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
↓ 2 callers
Method
spent
(self)
ci/report_views.py:288
↓ 2 callers
Function
stage
()
tests/test_harbor_runtime_skill_isolation.py:5260
↓ 2 callers
Method
stage_additional
( self, agent: str, emit_control: Callable[[str, Sequence[Path]], Sequence[Path]],
src/skillevaluator/tier3/plugin_comparisons.py:230
↓ 2 callers
Method
staged_name
The OpenCode file stem (and so the agent or command name) for a plugin component. OpenCode has no plugin namespace, so an agent named like a
src/skillevaluator/tier3/plugin_native.py:2094
↓ 2 callers
Function
stamp_publication_identity
Atomically stamp an already captured, producer-owned target identity.
src/skillevaluator/publication_identity.py:737
↓ 2 callers
Function
start_runner_with_endless_scans
Start the runner alone in its session, as Docker runs it, and wait for two endless scans.
tests/ci/test_run_tier1.py:1662
↓ 2 callers
Method
status
Canonical report status: ``passed``, ``failed``, or ``incomplete``.
src/skillevaluator/models/result.py:447
↓ 2 callers
Function
status_counts
Count rows per state, always including every state key.
src/skillevaluator/plugin_dependencies.py:590
↓ 2 callers
Method
stop
(self, delete: bool)
tests/tier3/test_harbor_local_agents.py:86
↓ 2 callers
Function
summarize_external_metrics
Reuse custom reward aggregation, keeping imported evidence out of canonical scores.
src/skillevaluator/tier3/external_metrics.py:237
↓ 2 callers
Function
synthetic_frame
(text: str, *, stop_reason: str, error: str | None = None)
tests/claude_refusal_fixtures.py:95
↓ 2 callers
Function
synthetic_trajectory_from_cursor_cli
Build a minimal trajectory from ``cursor-agent`` plain-text output (heuristic). Cursor CLI does not emit ATIF. We preserve the full transcript f
src/skillevaluator/tier3/eval_core/log_converters.py:191
↓ 2 callers
Method
take
Take out and return, in order, the items not taken yet whose spans overlap *start*:*end*.
src/skillevaluator/tier3/eval_core/checks.py:4528
↓ 2 callers
Function
take_credentials_at
(start: int)
src/skillevaluator/tier3/eval_core/checks.py:4461
↓ 2 callers
Method
terminate
(self)
tests/test_harbor_secure_docker_environment.py:3589
↓ 2 callers
Method
tick
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
tier1_unscanned_publication_paths
Return certified source paths that recognized Tier 1 runs skipped by name. A malformed record fails closed with a placeholder entry.
src/skillevaluator/reporting/base.py:2084
↓ 2 callers
Function
tier3_fix_rows
Tier 3 "How to improve" items as the tier's "what to fix" section.
src/skillevaluator/reporting/console_ui.py:449
↓ 2 callers
Function
to_claude_root
A string with each *pattern* placeholder rewritten to ``${CLAUDE_PLUGIN_ROOT}``; other values unchanged.
src/skillevaluator/tier3/plugin_native.py:493
↓ 2 callers
Method
to_dict
(self)
src/skillevaluator/plugin_component_risk.py:1134
↓ 2 callers
Method
to_dict
(self)
src/skillevaluator/plugin_context_cost.py:55
↓ 2 callers
Method
to_dict
Serialize for inclusion in reports (read-only summary).
src/skillevaluator/validators/policy.py:126
↓ 2 callers
Method
to_dict
(self)
src/skillevaluator/tier3/loose_parts.py:285
↓ 2 callers
Method
to_dict
(self)
src/skillevaluator/models/quality.py:149
↓ 2 callers
Method
toggleDataVisibility
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
token_efficiency_score
Soft-asymptote token score: ``1 / (1 + tokens / TOKEN_EFFICIENCY_HALF_LIFE)``.
src/skillevaluator/tier3/harbor/stats.py:1245
↓ 2 callers
Method
tooltipPosition
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
total_issue_count
Return the headline issue count: unique findings plus independent errors and warnings.
src/skillevaluator/reporting/finding_summary.py:315
↓ 2 callers
Function
trajectory
()
tests/tier3/test_shared_atif_staging.py:16
↓ 2 callers
Function
trial_coverage
Prefer per-condition counters; rows are only retained observations. Physical retry rows, preview limits and early-stop policy cannot establish
src/skillevaluator/reporting/presentation.py:180
↓ 2 callers
Function
trial_dimension_scores
Return one trial's score per dimension, skipping dimensions it cannot score. Dimension scores are linear in the metrics. A legacy reward without
src/skillevaluator/tier3/harbor/stats.py:335
↓ 2 callers
Function
trial_value
(trial: dict[str, Any], field: str)
src/skillevaluator/tier3/harbor/resource_usage.py:70
↓ 2 callers
Function
trials
(score: float)
tests/evaluation/test_tier3_feedback_chat.py:1083
↓ 2 callers
Function
trusted_observed_activation
Project collector-issued invocation evidence; never inspect trajectory or scores as proof.
