MCPcopy Create free account

hub / github.com/NVIDIA/SkillEvaluator / functions

Functions24,741 in github.com/NVIDIA/SkillEvaluator

↓ 4 callersFunctionsanitize_tier2_results
Remove supplied host paths from every user-visible result field in place.
src/skillevaluator/utils/tier2_paths.py:157
↓ 4 callersFunctionscan_excluded_name
The :data:`~skillevaluator.constants.SCAN_EXCLUDED_DIRS` name a folder is pruned as, or ``None``. Whole-tree discovery also prunes a differently
src/skillevaluator/plugin_paths.py:353
↓ 4 callersFunctionscan_shim
Put a fake ``skillevaluator`` on a PATH and return that PATH. A scan records its PID and its background tool's PID in ``scan.pids``, then run
tests/ci/test_run_tier1.py:1477
↓ 4 callersFunctionscanner_reward
(calls, *, root="/workspace", read_files=False)
tests/tier3/test_private_runtime_evidence.py:147
↓ 4 callersFunctionselect_agent_eval_candidate
Select the strongest Tier 3 result and payload independent of input order.
src/skillevaluator/reporting/base.py:1849
↓ 4 callersFunctionselect_catalog_models
Return a bounded filtered view with the configured model first.
src/skillevaluator/model_catalog.py:497
↓ 4 callersFunctionsensitive_values
Find literal values to redact without ever emitting them in diagnostics.
src/skillevaluator/utils/secret_redaction.py:36
↓ 4 callersFunctionskillspector_provenance_records
Return the provenance record of one result for human-readable reports, if any.
src/skillevaluator/skillspector_provenance.py:440
↓ 4 callersFunctionsource_validation_errors
Return the failed Tier 3 source checks recorded in ``metadata``, in order. Tier 3 source preflight stores its checks under ``source_validation``
src/skillevaluator/reporting/base.py:2583
↓ 4 callersFunctionsplit_into_paragraphs
Split an oversized section into paragraph-level chunks.
src/skillevaluator/deduplication/utils/chunker.py:214
↓ 4 callersMethodstages_members_only
Whether only the inventoried member skills are staged, without a copy of the plugin tree. A native OpenClaw plugin's members are exactly its
src/skillevaluator/tier3/plugin_native.py:1383
↓ 4 callersMethodstartOf
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 4 callersFunctionsummarize_dataset
Count normalized cases without interpreting any count as sufficiency.
src/skillevaluator/tier3/dataset_coverage.py:18
↓ 4 callersFunctionsummarize_dataset_entries
Summarize unique task identities and their activation intent.
src/skillevaluator/tier3/harbor/report_data.py:127
↓ 4 callersFunctionsupplementary_total_bytes
Return the configured supplement budget, which never widens the declared root. The operator may raise or lower how many bytes of skill-owned guid
src/skillevaluator/validators/rubric_eval.py:450
↓ 4 callersFunctionsymbol
Keep status labels readable in terminals without Unicode support.
src/skillevaluator/ci_onboarding.py:39
↓ 4 callersMethodthrottled_response
Build an HTTP 429 ThrottlingException response.
tests/test_retry.py:441
↓ 4 callersMethodto_metadata
(self)
src/skillevaluator/embedding/extractor.py:116
↓ 4 callersFunctiontoken_usage_table
Return the HTML report's token usage table, or ``None`` without rows. Each agent shows its first :data:`HTML_CASE_ROWS_PER_AGENT` case rows;
src/skillevaluator/reporting/tier3_token_usage.py:350
↓ 4 callersFunctiontt
(t,e,i,s=1e-6)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 4 callersMethodupdateElement
(t,e,i,s)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 4 callersFunctionvalidate_agents
Return unsupported agent names.
src/skillevaluator/tier3/commands.py:426
↓ 4 callersFunctionvalidate_embedding_vector
Validate a provider vector before it reaches similarity arithmetic.
src/skillevaluator/embedding/client.py:42
↓ 4 callersFunctionvalidate_line_count
Validate that a file doesn't exceed recommended line count. Args: line_count: Actual number of lines max_lines: Maximum recommend
src/skillevaluator/validators/naming_utils.py:56
↓ 4 callersFunctionvalidate_output_provenance_key_location
Keep the private marker key out of every tree that may reach an agent.
src/skillevaluator/tier3/harbor/adapter.py:438
↓ 4 callersFunctionvalidate_provenance_key_outside
Reject a private-key location that could enter an evaluated or generated tree.
src/skillevaluator/tier3/output_provenance.py:521
↓ 4 callersFunctionverify
(verifier, record, payload)
tests/ci/test_verify_provenance.py:84
↓ 4 callersMethodverify
Convenience wrapper around ``process()``.
