier3/harbor/gke_environment.py:787
↓ 3 callersMethod_get_env
Expose the sentinel key while suppressing Codex's legacy base-url config.
src/skillevaluator/tier3/harbor/local_agents.py:986
↓ 3 callersFunction_gitlab_origin
(api_url)
ci/feedback.py:542
↓ 3 callersFunction_graded_names_a_skill_script
Whether graded text names review.py or review.sh, as written or as `tier3 validate` normalizes it.
tests/tier3/test_dataset_baseline_fairness.py:969
↓ 3 callersFunction_harbor_bin
Return the Harbor executable installed with the active interpreter.
src/skillevaluator/tier3/harbor/runner.py:1445
↓ 3 callersFunction_harbor_result
Real Harbor model shapes, deterministic complete rewards, no agent/provider.
tests/tier3/test_private_comparison_engine_acceptance.py:48
↓ 3 callersFunction_harbor_task_directory
Return the task directory a Harbor trial ran, from its copied ``result.json`` or ``config.json``. A generated task's directory is named after its
src/skillevaluator/tier3/harbor/report_data.py:755
↓ 3 callersFunction_has_authored_ground_truth
Only null and blank strings leave an authored expected outcome unspecified.
src/skillevaluator/tier3/generate_dataset.py:1291
↓ 3 callersFunction_has_symlink_component
(path: Path, root: Path)
src/skillevaluator/tier3/harbor/adapter.py:2446
↓ 3 callersFunction_hash_descriptor
(descriptor: int)
src/skillevaluator/publication_identity.py:274
↓ 3 callersFunction_heading_entries
Return token index, level, line, title, and structural trust.
scripts/ci/check_public_benchmarks.py:1279
↓ 3 callersFunction_high_checks
(result: ValidationResult)
tests/validators/test_security.py:11200
↓ 3 callersMethod_hook
(self, source: _JsonSource, *, profile: FormatProfile | None = None)
src/skillevaluator/plugin_discovery.py:1377
↓ 3 callersFunction_host_text
(host: Any)
tests/tier3/test_ragas_judge_evidence.py:60
↓ 3 callersFunction_html_element_is_visually_hidden
Return whether an HTML element is removed from visual rendering.
scripts/ci/check_public_benchmarks.py:484
↓ 3 callersFunction_html_structure
( content: str, )
scripts/ci/check_public_benchmarks.py:726
↓ 3 callersFunction_http_error_body
Return a failed response's body text, reading it only once.
src/skillevaluator/tier3/harbor/templates/eval.py:1216
↓ 3 callersFunction_http_hook_result
(tmp_path: Path, url: str, policy: ValidationPolicy | None = None)
tests/validators/test_plugin_component_risk.py:1139
↓ 3 callersFunction_human_reasoning_bullets
Render 2-3 short, data-focused reasoning bullets for one dimension. Bullet 1 -- score + verdict relative to the configured PASS / NEUTRAL / F
src/skillevaluator/evaluation/dimension_judge.py:344
↓ 3 callersMethod_in_code
The phrase that ends at ``end`` is followed by "in parser.py", "from the script", ...
src/skillevaluator/evaluation/dataset_case_removal.py:324
↓ 3 callersFunction_in_unscanned_folder
Whether a plugin-root-relative path is inside a folder that Tier 1 whole-tree scans skip.
src/skillevaluator/plugin_paths.py:394
↓ 3 callersFunction_incomplete_reason
Return the line that explains an INCOMPLETE run verdict.
tests/test_tier3_result_display.py:325
↓ 3 callersFunction_inline_credentials
``(index, flag)`` for each of ``words`` that holds an inline credential, in order. The value of a credential-named flag counts: ``--token=x`` its
src/skillevaluator/validators/mcp_static.py:206
↓ 3 callersFunction_inputs
(wrapper, entry: dict, destination: Path)
tests/tier3/test_plugin_prepare_review.py:70
↓ 3 callersMethod_install_script_case
The notes writer with a second line that runs ``versions/install.sh``, which the scanned copy leaves out. The report is the capture for the f
tests/validators/test_security.py:9913
↓ 3 callersFunction_installer_dependencies
( environment: BaseEnvironment, dependencies: tuple[str, ...], *, system_npm_predicate: str, )
src/skillevaluator/tier3/harbor/installed_agents.py:110
↓ 3 callersFunction_instructions
Dockerfile instructions as (keyword, arguments), with continuations joined and comments dropped.
tests/ci/test_image_build.py:16
↓ 3 callersFunction_inventory
(root: Path)
tests/tier3/test_plugin_prepare_review.py:39
↓ 3 callersFunction_is_ascii_alnum
(character: str)
src/skillevaluator/tier3/harbor/stream_redaction.py:320
↓ 3 callersFunction_is_certificate_failure
Return whether ``error`` is, or wraps, a failed TLS certificate check; a retry cannot fix it.
src/skillevaluator/tier3/harbor/templates/eval.py:1660
↓ 3 callersFunction_is_claude_5_5_or_later
(model)
src/skillevaluator/tier3/harbor/templates/eval.py:1280
↓ 3 callersMethod_is_compose_client_operational_name
(name: str)
src/skillevaluator/tier3/harbor/secure_docker_environment.py:2994
↓ 3 callersFunction_is_count
Whether *value* is a JSON integer, not a boolean.
src/skillevaluator/validators/security.py:2737
↓ 3 callersFunction_is_generated_artifact
(source: Path, parts: tuple[str, ...], kind: str)
src/skillevaluator/publication_identity.py:134
↓ 3 callersFunction_is_k8s_http_status
Return True when a Kubernetes exception matches the given HTTP status code or status phrase.
src/skillevaluator/tier3/harbor/gke_environment.py:632
↓ 3 callersFunction_is_native_catalog_endpoint
Return whether catalog HTTP status has the provider's native meaning.
src/skillevaluator/tier3/harbor/runtime_preflight.py:709
← previousnext →2,701–2,800 of 24,741, ranked by callers