MCPcopy Create free account

hub / github.com/NVIDIA/SkillEvaluator / functions

Functions24,741 in github.com/NVIDIA/SkillEvaluator

↓ 1 callersMethod_updateRadius
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callersMethod_updateScales
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callersFunction_url_credentials
Whether a command word holds a URL with credentials in its userinfo, query, or fragment (as hook commands).
src/skillevaluator/validators/mcp_static.py:247
↓ 1 callersFunction_url_display_mask
For each character of ``url``: whether a finding may show it (scheme, host, port, and path). User information (through the last ``@`` of the auth
src/skillevaluator/validators/mcp_static.py:303
↓ 1 callersFunction_url_is_under
(target: _UrlParts, prefix: _UrlParts)
src/skillevaluator/plugin_component_risk.py:2758
↓ 1 callersFunction_url_path_secret
Whether a URL's path has a segment shaped like a token (``https://hooks.example/ghp_...``).
src/skillevaluator/plugin_component_risk.py:2817
↓ 1 callersFunction_url_path_segments
The percent-decoded path segments of ``url``, read from its raw text (no userinfo, query, or fragment).
src/skillevaluator/validators/mcp_static.py:443
↓ 1 callersFunction_urlopen_with_retry
Open a URL request with exponential backoff and full jitter on transient failures.
src/skillevaluator/tier3/harbor/templates/eval.py:1808
↓ 1 callersFunction_urlopen_without_redirects
(request: Request, *, timeout: float)
src/skillevaluator/model_catalog.py:317
↓ 1 callersFunction_usable_auto_trajectory
Require real agent text/actions and the shape consumed by refinement.
src/skillevaluator/tier3/generate_dataset.py:1738
↓ 1 callersFunction_usage
(command: click.Command, ctx: click.Context)
src/skillevaluator/cli_help.py:140
↓ 1 callersFunction_usage
(trial: dict[str, Any], *, zero_tokens_missing: bool = False)
src/skillevaluator/reporting/dashboard_data.py:250
↓ 1 callersFunction_usage_number
(usage: dict[str, Any] | None, *keys: str)
src/skillevaluator/tier3/harbor/run_budget.py:108
↓ 1 callersFunction_usage_tokens
Return a report trial's ``tokens`` object, or ``None`` without prompt and completion counts.
src/skillevaluator/tier3/harbor/report_data.py:726
↓ 1 callersFunction_user_copies
``(root, path, depth)`` of same-named skills in the user-level roots, for a skill installed in a project root. ``depth`` counts the grouping fold
src/skillevaluator/harness/overlay.py:710
↓ 1 callersFunction_userinfo_carries_credential
(user: str, password: str, *, rule: UserinfoRule)
src/skillevaluator/validators/url_policy.py:516
↓ 1 callersFunction_userns_enabled
()
src/skillevaluator/tier3/harbor/local_sandbox.py:428
↓ 1 callersFunction_uses_skill_files
Whether a read path or shell command touches SKILL.md, the skill's directory, or one of its scripts. The skill name and scripts can come from LLM
src/skillevaluator/tier3/generate_dataset.py:974
↓ 1 callersFunction_uses_variable
(text: str, names: frozenset[str])
src/skillevaluator/plugin_component_risk.py:2163
↓ 1 callersFunction_utc_aware_datetime
Normalize Harbor timestamps before ordering mixed local/UTC results. Harbor 0.22 job summaries can contain naive host-local timestamps while
src/skillevaluator/tier3/harbor/runner.py:4341
↓ 1 callersFunction_utf16_units
(value: Any)
src/skillevaluator/harness/overlay.py:738
↓ 1 callersFunction_utf8_prefix
Return a UTF-8-safe prefix bounded by encoded byte length.
src/skillevaluator/tier3/harbor/runner.py:4352
↓ 1 callersFunction_uv_invocation
``uvx`` / ``uv tool run``: the ``--from`` requirement (else the command), then every ``--with`` requirement. These options count only before the
src/skillevaluator/validators/mcp_static.py:2214
↓ 1 callersFunction_uv_pins
Every place CI chooses a uv version, as ``(where, version)``. ``version`` is ``None`` for a ``setup-uv`` step that names none, because that s
tests/test_ci_workflows.py:504
↓ 1 callersFunction_valid_agent_model_states
Validate a comma-delimited agent list without a backtracking list regex.
