Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/NVIDIA/SkillEvaluator
/ functions
Functions
24,741 in github.com/NVIDIA/SkillEvaluator
⨍
Functions
24,741
◇
Types & classes
1,292
↳
Endpoints
703
↓ 1 callers
Method
_passive
(self, start: int, end: int, kind: str)
src/skillevaluator/evaluation/dataset_case_removal.py:539
↓ 1 callers
Method
_passive_verb_at
(self, index: int)
src/skillevaluator/evaluation/dataset_case_removal.py:514
↓ 1 callers
Function
_password
()
tests/validators/test_plugin_mcp.py:90
↓ 1 callers
Function
_password_tails
Return the ends of a short URL password, each with the "@" after it.
src/skillevaluator/utils/secret_redaction.py:75
↓ 1 callers
Function
_patch_header_paths
Distinct paths the ``Add``/``Update``/``Delete File`` headers name, in patch order.
src/skillevaluator/tier3/eval_core/atif_helpers.py:384
↓ 1 callers
Function
_patch_targets
Paths an apply_patch tool call writes (see :func:`_patch_written_paths`).
src/skillevaluator/tier3/eval_core/plugin_signals.py:4385
↓ 1 callers
Function
_path
(value, field: str, *, skill: bool = False)
ci/tier_manifest.py:57
↓ 1 callers
Function
_path
(value: object, *, skill: bool = False)
ci/security_baselines.py:42
↓ 1 callers
Function
_path_entry_in
(entry: str, text: str)
src/skillevaluator/tier3/eval_core/checks.py:932
↓ 1 callers
Function
_path_is_contained
(path: Path, root: Path)
src/skillevaluator/tier3/output_provenance.py:506
↓ 1 callers
Method
_path_is_within_allowed_write_roots
(self, path: Path)
src/skillevaluator/tier3/harbor/local_environment.py:1803
↓ 1 callers
Function
_path_key
(path: Path)
src/skillevaluator/tier3/harbor/adapter.py:1015
↓ 1 callers
Function
_path_scope
(file_path: str, *, artifact_wide: bool)
src/skillevaluator/validators/skillspector_scope.py:552
↓ 1 callers
Function
_path_spellings
``path`` as written natively, with ``/``, and with ``\\`` escaped as in JSON.
src/skillevaluator/evaluation/tier3_feedback_chat.py:554
↓ 1 callers
Function
_path_start
Index where the path, query, or fragment of a special-scheme URL starts, after ``scheme:`` and its authority.
src/skillevaluator/validators/url_policy.py:592
↓ 1 callers
Function
_path_start
Where the absolute path that ends at ``end`` starts, searching from ``start``. It starts at the first root, unless a later root follows a path se
src/skillevaluator/evaluation/tier3_feedback_chat.py:521
↓ 1 callers
Function
_paths
(items: list[str])
src/skillevaluator/harness/overlay.py:281
↓ 1 callers
Function
_paths_safe
(workspace: Path, reports: Path)
ci/run_tier1_image.py:61
↓ 1 callers
Function
_pattern
(*parts: str, flags: int = 0)
scripts/check_oss_boundary.py:118
↓ 1 callers
Function
_payloads
Prefer authoritative per-validator results over a top-level Tier 3 mirror.
src/skillevaluator/reporting/dashboard_data.py:157
↓ 1 callers
Function
_percentile
Linearly interpolated percentile of an already sorted sequence.
src/skillevaluator/tier3/harbor/stats.py:471
↓ 1 callers
Function
_performance_column_config
()
src/skillevaluator/reporting/dashboard_app.py:154
↓ 1 callers
Function
_perl_version
()
scripts/script_invocation_differential.py:109
↓ 1 callers
Function
_permission_flag_issue
(flag: str, path: str, risk: _OptionRisk)
src/skillevaluator/validators/mcp_static.py:3341
↓ 1 callers
Function
_persist_pre_execution_failure
Retain redacted probe provenance for failures after run reservation.
src/skillevaluator/tier3/harbor/runner.py:6138
↓ 1 callers
Function
_persistable_name
Whether an argument-derived component name may be persisted (identifier-shaped or declared).
src/skillevaluator/tier3/eval_core/plugin_signals.py:2803
↓ 1 callers
Function
_persisted_result_projection
Build a compact on-disk result while leaving the returned result complete.
