MCPcopy Create free account

hub / github.com/NVIDIA/SkillEvaluator / functions

Functions24,741 in github.com/NVIDIA/SkillEvaluator

↓ 1,826 callersMethodget
(self, key: str, default: Any = None)
src/skillevaluator/harness/overlay.py:1000
↓ 1,396 callersMethodread_text
(self, **_kwargs)
tests/reporting/test_chart_dependency.py:288
↓ 886 callersFunctionset
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 780 callersMethodsplit
(self, *args: object, **kwargs: object)
tests/test_harbor_secure_docker_environment.py:4963
↓ 661 callersMethodexists
(path: Path)
tests/test_harbor_local_sandbox.py:358
↓ 612 callersMethodrun
(self, args: list[str], **kwargs: Any)
tests/validators/test_private_claude_parity.py:32
↓ 604 callersMethoditems
(self, value: object)
src/skillevaluator/reporting/telemetry_view.py:93
↓ 564 callersMethodreplace
(match: re.Match[str])
src/skillevaluator/tier3/harbor/local_agents.py:469
↓ 536 callersMethodrender_all
(self, results: list[ValidationResult])
tests/reporting/test_secure_report_output.py:31
↓ 516 callersMethodresolve
(path: Path, strict: bool = False)
tests/test_harbor_local_sandbox.py:352
↓ 443 callersMethodinvoke
(self, ctx: click.Context)
src/skillevaluator/tier_group.py:34
↓ 416 callersMethodget
(self, path)
tests/ci/test_feedback.py:103
↓ 414 callersMethodupdate
(self, renderable, *, refresh: bool = False)
tests/test_tier3_progress.py:55
↓ 381 callersMethodpop
(self, index: int = -1)
tests/test_harbor_secure_docker_environment.py:4955
↓ 358 callersMethodvalidate
(self, skill_path)
tests/validators/test_base.py:99
↓ 334 callersMethodis_file
(self)
tests/test_harbor_local_mode.py:5734
↓ 282 callersMethodsplitlines
(self, keepends=False)
tests/deduplication/utils/test_chunker.py:91
↓ 281 callersMethodget
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 274 callersMethodvalues
(self, label: str, value: object)
src/skillevaluator/reporting/native_evidence.py:148
↓ 273 callersMethoditems
(self)
tests/reporting/test_advisory_skip_verdict.py:1786
↓ 253 callersMethodupdate
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 248 callersMethodcount
(self, sub, start=None, end=None)
tests/deduplication/utils/test_chunker.py:86
↓ 245 callersMethodread_bytes
Read a bounded regular file without redirects; hard links require an explicit opt-in.
src/skillevaluator/utils/secure_fs.py:371
↓ 235 callersMethodsearch
(self, content: str, start: int = 0)
src/skillevaluator/validators/markdown.py:145
↓ 223 callersMethodmatch
Return where group 1 and the match of *pattern* end in *text* when read at *position* (None: no match). *pattern* is read in the masked text,
src/skillevaluator/tier3/eval_core/checks.py:4568
↓ 210 callersMethodadd_success
Add a success detail. Args: check_name: Identifier for the check message: Human-readable description **me
src/skillevaluator/models/result.py:516
↓ 210 callersMethodis_dir
(self, *, follow_symlinks=True)
tests/utils/test_secure_fs.py:308
↓ 197 callersMethodvalues
(self)
tests/reporting/test_advisory_skip_verdict.py:1789
↓ 196 callersMethodindex
(self, sub, start=None, end=None)
tests/deduplication/utils/test_chunker.py:80
↓ 184 callersFunction_write_projection_fixture
(tmp_path: Path)
tests/test_harbor_runtime_skill_isolation.py:104
↓ 171 callersMethodglob
The compiled glob, as :func:`fnmatch.fnmatchcase` compiles it.
src/skillevaluator/tier3/eval_core/plugin_signals.py:3430
↓ 169 callersMethodadd
(self, where: str, problem: str)
src/skillevaluator/tier3/eval_core/plugin_signals.py:629
↓ 166 callersFunctionbuild_agent_eval_payload
Assemble the canonical Tier 3 ``agent_eval`` payload from loaded agent data. ``agents`` is the structure produced by :func:`skillevaluator.ti
src/skillevaluator/evaluation/tier3_report.py:773
↓ 156 callersMethodadd_finding
Add a structured finding and update summary. Supports multiple calling conventions for backward compatibility: 1. New style: add_find
src/skillevaluator/models/result.py:453
↓ 153 callersMethodsearch
(self, *args)
tests/validators/test_mcp_static.py:1320
↓ 151 callersFunctiongenerate_harbor_tasks
Generate tasks from one private evals snapshot, then publish exactly.