scripts/ci/check_public_benchmarks.py:1071
↓ 1 callersMethod_valid_pip_audit_vulnerability
Validate fields consumed from one pip-audit vulnerability entry.
src/skillevaluator/validators/dependencies.py:685
↓ 1 callersFunction_valid_recorded_agent_models
(value: str)
scripts/ci/check_public_benchmarks.py:2026
↓ 1 callersFunction_valid_unrecorded_agent_models
(value: str)
scripts/ci/check_public_benchmarks.py:1092
↓ 1 callersMethod_validate_author
Classify ``metadata.author`` and apply the active policy. Four outcomes, each with its own check name so a policy can set its severit
src/skillevaluator/validators/schema.py:995
↓ 1 callersFunction_validate_bedrock_local_sdk_configuration
Force local values that the probe's bounded overrides could mask.
src/skillevaluator/tier3/harbor/runtime_preflight.py:338
↓ 1 callersMethod_validate_body_content
Validate SKILL.md body has required heading and sections. Per agentskills.io spec, the body has no format restrictions, so we only ha
src/skillevaluator/validators/schema.py:890
↓ 1 callersFunction_validate_canonical_ref
Validate a canonical-ID string ref against the selector source+repo rules. A canonical ID has the shape ``<source>::<repo>::<type>::<name>`` (e.g
src/skillevaluator/models/plugin.py:95
↓ 1 callersFunction_validate_catalog
Run the full validate pipeline once per skill in the catalog. Each skill is an independent job with its own pipeline view, reports (under ``<
src/skillevaluator/cli.py:1779
↓ 1 callersFunction_validate_catalog_parallel
Validate catalog skills concurrently in isolated child processes.
src/skillevaluator/cli.py:1879
↓ 1 callersFunction_validate_chat_tool_call
(tool_call: Any)
src/skillevaluator/tier3/harbor/nvidia_build_bridge.py:1922
↓ 1 callersFunction_validate_checked_destination
(output_path: Path)
src/skillevaluator/reporting/base.py:331
↓ 1 callersFunction_validate_codex_overlay
Field problems of a ``.codex-plugin/plugin.json`` overlay beside a root Agent Plugins manifest. OpenAI documents this overlay as the source of th
src/skillevaluator/plugin_format_schema.py:1617
↓ 1 callersFunction_validate_compose_build
( service_name: str, build: object, environment_dir: Path, *, allowed_env: set[str], )
src/skillevaluator/tier3/harbor/adapter.py:3360
↓ 1 callersFunction_validate_compose_interpolation
(content: dict[str, Any], allowed_env: set[str])
src/skillevaluator/tier3/harbor/adapter.py:3304
↓ 1 callersFunction_validate_compose_model_parse_budget
Reject oversized YAML graphs before PyYAML constructs them.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:221
↓ 1 callersMethod_validate_contained_manifest
Shallow-validate a contained ``.claude-plugin/plugin.json`` file.
src/skillevaluator/validators/plugin_schema.py:209
↓ 1 callersFunction_validate_delegated_caller
Prove the only root workflow job calls the explicitly trusted delegate. The run API lists referenced workflows without parent/child edges. Requir
ci/publish_reports.py:538
↓ 1 callersMethod_validate_directory_structure
Verify workflow has required directory structure.
src/skillevaluator/validators/workflows_schema.py:146
↓ 1 callersFunction_validate_discovery_depth
Validate the optional shallow-discovery cutoff against the hard cap.
src/skillevaluator/utils/secure_fs.py:933
↓ 1 callersFunction_validate_empty_named_input_context_sources
Reject specific reads from an explicitly empty named input context.
src/skillevaluator/tier3/harbor/adapter.py:3158
↓ 1 callersFunction_validate_env_name
(name: Any, config_path: Path, field: str)
src/skillevaluator/tier3/evals_config.py:776
↓ 1 callersFunction_validate_evals_config
(path: Path, results: list[CheckResult])
src/skillevaluator/tier3/evals_spec.py:1214
↓ 1 callersFunction_validate_exact_fields
(data: dict[str, object], expected: frozenset[str], label: str)
src/skillevaluator/embedding/registry.py:894
↓ 1 callersFunction_validate_existing_metadata_network_policy
Raise RuntimeError if a NetworkPolicy object does not block both GKE metadata CIDRs.
