MCPcopy Create free account

hub / github.com/NVIDIA/SkillEvaluator / functions

Functions24,741 in github.com/NVIDIA/SkillEvaluator

↓ 2 callersFunction_check_bridge_health
Require one authenticated health response from a bound bridge.
src/skillevaluator/tier3/harbor/nvidia_build_bridge.py:734
↓ 2 callersFunction_check_claude_monitors
``monitors``: a path to a monitors file, or an array of monitor objects with unique names.
src/skillevaluator/plugin_format_schema.py:736
↓ 2 callersFunction_check_claude_user_config
(value: Any, where: str, issues: list[ManifestIssue], *, label: str)
src/skillevaluator/plugin_format_schema.py:803
↓ 2 callersFunction_check_codex_interface
( interface: Any, issues: list[ManifestIssue], *, label: str, require_complete: bool = True )
src/skillevaluator/plugin_format_schema.py:1470
↓ 2 callersFunction_check_codex_keywords
Codex reads 'keywords' as an array of strings: it may be left out, but an explicit null is refused too.
src/skillevaluator/plugin_format_schema.py:1541
↓ 2 callersFunction_check_codex_types
Report each of *fields* that is set to a type Codex does not accept (see ``_CODEX_FIELD_TYPES``).
src/skillevaluator/plugin_format_schema.py:1379
↓ 2 callersFunction_check_command
(server: _ServerFindings, config: dict[str, Any])
src/skillevaluator/validators/mcp_static.py:731
↓ 2 callersFunction_check_endpoint_url
Scheme, host, credential, and endpoint checks for one URL field of an MCP server (``label`` names it).
src/skillevaluator/validators/mcp_static.py:1008
↓ 2 callersFunction_check_pinning
HIGH for a moving tag (``@latest``, ``:main``) where the runner reads the version; MEDIUM for no exact version. The tag is read from the parsed p
src/skillevaluator/validators/mcp_static.py:3520
↓ 2 callersMethod_check_pod_terminated
Raise immediately if the pod was deleted/preempted or is in a terminal state.
src/skillevaluator/tier3/harbor/gke_environment.py:1127
↓ 2 callersFunction_check_semver
(data: dict[str, Any], issues: list[ManifestIssue], *, label: str, level: IssueLevel)
src/skillevaluator/plugin_format_schema.py:239
↓ 2 callersFunction_check_staged_baseline_does_not_contain_target
Fail closed if any agent-visible staged file is the evaluated instructions.
src/skillevaluator/tier3/harbor/adapter.py:1065
↓ 2 callersFunction_check_url_inline_secrets
Flag credentials written into an MCP URL; return whether its user information carries one. The URL ``text`` (as the client builds it) goes throug
src/skillevaluator/validators/mcp_static.py:451
↓ 2 callersFunction_check_url_scheme
Report a dangerous, plaintext, or not-allowed (non-empty, not https/wss) scheme of one URL field.
src/skillevaluator/validators/mcp_static.py:1102
↓ 2 callersFunction_checkable_tokens
Extract ``(claim, match_mode)`` pairs from *text*. Modes are ``"path"`` for an artifact path and ``"ci"`` for a backtick-quoted snippet of fo
src/skillevaluator/tier3/eval_core/atif_helpers.py:992
↓ 2 callersFunction_checked_content
Return model text, raising typed errors for refusals, truncation and empty replies.
src/skillevaluator/inference/client.py:327
↓ 2 callersFunction_checks
(findings)
tests/test_openclaw_plugin_discovery.py:59
↓ 2 callersFunction_child_env
Build the child environment from an allowlist, with HOME and the Claude config in *home*.
src/skillevaluator/validators/claude_plugin_validate.py:180
↓ 2 callersFunction_child_environment
What the command at ``stop`` inherits from this shell. The exported names, then the segment's own ``NAME=value`` prefix, then each ``env`` st
src/skillevaluator/tier3/eval_core/script_invocation.py:1377
↓ 2 callersFunction_chmod_named_entry
Change a named entry's mode without following a final link. CPython has no ``lchmod`` on Linux and maps ``dir_fd`` plus ``follow_symlinks=False``
src/skillevaluator/tier3/harbor/secure_copy.py:885
↓ 2 callersFunction_classify_address
``(classification, embedded)``: an embedded IPv4 (mapped, 6to4, Teredo, NAT64, compatible) decides for it.