src/skillevaluator/tier3/harbor/adapter.py:6737
↓ 151 callersFunctionpatch
(path, payload)
tests/ci/test_benchmark_publish.py:407
↓ 149 callersMethodread_text
Bounded, anchored, no-follow UTF-8 read; raises :class:`SecurePathError`. ``config`` charges the read to the separate config budget.
src/skillevaluator/plugin_paths.py:193
↓ 148 callersMethodfail
(self, reason: str)
src/skillevaluator/tier3/mcp_proof.py:566
↓ 148 callersMethodstat
(self, *, follow_symlinks=True)
tests/utils/test_secure_fs.py:305
↓ 141 callersMethodget
(self, path)
ci/feedback.py:622
↓ 134 callersMethodadd_error
Add a legacy error string and mark validation as failed. For backward compatibility with existing validators.
src/skillevaluator/models/result.py:662
↓ 132 callersMethodadd
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 127 callersMethodis_symlink
(path: Path)
tests/test_harbor_local_sandbox.py:361
↓ 126 callersFunction_validate
Run one captured report through validate_security_only.
tests/validators/test_skillspector_hook_surface.py:285
↓ 121 callersMethodvalidate_pii_only
Run only PII detection (no skillspector security scan).
src/skillevaluator/validators/security.py:3199
↓ 118 callersFunctionvalidate_mcp_server_declaration
Statically validate one contained ``mcpServers`` entry (``name`` -> config). ``allowed_private_hosts`` comes from the validation policy (``mc
src/skillevaluator/validators/mcp_static.py:3582
↓ 116 callersFunction_validate_skillspector_payload
Run one deterministic SkillSpector payload through the public validation seam.
tests/validators/test_security.py:219
↓ 111 callersFunctioncheck
(*args, **kwargs)
tests/tier3/test_error_recovery.py:23
↓ 111 callersMethodfinditer
(self, *args)
tests/validators/test_mcp_static.py:1316
↓ 109 callersMethodrender
(self, result: ValidationResult)
tests/reporting/test_secure_report_output.py:27
↓ 108 callersMethodset
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 107 callersFunction_valid_benchmark
()
tests/test_public_benchmark_gate.py:76
↓ 103 callersFunction_initialized_secure_docker_environment
( tmp_path: Path, *, persistent_env: dict[str, str] | None = None, )
tests/test_harbor_secure_docker_environment.py:83
↓ 102 callersMethodopen
(self, bracket: str)
src/skillevaluator/utils/structured_data.py:360
↓ 100 callersFunction_bash
(command: str, observation: str = "Exit code 0")
tests/tier3/test_script_invocation_evidence.py:40
↓ 96 callersFunctioncollect_harbor_results
Collect results from Harbor jobs into evals/results/<agent>/ structure. Returns a dict with per-agent scores, lift, and a cross-agent comparison.
src/skillevaluator/tier3/harbor/collector.py:7581
↓ 96 callersFunctionrender_result
Return the stable non-color rendering used by tests and API callers.
src/skillevaluator/tier3/result_display.py:1461
↓ 95 callersMethodobject
(self, value: object)
src/skillevaluator/utils/secret_redaction.py:155
↓ 93 callersMethodopen
(self, request, timeout)
tests/ci/test_github_api.py:40
↓ 92 callersFunction_checks
(findings)
tests/validators/test_mcp_static.py:29
↓ 90 callersFunction_agent
(with_cases: dict, without_cases: dict)
tests/test_tier3_case_summary.py:27
↓ 90 callersFunction_skillspector_json_report
Return the pinned SkillSpector JSON report shape used by contract tests.
tests/validators/test_security.py:129
↓ 88 callersMethodclear
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 88 callersMethodmark_scan_incomplete
Record that an external scanner did not produce complete trustworthy results. Partial, missing, timed-out, crashed, and malformed scanner run
src/skillevaluator/models/result.py:694
↓ 88 callersMethodstart
(self, *, refresh: bool = False)
tests/test_tier3_progress.py:51
↓ 87 callersFunctionstage_native_harbor_tasks
Stage native tasks privately, then publish one exact output snapshot.
src/skillevaluator/tier3/harbor/adapter.py:6045
↓ 85 callersFunction_trajectory
(source: str, observation: str = "outer observation")
tests/tier3/test_codex_tool_call_normalization.py:47
↓ 85 callersMethodadd_warning
Add a legacy warning string. For backward compatibility with existing validators.