src/skillevaluator/tier3/harbor/gke_environment.py:640
↓ 1 callersFunction_validate_existing_output
Allow regenerating saved manifests, including ones with stale source hashes.
src/skillevaluator/tier3/decoy_commands.py:114
↓ 1 callersFunction_validate_fallback_components
(path: Path)
src/skillevaluator/publication_identity.py:450
↓ 1 callersFunction_validate_fallback_stage_exact
(stage: Path, manifests: tuple[_TreeManifest, ...])
src/skillevaluator/tier3/harbor/secure_copy.py:1912
↓ 1 callersFunction_validate_fallback_tree_exact_pass
( stage: Path, manifests: tuple[_TreeManifest, ...], *, child_final_modes: bool, root_mode
src/skillevaluator/tier3/harbor/secure_copy.py:1828
↓ 1 callersMethod_validate_folder
Validate all skills in a folder, aggregating quality results.
src/skillevaluator/validators/quality_score.py:357
↓ 1 callersMethod_validate_folder
(self, root: Path)
src/skillevaluator/validators/rubric_eval.py:640
↓ 1 callersMethod_validate_folder_compliance
Check if folder contains standard SkillEvaluator directories (skills/, team-skills/).
src/skillevaluator/validators/schema.py:1152
↓ 1 callersMethod_validate_folder_compliance
Check if folder follows SkillEvaluator team-rules/ structure.
src/skillevaluator/validators/rules_schema.py:231
↓ 1 callersMethod_validate_folder_compliance
Check if folder follows SkillEvaluator workflows structure.
src/skillevaluator/validators/workflows_schema.py:342
↓ 1 callersMethod_validate_folder_structure
Verify skill is in valid folder hierarchy (skills/ or team-skills/).
src/skillevaluator/validators/schema.py:546
↓ 1 callersMethod_validate_folder_structure
Verify rule is in valid folder hierarchy (team-rules/).
src/skillevaluator/validators/rules_schema.py:172
↓ 1 callersMethod_validate_folder_structure
Verify workflow is in valid folder hierarchy.
src/skillevaluator/validators/workflows_schema.py:231
↓ 1 callersFunction_validate_harbor_job_result
Require Harbor's persisted trial state to be complete and error-free.
src/skillevaluator/tier3/harbor/runner.py:4241
↓ 1 callersMethod_validate_in_plugin_skills
Merge schema results for live skills bundled below ``skills/``.
src/skillevaluator/validators/plugin_schema.py:273
↓ 1 callersFunction_validate_key_metadata
(path: Path, metadata: os.stat_result)
src/skillevaluator/tier3/output_provenance.py:350
↓ 1 callersMethod_validate_license
Validate detected license against allowlist/blocklist.
src/skillevaluator/validators/license.py:680
↓ 1 callersMethod_validate_line_count
Warn if SKILL.md exceeds recommended line count.
src/skillevaluator/validators/schema.py:847
↓ 1 callersMethod_validate_line_count
Warn if rule file exceeds recommended line count.
src/skillevaluator/validators/rules_schema.py:199
↓ 1 callersFunction_validate_manifest
(manifest: object, target: Path)
src/skillevaluator/tier3/decoys.py:279
↓ 1 callersFunction_validate_mcp_toml
(path: Path, results: list[CheckResult])
src/skillevaluator/tier3/evals_spec.py:1223
↓ 1 callersMethod_validate_metadata_fields
Check for recommended metadata fields.
src/skillevaluator/validators/rules_schema.py:209
↓ 1 callersMethod_validate_metadata_id
Validate an optional ``metadata.id`` against the stable-ID contract. A present ID must be an exact 1-64 character lowercase ASCII slug in
src/skillevaluator/validators/schema.py:1089
↓ 1 callersMethod_validate_metadata_id_uniqueness
Report ``metadata.id`` values declared by more than one skill in this collection. Uniqueness is evaluated only across the skills discovered t
src/skillevaluator/validators/schema.py:323
↓ 1 callersMethod_validate_name_matches_directory
Ensure directory name matches frontmatter 'name' field.