src/skillevaluator/validators/mcp_static.py:2899
↓ 2 callersFunction_classify_bedrock_retry_error
Return (is_retriable, status_label, retry_after_str) for a Bedrock Converse exception.
src/skillevaluator/tier3/harbor/templates/eval.py:2057
↓ 2 callersFunction_classify_image
Classify a container image reference (``repo[:tag][@digest]``).
src/skillevaluator/validators/mcp_static.py:2479
↓ 2 callersFunction_classify_ref
:func:`classify_ref` for a plugin root that is already resolved (``plugin_real``).
src/skillevaluator/plugin_dependencies.py:628
↓ 2 callersFunction_claude_error_result
(text: str, **extra: Any)
tests/tier3/test_f90_mcp_execution.py:73
↓ 2 callersFunction_claude_flag
Whether Claude Code reads a frontmatter flag as on. A Claude Code 2.1.284 probe forced an output style for ``true``, ``"true"``, ``yes`` and
src/skillevaluator/plugin_context_cost.py:189
↓ 2 callersFunction_claude_read
(*, is_error: bool, content: str)
tests/tier3/test_failed_skill_reads.py:61
↓ 2 callersFunction_claude_root_vars
Rewrite a format's plugin-root placeholder to ``${CLAUDE_PLUGIN_ROOT}`` in a command handler.
src/skillevaluator/tier3/plugin_native.py:595
↓ 2 callersFunction_claude_rule_file_name
The staged user-rule file for one plugin rule: the full rule name, ending in ``.md``. Keeping the full name keeps ``style.md`` and ``style.mdc``
src/skillevaluator/tier3/plugin_native.py:1318
↓ 2 callersFunction_claude_user_rule
The staged Claude Code user rule for one plugin rule, or ``(None, reason)``. Claude Code loads a user rule on every task unless its frontmatter h
src/skillevaluator/tier3/plugin_native.py:1208
↓ 2 callersFunction_claude_version
Return the ``(major, minor)`` version of a Claude model ID, or ``None``.
src/skillevaluator/provider_config.py:154
↓ 2 callersFunction_claude_version
(model)
src/skillevaluator/tier3/harbor/templates/eval.py:1263
↓ 2 callersFunction_clean_load_failures
``{case: {arm: reason}}`` from either shape :data:`LoadFailures` accepts; an empty entry is no failure.
src/skillevaluator/tier3/harbor/stats.py:886
↓ 2 callersFunction_clean_lsp_servers
Bounded ``server -> file extensions`` (casefolded, each with its leading dot) of staged LSP servers.
src/skillevaluator/tier3/eval_core/plugin_signals.py:1494
↓ 2 callersFunction_clean_names
(values: Iterable[Any])
src/skillevaluator/tier3/eval_core/plugin_signals.py:1271
↓ 2 callersFunction_cleanup_fallback_backup
(backup: Path)
src/skillevaluator/tier3/harbor/secure_copy.py:1982
↓ 2 callersFunction_cli
( payload: dict, *, title: str = "Plugin static evidence", notice: str = "Static evidence does
src/skillevaluator/reporting/plugin_static_render.py:74
↓ 2 callersFunction_client_provider
(client: EmbeddingClient)
src/skillevaluator/embedding/registry.py:752
↓ 2 callersFunction_clone
(source, target, *options)
tests/ci/test_select_changed_skills.py:301
↓ 2 callersMethod_close
(self, *, after_failure: bool)
src/skillevaluator/tier3/harbor/progress.py:286
↓ 2 callersFunction_close_running_bridge
( running: RunningBridge, *, preserve_cancellation: bool = True, )
src/skillevaluator/tier3/harbor/local_agents.py:85
↓ 2 callersFunction_codex_args_key
Canonical JSON text of a Codex call's arguments (a JSON string or an object), bounded.
src/skillevaluator/tier3/plugin_trace_artifacts.py:113
↓ 2 callersMethod_codex_drops
Whether Codex ignores this declared path value (and loads the default location instead).
src/skillevaluator/plugin_discovery.py:318
↓ 2 callersFunction_codex_exec
(source: str, call_id: str = "exec-1")
tests/tier3/test_plugin_signals.py:96
↓ 2 callersFunction_codex_json_lines
Parsed JSON lines that mention MCP (bounded, no-follow read; unparsable lines are skipped).