src/skillevaluator/tier3/harbor/runner.py:610
↓ 1 callers
Function
_persisted_run_result
Read the run's own ``result.json`` so re-rendered reports keep the engine status.
src/skillevaluator/evaluation/tier3_report.py:592
↓ 1 callers
Function
_persisted_trial_name
Derive output and source names only from physical Harbor path components.
src/skillevaluator/tier3/harbor/collector.py:5914
↓ 1 callers
Method
_phrases
Every (start, end, kind) token span that reads as a case in the dataset.
src/skillevaluator/evaluation/dataset_case_removal.py:393
↓ 1 callers
Function
_physical_location
( finding: Finding, workspace_root: Path | None, scan_root: Path | None = None, )
src/skillevaluator/reporting/sarif_reporter.py:202
↓ 1 callers
Function
_physical_steps_layout_present
Return whether a trial has any physical steps entry, failing closed on I/O errors.
src/skillevaluator/tier3/harbor/collector.py:3238
↓ 1 callers
Function
_piece_digest
One piece's sha256 in one staged arm, or ``None`` when the arm does not have it.
src/skillevaluator/tier3/loose_parts.py:1426
↓ 1 callers
Function
_pipeline_last_stage_keeps_bindings
Whether a binding made in a pipeline's last command survives it. ``True`` and ``False`` when the shell settles it; ``None`` when the text doe
src/skillevaluator/tier3/eval_core/script_invocation.py:2728
↓ 1 callers
Function
_pipeline_upstream_names_script
Whether an earlier stage of the pipeline that the segment at ``idx`` ends names the expected script as one of its words.
src/skillevaluator/tier3/eval_core/script_invocation.py:2537
↓ 1 callers
Function
_pipes_fetch_into_interpreter
A pipeline stage that fetches, followed by a later stage that runs its standard input as a program.
src/skillevaluator/plugin_component_risk.py:2177
↓ 1 callers
Function
_place_kept_text
Return where each piece of *kept*, the text a masker kept between its markers, starts in *read*, the text it read. A piece placed at the same ind
src/skillevaluator/tier3/eval_core/checks.py:4120
↓ 1 callers
Function
_plain
Drop HTML tags, entities and Markdown emphasis, and collapse whitespace.
tests/changelog_entries.py:14
↓ 1 callers
Function
_plain_pipeline
Whether ``unit``, its comments blanked, is one pipeline of simple commands that the walk reads as the shell does. Words and complete redirect
src/skillevaluator/tier3/eval_core/script_invocation.py:1830
↓ 1 callers
Function
_plaintext_registry_urls
``(option, url)`` for each registry or index URL of a package runner that uses plaintext http. A loopback host (a local registry mirror) does not
src/skillevaluator/validators/mcp_static.py:627
↓ 1 callers
Method
_plan_lines
(self, plan: Tier3RunPlan)
src/skillevaluator/tier3/harbor/progress.py:763
↓ 1 callers
Function
_plugin
(payload: dict, budget: _Budget, warnings: list[NativeNotice])
src/skillevaluator/reporting/native_evidence.py:222
↓ 1 callers
Method
_plugin_file
(self, rel: str, what: str)
src/skillevaluator/tier3/loose_parts.py:1089
↓ 1 callers
Function
_plugin_file_checker
A predicate for a relative command word that names a regular file inside *plugin_root*.
src/skillevaluator/tier3/plugin_native.py:532
↓ 1 callers
Function
_plugin_ignored_agent_fields
The frontmatter fields of a subagent that Claude Code ignores in a plugin agent (in declared spelling).
src/skillevaluator/tier3/loose_parts.py:1255
↓ 1 callers
Function
_plugin_launch_path
Anchor explicit ./ launch paths when Claude copies the plugin tree.
src/skillevaluator/tier3/plugin_prepare.py:537
↓ 1 callers
Function
_plugin_rel
(plugin_root: Path, path: Path)
src/skillevaluator/tier3/loose_parts.py:1277
↓ 1 callers
Function
_plugin_root_refs
Root placeholders a hook command may use, each also in its unbraced ``$VAR`` form. ``${CLAUDE_PLUGIN_ROOT}`` is always included, plus the manifes
src/skillevaluator/plugin_component_risk.py:1707
↓ 1 callers
Function
_plugin_signal_trajectory
Reuse the collector's bounded, topology-authorized continuation/child graph.