src/skillevaluator/validators/schema.py:679
↓ 1 callersFunction_validate_named_hidden_file
( parent: int, name: str, manifest: _FileManifest, )
src/skillevaluator/tier3/harbor/secure_copy.py:2444
↓ 1 callersFunction_validate_named_hidden_tree
( parent_descriptor: int, name: str, manifest: _TreeManifest, )
src/skillevaluator/tier3/harbor/secure_copy.py:1286
↓ 1 callersFunction_validate_namespace_name
(name: str, field_name: str)
src/skillevaluator/tier3/harbor/nvidia_build_bridge.py:1759
↓ 1 callersMethod_validate_naming_conventions
Verify folder name follows kebab-case convention (1-64 chars).
src/skillevaluator/validators/schema.py:619
↓ 1 callersMethod_validate_naming_conventions
Verify filename follows kebab-case convention.
src/skillevaluator/validators/rules_schema.py:191
↓ 1 callersMethod_validate_naming_conventions
Verify workflow directory name follows kebab-case convention.
src/skillevaluator/validators/workflows_schema.py:256
↓ 1 callersFunction_validate_native_agent_judge_model_controls
Keep host judge controls out of the native task's agent environment.
src/skillevaluator/tier3/harbor/adapter.py:5064
↓ 1 callersFunction_validate_native_task_execution_compatibility
Fail closed where Harbor execution can bypass evaluator projections.
src/skillevaluator/tier3/harbor/adapter.py:5509
↓ 1 callersFunction_validate_native_task_host_env_boundaries
Validate every Harbor 0.22 task field that resolves from parent env.
src/skillevaluator/tier3/harbor/adapter.py:5118
↓ 1 callersFunction_validate_native_task_mcp_policy
Check native MCP declarations before Harbor can resolve their headers.
src/skillevaluator/tier3/harbor/adapter.py:5182
↓ 1 callersFunction_validate_native_verifier_compose_boundaries
Keep separate-verifier Compose models inside the verifier env boundary.
src/skillevaluator/tier3/harbor/adapter.py:5239
↓ 1 callersMethod_validate_optional_files
Check for optional supporting files and warn about unexpected items.
src/skillevaluator/validators/schema.py:1223
↓ 1 callersFunction_validate_output_path
Keep manifest output from replacing inputs or changing selected sources.
src/skillevaluator/tier3/decoy_commands.py:77
↓ 1 callersFunction_validate_pages_privacy
Recheck access control immediately before retaining source-bearing reports.
ci/publish_reports.py:911
↓ 1 callersMethod_validate_portability
Flag spec-valid patterns that common agent harnesses load differently or not at all.
src/skillevaluator/validators/schema.py:724
↓ 1 callersFunction_validate_posix_destination
(parent_descriptor: int, output_path: Path)
src/skillevaluator/reporting/base.py:230
↓ 1 callersMethod_validate_previous_version
Validate the optional previous version bound. Returns ``True`` when validation can continue.
src/skillevaluator/validators/version.py:76
↓ 1 callersFunction_validate_private_file_node
(parent: int, staged: _StagedNode, manifest: _FileManifest)
src/skillevaluator/tier3/harbor/secure_copy.py:1672
↓ 1 callersFunction_validate_private_tree
( descriptor: int, manifests: tuple[_TreeManifest, ...], *, root_device: int, )
src/skillevaluator/tier3/harbor/secure_copy.py:1218
↓ 1 callersMethod_validate_reference_file
Validate a single reference .mdc file.
src/skillevaluator/validators/workflows_schema.py:290
↓ 1 callersMethod_validate_references
Validate reference .mdc files in references/ directory.
src/skillevaluator/validators/workflows_schema.py:265
↓ 1 callersFunction_validate_revision
(value: str)
scripts/classify_ci_changes.py:132
↓ 1 callersMethod_validate_rules_directory
Validate a directory containing .mdc files.
src/skillevaluator/validators/rules_schema.py:127
↓ 1 callersFunction_validate_run
(run: dict, repository: str, run_id: int)
ci/publish_reports.py:489
↓ 1 callersMethod_validate_scripts
Validate scripts in references/scripts/ directory.
src/skillevaluator/validators/workflows_schema.py:308
↓ 1 callersMethod_validate_single_asset
Validate license compliance for a single asset directory.
src/skillevaluator/validators/license.py:259
↓ 1 callersMethod_validate_single_skill
Audit dependencies for a single skill directory.
src/skillevaluator/validators/dependencies.py:155
← previousnext →8,101–8,200 of 24,741, ranked by callers