src/skillevaluator/tier3/plugin_trace_artifacts.py:45
↓ 2 callersFunction_codex_patch_trajectory
(patch: str)
tests/tier3/test_behavior_evidence.py:1436
↓ 2 callersFunction_codex_server
(value: Any)
src/skillevaluator/tier3/plugin_trace_artifacts.py:38
↓ 2 callersFunction_codex_server_key
(server: str)
src/skillevaluator/tier3/plugin_trace_artifacts.py:208
↓ 2 callersFunction_codex_wrappers
(module: types.ModuleType)
tests/test_custom_execution_backends.py:508
↓ 2 callersFunction_collect_cost_limited
Collect attempt jobs a cost limit cut short; opencode's second case-a baseline attempt timed out. The limit admitted no attempt of case-b for ope
tests/tier3/test_baseline_timeouts.py:1353
↓ 2 callersFunction_collect_incomplete_scans
(results: list[ValidationResult])
src/skillevaluator/reporting/sarif_reporter.py:252
↓ 2 callersMethod_collect_license_files
Read every conventional license filename and collect all matches.
src/skillevaluator/validators/license.py:349
↓ 2 callersMethod_collect_statistics
(self, result: dict[str, Any], jobs_dir: Path)
src/skillevaluator/tier3/plugin_comparisons.py:347
↓ 2 callersMethod_collect_streamed_output
Stream output within one hard raw-byte budget.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:1354
↓ 2 callersFunction_collect_txt_deps
Collect non-comment lines from all instances of ``filename`` inside skill dirs.
src/skillevaluator/tier3/harbor/adapter.py:2502
↓ 2 callersFunction_collected_default_run
Collect a default run with its baseline from Harbor verifier output and build its Tier 3 result.
tests/golden/test_benchmark_card.py:1196
↓ 2 callersFunction_combine_continuation_trajectories
( base: dict[str, Any], continuation: dict[str, Any], *, state: _TrajectoryReferenceState, )
src/skillevaluator/tier3/harbor/collector.py:2384
↓ 2 callersFunction_combined
(statuses: list[str])
ci/run_tiers.py:310
↓ 2 callersFunction_combined_notes
(*values: Any)
src/skillevaluator/tier3/harbor/collector.py:2137
↓ 2 callersFunction_command_input_args
(command: list[str], cmd_idx: int, assignments: dict[str, str])
src/skillevaluator/tier3/eval_core/script_invocation.py:78
↓ 2 callersFunction_command_io
(command: Sequence[str])
src/skillevaluator/tier3/eval_core/plugin_signals.py:2435
↓ 2 callersFunction_command_line_switch
Whether ``token`` is a switch of command-line parser ``base`` that takes the rest as a command line.
src/skillevaluator/validators/mcp_static.py:544
↓ 2 callersFunction_command_looks_like_write
(command: str)
src/skillevaluator/tier3/eval_core/atif_helpers.py:428
↓ 2 callersFunction_command_names_script
Whether the expected script is named anywhere in this command's words. Inline code, a module, or text handed to ``eval`` can run the script w
src/skillevaluator/tier3/eval_core/script_invocation.py:1514
↓ 2 callersFunction_command_position_words
The words of a segment that stand where a command may: the first, and each that follows ``then``, ``do``, ``else``, ``elif``, ``!``, ``{`` or ``(`
src/skillevaluator/tier3/eval_core/script_invocation.py:1929
↓ 2 callersFunction_command_prefix_end
Skip the compound syntax in front of a command word. Returns the index of the command word, the number of ``case`` commands opened, and the n
src/skillevaluator/tier3/eval_core/checks.py:2587
↓ 2 callersFunction_command_reads_skill_md_arg
(command: list[str], cmd_idx: int, assignments: dict[str, str])
src/skillevaluator/tier3/eval_core/checks.py:1709
↓ 2 callersFunction_comment_line_numbers
Lines to skip as comments for the file's actual comment syntax.
src/skillevaluator/validators/security.py:443
↓ 2 callersFunction_commit_change
(consumer, *, write=(), rename=())
tests/ci/test_run_tiers.py:674
↓ 2 callersFunction_compact_dimensions
Summarize dimensions with each explanation masked before it is shortened. An explanation that is not text is dropped, so ``_summarize_dimensions`
src/skillevaluator/evaluation/tier3_feedback_chat.py:725
↓ 2 callersMethod_complete_messages
Send ``turns`` after an optional system prompt and return the response text.
src/skillevaluator/inference/client.py:717
↓ 2 callersMethod_complete_progress
(self, index, state, detail=None)
src/skillevaluator/reporting/console_ui.py:682
↓ 2 callersFunction_completion_counter_jobs
Return the launch errors of jobs whose timed-out baseline attempt reports ``count`` output tokens.