src/skillevaluator/tier3/plugin_trace_artifacts.py:25
↓ 1 callers
Function
_plural
(count: int, word: str)
src/skillevaluator/harness/report.py:83
↓ 1 callers
Function
_png_bytes
The 1x1 PNG that the settings-hook image report scanned.
tests/validators/test_skillspector_hook_surface.py:196
↓ 1 callers
Function
_policy
(records: dict[int, dict], expected: list[int])
ci/comment_summary.py:101
↓ 1 callers
Function
_populate_reward_entry_id
( reward_data: dict[str, Any], result_payload: dict[str, Any] | None, trial_name_val:
src/skillevaluator/tier3/harbor/collector.py:3679
↓ 1 callers
Method
_portability_finding
( self, check_name: str, default: Severity, message: str, file_path: s
src/skillevaluator/validators/schema.py:705
↓ 1 callers
Function
_portable_fingerprint_mode
Normalize mode bits to the semantics the active copier preserves.
src/skillevaluator/tier3/harbor/secure_copy.py:250
↓ 1 callers
Function
_portable_mode
Keep only the execute state that a git checkout reproduces. Git records whether a file is executable and nothing for directories; the group a
src/skillevaluator/tier3/decoy_identity.py:23
↓ 1 callers
Function
_portable_trial_name_key
Normalize one output name for case-insensitive and Win32-compatible collision checks.
src/skillevaluator/tier3/harbor/collector.py:5922
↓ 1 callers
Method
_position_is_covered
(self, position: int)
src/skillevaluator/tier3/harbor/secure_docker_environment.py:462
↓ 1 callers
Function
_positive_token_count
Return a positive whole token count the run record can hold, or None for anything else (never bool).
src/skillevaluator/tier3/harbor/collector.py:7291
↓ 1 callers
Function
_posix_directory_flags
No-follow directory open flags (``AttributeError`` where the platform lacks them).
src/skillevaluator/utils/secure_fs.py:1166
↓ 1 callers
Function
_possibly_references_target_directory
(value: Any, target_skill: str)
src/skillevaluator/tier3/eval_core/checks.py:1884
↓ 1 callers
Function
_powershell_encoded_command
Whether ``argv`` runs PowerShell with ``-EncodedCommand``: a base64 program that cannot be read statically.
src/skillevaluator/validators/mcp_static.py:561
↓ 1 callers
Function
_pre_agent_setup
(value: Any, config_path: Path)
src/skillevaluator/tier3/evals_config.py:804
↓ 1 callers
Function
_pre_agent_setup_healthcheck_toml_block
(pre_agent_setup: list[str] | None)
src/skillevaluator/tier3/harbor/adapter.py:2338
↓ 1 callers
Function
_pre_dispatch_exec_api
Return a fake CoreV1Api whose routed exec handshake fails with each queued error before dispatch.
tests/test_harbor_gke_environment.py:1016
↓ 1 callers
Function
_prefixed
(path: str)
src/skillevaluator/validators/skillspector_scope.py:998
↓ 1 callers
Function
_preflight_generated_tasks
(entries: list[dict[str, Any]], output_dir: Path)
src/skillevaluator/tier3/harbor/adapter.py:1915
↓ 1 callers
Function
_preflight_yaml
(raw: str)
src/skillevaluator/utils/structured_data.py:163
↓ 1 callers
Function
_preflight_yaml_merge_work
Bound flattened pairs, including duplicate keys that construction discards. PyYAML flattens merges into lists before constructing dictionaries. C
src/skillevaluator/utils/structured_data.py:99
↓ 1 callers
Method
_prepare_bridge_client
Perform optional agent-specific setup after the bridge is healthy.
src/skillevaluator/tier3/harbor/local_agents.py:705
↓ 1 callers
Function
_prepare_checked_parent
Create and snapshot parent components for platforms without dir-fd writes.
src/skillevaluator/reporting/base.py:304
↓ 1 callers
Function
_prepare_dataset
Reuse valid authored sources; create one starter case only when absent.
src/skillevaluator/tier3/workflow.py:343
↓ 1 callers
Function
_prepare_generated_outputs
Validate the complete cleanup plan, then reset current and omitted generated outputs.
src/skillevaluator/tier3/harbor/collector.py:570
↓ 1 callers
Function
_prepare_hidden_fallback_tree
(stage: Path, manifests: tuple[_TreeManifest, ...])