src/skillevaluator/inference/finding_verifier.py:193
↓ 4 callersMethodwait_for_workers
(self, timeout: float)
src/skillevaluator/tier3/harbor/nvidia_build_bridge.py:913
↓ 4 callersFunctionwithout_userinfo
``url`` without the userinfo of its raw text, which :func:`url_credentials` checks for credentials. A WHATWG client can read part of that userinf
src/skillevaluator/validators/url_policy.py:507
↓ 4 callersFunctionworkflow
(path)
tests/ci/test_repository_skill_ci.py:16
↓ 4 callersMethodwrap
(self, *_args, **_kwargs)
tests/test_harbor_local_mode.py:4399
↓ 4 callersFunctionwrapper_decision
A wrapper-mode decision with ``reason`` (the default, or an ``auto`` fallback).
src/skillevaluator/tier3/plugin_native.py:2189
↓ 4 callersFunctionwrite_json
(path: Path, payload: dict)
ci/run_tiers.py:157
↓ 4 callersFunctionwrite_tree
(root: Path, files: dict)
tests/test_plugin_markdown.py:22
↓ 4 callersFunctionwt
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 4 callersFunctionxn
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 4 callersFunctionze
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionBn
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionFt
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionHe
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionIt
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionNs
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionP
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 3 callersFunctionPi
(t,e,i,s)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionPo
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionV
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 3 callersFunctionXi
(t,e,i,s,n,o)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionY
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 3 callersFunctionYi
(t,e,i,s,n)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersFunctionZt
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 3 callersMethod__init__
( self, message: str, *, kind: ModelCatalogFailureKind | str = ModelCatalogFai
src/skillevaluator/model_catalog.py:73
↓ 3 callersMethod__init__
(self, *, suppress_visually_hidden: bool = True)
scripts/ci/check_public_benchmarks.py:525
↓ 3 callersMethod__init__
(self, chunks: list[bytes])
tests/test_harbor_secure_docker_environment.py:165
↓ 3 callersMethod__init__
(self, *, branch_sha="a" * 40, record=None)
tests/ci/test_publish_reports.py:449
↓ 3 callersFunction_absolute_no_resolve
Return an absolute lexical path without resolving links.
src/skillevaluator/utils/secure_fs.py:200
↓ 3 callersFunction_across_limit
Filler, then ``text`` placed so that only its first ``inside`` characters fit within ``limit``.
tests/evaluation/test_tier3_feedback_chat.py:852
↓ 3 callersFunction_action_repository
The ``owner/repo`` half of a ``uses:`` value, folded for comparison. GitHub resolves repository names case-insensitively, so ``Actions/Checkout``
tests/test_ci_workflows.py:95
↓ 3 callersFunction_advisory_agent_eval_payload
Build the canonical (but empty) ``agent_eval`` payload for a skipped Tier 3 run. The combined HTML/JSON report only renders a Tier 3 section when
src/skillevaluator/evaluation/tier3_report.py:257
↓ 3 callersFunction_advisory_skip_result
(payload: str)
tests/reporting/test_rich_markup_escaping.py:171
↓ 3 callersFunction_agent
(status: str = "succeeded", baseline: str = "succeeded")
tests/tier3/test_baseline_timeouts.py:1516
↓ 3 callersFunction_agent
(call_id: str, fn: str, args: dict[str, Any], content: str, **extra: Any)
tests/tier3/test_f90_signals_handoff_prompt.py:34
↓ 3 callersFunction_agent_dimension
(agent: dict[str, Any], dim_id: str)
src/skillevaluator/reporting/benchmark.py:1759
↓ 3 callersFunction_agent_eval_dimension_scores
Return with-skill scores only when matching baseline evidence exists.
src/skillevaluator/reporting/base.py:1172
↓ 3 callersFunction_agent_eval_has_publication_target_conflict
(payload: dict[str, Any] | None)
src/skillevaluator/reporting/base.py:883
↓ 3 callersFunction_agent_eval_result
(payload: str)
tests/reporting/test_rich_markup_escaping.py:149
↓ 3 callersFunction_agent_plugin
(root: Path, files: dict)
tests/validators/test_plugin_static_risk_review_fixes.py:51
↓ 3 callersFunction_agent_quality_verdict_reason
(agent: dict[str, Any])
src/skillevaluator/evaluation/tier3_report.py:4680
↓ 3 callersFunction_agent_scorecard_bar_widths
(output: str)
tests/test_harbor_html_report_failures.py:73
↓ 3 callersFunction_agent_statistics
(arms: dict[str, ArmObservations], **kwargs: Any)
tests/tier3/test_private_ladder_stats.py:550
↓ 3 callersFunction_agents
()
tests/reporting/test_execution_backend_reporting.py:19
↓ 3 callersFunction_agents
()
tests/reporting/test_budget_partial_reporting.py:18
↓ 3 callersFunction_aggregate_execution_error_details
Aggregate hidden diagnostic occurrences without duplicating display text.