tests/tier3/test_baseline_timeouts.py:362
↓ 2 callersFunction_component_ident
( kind: str, name: str, fn: str, tool_label: str, *, persist: bool = True, declare
src/skillevaluator/tier3/eval_core/plugin_signals.py:2772
↓ 2 callersFunction_compose_interpolation_names
Extract Compose variable names while respecting ``$$`` escapes.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:559
↓ 2 callersFunction_compose_strings
(value: Any)
src/skillevaluator/tier3/harbor/adapter.py:3277
↓ 2 callersFunction_compound_after_assignment
Whether a compound's opening word follows an assignment in a command. Every modelled shell refuses the text before running the line that holds
src/skillevaluator/tier3/eval_core/script_invocation.py:1562
↓ 2 callersFunction_compound_keywords
Count the compound commands a simple command opens and closes. ``while read -r l``, ``do if true`` and the ``f() {`` or ``g() (`` of a functi
src/skillevaluator/tier3/eval_core/checks.py:2645
↓ 2 callersFunction_compute_bounded_retry_delay
Compute a bounded retry sleep duration and verify the judge deadline allows it.
src/skillevaluator/tier3/harbor/templates/eval.py:1791
↓ 2 callersFunction_compute_custom_lift
Compute lift for user-owned custom metrics.
src/skillevaluator/tier3/harbor/collector.py:5807
↓ 2 callersMethod_compute_top_issues
Compute cross-skill aggregation of top issues for the executive summary. For SECURITY findings, groups by category prefix (e.g., all
src/skillevaluator/reporting/html.py:2713
↓ 2 callersFunction_condition_has_mixed_metric_contracts
Prefer collector-owned contract truth while retaining legacy inference.
src/skillevaluator/evaluation/tier3_report.py:5141
↓ 2 callersFunction_conditional_approve_hook
(tmp_path: Path, matcher: str, condition: str)
tests/validators/test_plugin_component_risk.py:356
↓ 2 callersFunction_config_key_names_a_credential
Return whether the last part of config set key *key* names a credential.
src/skillevaluator/tier3/eval_core/checks.py:4735
↓ 2 callersFunction_configured_public_providers
()
src/skillevaluator/tier3/harbor/templates/eval.py:1474
↓ 2 callersFunction_configured_secret_values
(extra_secret_values=())
src/skillevaluator/tier3/harbor/templates/eval.py:1147
↓ 2 callersMethod_confined_project_env
(self, env: dict[str, str] | None = None)
src/skillevaluator/tier3/harbor/local_agents.py:564
↓ 2 callersFunction_constrain_native_html
(document: str, trusted_code: frozenset[str])
ci/native_reports.py:87
↓ 2 callersMethod_consume_path
(self, relative: Path)
src/skillevaluator/utils/secure_fs.py:1101
↓ 2 callersFunction_consume_validate_json_report
()
src/skillevaluator/cli.py:720
↓ 2 callersMethod_contain_main_container
Remove only this project's main container after an interrupted exec.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:1811
↓ 2 callersMethod_contain_sidecar_service
Stop only the target sidecar and retain its exact IDs for restart.
src/skillevaluator/tier3/harbor/secure_docker_environment.py:1869
↓ 2 callersFunction_contained_transfer_files
(root: Path)
tests/test_harbor_local_environment.py:496
↓ 2 callersFunction_contained_transfer_payload
()
tests/test_harbor_local_environment.py:481
↓ 2 callersFunction_contains_any_word
(text: str, words: Iterable[str])
src/skillevaluator/validators/quality_score.py:187
↓ 2 callersFunction_contains_credential_bearing_uri
Reject URI userinfo without parsing or rendering the supplied value.
src/skillevaluator/tier3/evals_config.py:544
↓ 2 callersFunction_contains_target_payload
(path: Path, target: Path)
src/skillevaluator/tier3/decoys.py:153
↓ 2 callersFunction_contains_term
Match a word or phrase without accepting it inside another word.
src/skillevaluator/validators/quality_score.py:171
↓ 2 callersFunction_contrast_ratio
(foreground: str, background: str)
tests/reporting/test_report_redesign.py:522
↓ 2 callersFunction_copy_metric_py
Copy the custom metric.py for Harbor dataset-level aggregation.
src/skillevaluator/tier3/harbor/adapter.py:6192
← previousnext →3,701–3,800 of 24,741, ranked by callers