src/skillevaluator/tier3/harbor/secure_copy.py:1921
↓ 1 callers
Function
_prepare_hidden_final_tree
(node: _StagedNode, manifests: tuple[_TreeManifest, ...])
src/skillevaluator/tier3/harbor/secure_copy.py:1269
↓ 1 callers
Method
_prepare_log_dirs
(self)
src/skillevaluator/tier3/harbor/local_environment.py:563
↓ 1 callers
Function
_prepare_moved_backup
Verify a moved original and immediately make any residue private.
src/skillevaluator/tier3/harbor/secure_copy.py:1375
↓ 1 callers
Function
_prepare_native_environment
Stage SkillEvaluator runtime additions into a copied native Harbor task.
src/skillevaluator/tier3/harbor/adapter.py:4581
↓ 1 callers
Function
_prepend_facts
(text: str)
src/skillevaluator/tier3/eval_core/atif_helpers.py:1312
↓ 1 callers
Function
_prescan_local_pax_payload
Validate bounded PAX framing and return its effective size override.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:874
↓ 1 callers
Function
_prescan_uncompressed_tar_archive
Bound tar metadata before ``tarfile`` is allowed to allocate it.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:920
↓ 1 callers
Function
_preserve_auto_authored_cases
Keep authored tasks and execution metadata while proposing refined expectations.
src/skillevaluator/tier3/generate_dataset.py:1762
↓ 1 callers
Method
_preserve_gateway_model_name
Undo Harbor's default Codex model truncation when routing through an OpenAI-compatible gateway.
src/skillevaluator/tier3/harbor/local_agents.py:461
↓ 1 callers
Method
_primary_category
Determine the primary (highest-risk) category from a set.
src/skillevaluator/validators/unicode_smuggle.py:475
↓ 1 callers
Function
_print_baseline_fairness_warnings
Name generated cases that still grade skill use; `tier3 validate` reports the same warnings. The check does not read expected_script, and normali
src/skillevaluator/tier3/generate_dataset.py:953
↓ 1 callers
Function
_print_catalog_divider
A full-width rule announcing the next per-skill job in a catalog run.
src/skillevaluator/cli.py:1302
↓ 1 callers
Method
_print_finding
Print a single finding with structured details. Dynamic finding fields (message, location, content, suggestion) are escaped so values
src/skillevaluator/reporting/cli.py:800
↓ 1 callers
Method
_print_findings
Print findings with tree structure.
src/skillevaluator/reporting/cli.py:770
↓ 1 callers
Function
_print_gitlab_root_merge_steps
(wizard, existing)
src/skillevaluator/_ci_onboarding_all_tiers.py:580
↓ 1 callers
Function
_print_json
Print a stable, human-readable JSON object.
src/skillevaluator/tier3/reference_skills/api-caller/scripts/call_api.py:153
↓ 1 callers
Function
_print_qualification
(decision)
src/skillevaluator/tier3/grader_validation.py:798
↓ 1 callers
Method
_print_quality_table
Print a unified quality scores table for single or multi-skill runs.
src/skillevaluator/reporting/cli.py:344
↓ 1 callers
Method
_print_reference_summary
Print how many referenced artifacts SkillSpector did not completely inspect in a failed result.
src/skillevaluator/reporting/cli.py:326
↓ 1 callers
Method
_print_rubric_table
Print LLM rubric evaluation results as a table.
src/skillevaluator/reporting/cli.py:409
↓ 1 callers
Function
_print_run_banner
Print the pre-run header (target + detected type + active profile). Restores parity with SkillEvaluator's ``_print_validation_banner``: before an
src/skillevaluator/cli.py:2068
↓ 1 callers
Method
_print_summary_stats
Print summary statistics.
src/skillevaluator/reporting/cli.py:724
↓ 1 callers
Function
_print_tier_banner
Print a labeled per-tier section banner (parity with SkillEvaluator).
src/skillevaluator/cli.py:2062
↓ 1 callers
Function
_print_validate_results
(skill_path: Path, results: list[Any])
src/skillevaluator/tier3/commands.py:1296
↓ 1 callers
Function
_prioritized_evidence_rewards
(rewards: list[dict[str, Any]])
src/skillevaluator/tier3/harbor/report.py:1090
↓ 1 callers
Function
_private_agent_paths
Resolve both arms on the calling thread, before launching worker pools.
src/skillevaluator/tier3/harbor/runner.py:5117
← previous
next →
7,101–7,200 of 24,741, ranked by callers