src/skillevaluator/evaluation/tier3_report.py:5181
↓ 3 callersFunction_all_steps
(workflow: dict[str, Any])
tests/test_ci_workflows.py:91
↓ 3 callersMethod_allowed_write_roots
(self)
src/skillevaluator/tier3/harbor/local_environment.py:1819
↓ 3 callersFunction_anthropic_fallback_response
Return a reply whose server-side fallback block names the model that served it.
tests/inference/test_llm_client.py:1480
↓ 3 callersFunction_anthropic_request
(verifier, monkeypatch)
tests/tier3/test_judge_policies.py:149
↓ 3 callersFunction_anthropic_to_chat_request
( payload: Any, *, max_output_tokens: int = MAX_OUTPUT_TOKENS_PER_REQUEST )
src/skillevaluator/tier3/harbor/nvidia_build_bridge.py:1142
↓ 3 callersFunction_append_evaluator_runtime_lines
Append evaluator-owned instructions without trusting authored text matches.
src/skillevaluator/tier3/harbor/adapter.py:3947
↓ 3 callersMethod_append_visible
(self, data: str)
scripts/ci/check_public_benchmarks.py:576
↓ 3 callersFunction_applicable_accuracy_criteria
Return accuracy verdicts minus criteria the verifier excluded from the score.
src/skillevaluator/tier3/harbor/report.py:465
↓ 3 callersFunction_apply_case_findings
Lead the report with the case-based verdict and fixes; also saves them as what_to_fix.json for the CLI. ``fixes`` already saved for the run are r
src/skillevaluator/evaluation/tier3_report.py:4334
↓ 3 callersFunction_apply_patch
(header: str, **args: str)
tests/tier3/test_security_home_paths.py:84
↓ 3 callersFunction_apt_install_packages
Return the words the shell passes to ``apt-get install``. Docker joins a line that ends in ``\\`` with the next line before the shell sees th
tests/test_harbor_apt_install_lines.py:28
↓ 3 callersFunction_arguments
(stage: list[tuple[str, str]], keyword: str)
tests/ci/test_image_build.py:36
↓ 3 callersFunction_arm_not_applicable
Return only all-N/A facts validated against complete scored reward groups.
src/skillevaluator/evaluation/tier3_report.py:2063
↓ 3 callersFunction_assert_equal
(parts: LooseParts, tasks: dict[str, list[Path]])
tests/tier3/test_private_loose_dependency_scaffold_review.py:48
↓ 3 callersFunction_assert_finding_ids_match_text
Every finding ID in a shared report is hashed from that report's redacted text.
tests/ci/test_secret_redaction.py:52
↓ 3 callersFunction_assert_helper_runs
(skill: Path, *, eval_name: str = "eval")
tests/tier3/test_runtime_eval_directories.py:43
↓ 3 callersFunction_assert_lf_script
(path: Path, shebang: bytes)
tests/tier3/test_generated_script_line_endings.py:40
↓ 3 callersFunction_assert_no_controls_or_emoji
No ESC, C1 or other control byte reaches the output, and emoji codes stay literal.
tests/reporting/test_rich_markup_escaping.py:98
↓ 3 callersFunction_assert_no_path_filter
(workflow: dict[str, Any], event: str = "pull_request")
tests/test_ci_workflows.py:73
↓ 3 callersFunction_attach_harbor_evidence_to_conclusions
( conclusions: list[dict[str, Any]], evidence_links: list[dict[str, Any]], )
src/skillevaluator/evaluation/tier3_report.py:4144
↓ 3 callersFunction_attribute_scan_result
Map partition-local paths and grouped occurrences back to their owner.
src/skillevaluator/tier1/plugin_checks.py:186
↓ 3 callersMethod_audit_pyproject
Audit declared pins without installing the target or running build code.
src/skillevaluator/validators/dependencies.py:202
↓ 3 callersFunction_baseline_coverage
Say whether this run has a no-skill baseline score for one dimension.
src/skillevaluator/reporting/benchmark.py:1800
↓ 3 callersFunction_bash
(command: str)
tests/tier3/test_security_home_paths.py:50
↓ 3 callersFunction_bedrock_error
(code, status, message)
tests/tier3/test_judge_policies.py:458
↓ 3 callersMethod_begin_attempt
Scope a handoff while retaining coordinator data registered for its jobs.
src/skillevaluator/tier3/plugin_evidence.py:317
↓ 3 callersFunction_behavior_bundle
Build the behavior_check bundle with the module the judge belongs to.
tests/tier3/test_judge_contracts.py:393
← previousnext →2,501–2,600 of 24,741, ranked by callers