src/skillevaluator/models/result.py:671
↓ 81 callersMethodformat
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 79 callersFunction_case
(*passed: bool, missing: int = 0, prefix: str = "t")
tests/test_tier3_case_summary.py:17
↓ 79 callersFunction_traj
(*steps: dict[str, Any], prompt: str = "Please help.")
tests/tier3/test_plugin_signals.py:59
↓ 79 callersMethodget
(self, key, *default)
tests/reporting/test_private_static_reports.py:477
↓ 77 callersFunctionagent_eval_result_from_directory
Build the canonical ``AGENT_EVAL`` result for one explicit Harbor run. Read-only consumers that never embed the payload pass ``enforce_embedd
src/skillevaluator/evaluation/tier3_report.py:503
↓ 75 callersFunction_collect
( tmp_path: Path, *, skip_baseline: bool, case_ids: list[str], pass_threshold: float = 0.5
tests/test_issue55_collector_truth.py:160
↓ 75 callersMethodwrite
(self, data: Any)
tests/test_tier3_progress.py:1830
↓ 74 callersFunction_scanner
()
tests/test_oss_boundary.py:27
↓ 73 callersFunction_write
(root: Path, relative: str, text: str = _INJECTION)
tests/validators/test_skillspector_scope.py:131
↓ 73 callersMethodexpanduser
(self)
tests/test_harbor_local_mode.py:5731
↓ 71 callersMethodprint
Print single result directly to console.
src/skillevaluator/reporting/cli.py:177
↓ 70 callersFunction_one
(fn: str, args: dict[str, Any] | None = None, content: Any = "ok", call_id: str = "c1", **extra: Any)
tests/tier3/test_plugin_signals.py:63
↓ 70 callersFunction_progress_module
()
tests/test_tier3_progress.py:64
↓ 70 callersMethodadd
(self, keys: Iterable[str], record: HookRecord)
src/skillevaluator/plugin_component_risk.py:2900
↓ 69 callersFunction_write_minimal_native_task
(target: Path)
tests/test_harbor_runtime_skill_isolation.py:93
↓ 68 callersFunctionbuild_harbor_run_command
Build a Harbor invocation for a built-in or backend environment, or local mode.
src/skillevaluator/tier3/harbor/runner.py:1886
↓ 66 callersMethodevaluate
(cls, expression: str, names: dict[str, Any])
src/skillevaluator/harness/overlay.py:792
↓ 65 callersFunctionwhat_to_fix
Rank failed test cases across agents by attempts lost and attach one fix each.
src/skillevaluator/tier3/case_summary.py:1427
↓ 64 callersMethodfind
(self, x: int)
src/skillevaluator/deduplication/intra_skill/semantic_clustering.py:45
↓ 63 callersFunction_stub_runner
( monkeypatch: pytest.MonkeyPatch, tmp_path: Path, *, run_agent=None, collect=None, ht
tests/test_tier3_progress.py:2847
↓ 62 callersFunction_issue
(file_path: object, *, rule_id: str = "P1", category: str = "Prompt Injection")
tests/validators/test_skillspector_scope.py:37
↓ 62 callersFunction_write_issue_skill
Write a complete guide skill for issue #30 scoring regressions.
tests/validators/test_quality_score.py:97
↓ 62 callersFunction_write_skill
(root: Path, files: dict[str, str])
tests/validators/test_security.py:9308
↓ 61 callersFunction_default_reward
(entry_id: str, score: float)
tests/test_issue55_collector_truth.py:75
↓ 61 callersFunctionresolve_llm_provider
Resolve the public provider used for LLM-backed checks and judging.
src/skillevaluator/provider_config.py:534
↓ 61 callersMethodstart
(self, plan: Tier3RunPlan)
src/skillevaluator/tier3/harbor/progress.py:139
↓ 60 callersFunction_trajectory
( session_id: str, labels: tuple[str, ...], *, prompt_tokens: int, completion_tokens: int,
tests/test_harbor_collector_multistep_trajectory.py:29
↓ 59 callersFunction_write_complete_job_result
(job_dir: Path, trial_names: list[str])
tests/test_issue55_collector_truth.py:48
↓ 59 callersFunctioncollect_files
Collect bounded text files through a descriptor-anchored root.
src/skillevaluator/deduplication/utils/skill_collector.py:420
↓ 59 callersMethodvalidate_security_only
(self, _target: Path)
tests/validators/test_skillspector_scope.py:1795
↓ 58 callersFunction_complete_tier3_result
( marker: str, *, score: float, runtime_seconds: float, evaluated_at: str = "2026-08-24T12
tests/reporting/test_advisory_skip_verdict.py:107
next →1–100 of 24,741, ranked by callers