src/skillevaluator/tier3/harbor/metrics.py:464
↓ 2 callers
Function
ui
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Method
unbindEvents
()
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
uncached_tokens
Return uncached prompt plus completion tokens, or ``None`` if unknown. Prompt counters include cached tokens (ATIF ``total_prompt_tokens`` and
src/skillevaluator/tier3/harbor/stats.py:1231
↓ 2 callers
Function
unrecognized_tier2_results
Return executed Tier 2 results that lack a recognized built-in producer. Optional Tier 2 evidence may be absent or cleanly skipped. A present,
src/skillevaluator/reporting/base.py:2108
↓ 2 callers
Function
unverified_groups
Unresolved refs grouped for reporting: ``[(cause, rows, suggestion), ...]``. Refs unresolved because the repository identity is unknown share one
src/skillevaluator/plugin_dependencies.py:906
↓ 2 callers
Method
update
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Method
update
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Method
updateControlPoints
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Method
updateSharedOptions
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Method
upload_file
(self, source_path: Path | str, target_path: str)
src/skillevaluator/tier3/harbor/local_environment.py:785
↓ 2 callers
Method
upload_file
(self, source_path: Path | str, target_path: str)
tests/test_harbor_secure_docker_environment.py:8425
↓ 2 callers
Function
v
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 2 callers
Function
va
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Method
validate
(target: Path, reports: Path)
tests/validators/test_security.py:10596
↓ 2 callers
Function
validate_environment
(environment: object, field: str, *, allowed_env: dict[str, str] | None)
src/skillevaluator/tier3/harbor/adapter.py:5192
↓ 2 callers
Function
validate_harbor_contract
Validate the BYOT Harbor task contract without mutating user files.
src/skillevaluator/tier3/evals_spec.py:454
↓ 2 callers
Function
validate_http_url
Reject local-file and custom URL schemes before making a request.
src/skillevaluator/tier3/reference_skills/api-caller/scripts/url_safety.py:9
↓ 2 callers
Function
validate_local_agents
Return agents that are not supported by Harbor local mode.
src/skillevaluator/tier3/harbor/local_runtime.py:202
↓ 2 callers
Function
validate_mcp_command
Append the stdio command findings for one server's ``command`` and ``args`` to ``findings``. Shell metacharacters where a shell would read them,
src/skillevaluator/validators/mcp_static.py:710
↓ 2 callers
Function
validate_mcp_pinning
Append the pinning finding for one server to ``findings``, when its package runner is not pinned. A HIGH ``mcp_command_floating_version`` for a m
src/skillevaluator/validators/mcp_static.py:721
↓ 2 callers
Function
validate_native_mcp_source
Apply the task MCP policy to declarations before native config projection. Claude's installation placeholders are client-owned paths, not referen
src/skillevaluator/tier3/harbor/native_staging.py:89
↓ 2 callers
Function
validate_parent
(target: Path, handle)
src/skillevaluator/ci_onboarding.py:334
↓ 2 callers
Function
validate_summary
(summary: dict)
ci/publish_reports.py:132
↓ 2 callers
Function
validate_verifier
(verifier: object, field: str)
src/skillevaluator/tier3/harbor/adapter.py:5149
↓ 2 callers
Method
value_end
Return where the value the assignment rule masks after the name *match* ends, or None when it masks none.
src/skillevaluator/tier3/eval_core/checks.py:4863
↓ 2 callers
Function
verdict_word
Return the report's verdict word for a run or skill status.
ci/native_reports.py:199
↓ 2 callers
Function
verifier_judge_env
Return ``(task verifier env templates, Gym process env)`` carrying the judge settings. Mirrors the Harbor runner's staging. Gym has no job-level
src/skillevaluator/tier3/nemo_gym/runner.py:338
↓ 2 callers
Function
verify_directory
( directory: int, parts: tuple[str, ...], named_metadata: os.stat_result | None = None
src/skillevaluator/tier3/harbor/secure_copy.py:1114
↓ 2 callers
Function
verify_receipt
Return signed identity claims only after signature and context validation.
ci/verify_provenance.py:109
↓ 2 callers
Function
version_finding
Advisory for a harness release other than the one the profile was verified on.
src/skillevaluator/harness/overlay.py:877
↓ 2 callers
Function
vo
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
vt
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
wait
(timeout=None)
tests/tier3/test_budget_reservation.py:26
↓ 2 callers
Method
wait
(self, timeout: float | None = None)
tests/test_harbor_runner_status.py:748
↓ 2 callers
Method
wait_closed
(self)
tests/test_secure_docker_environment.py:138
↓ 2 callers
Function
weighted_dimension_score
Return one dimension's weighted evaluator mean, or ``None`` when no evaluator is scored. *config* is a ``DIMENSION_MAPPING`` entry and *value_of*
src/skillevaluator/tier3/harbor/metrics.py:667
↓ 2 callers
Method
with_agent_import_path
(self, agent: str, base_import_path: str | None)
src/skillevaluator/tier3/plugin_execution.py:125
↓ 2 callers
Function
wrap_hook_handler
Wrap one command handler with the hook census logger. Shell form becomes ``/bin/sh /skilleval/hook_census.sh <hook_id> <event> -- '<original
src/skillevaluator/tier3/plugin_native.py:395
↓ 2 callers
Function
wrap_hook_sources
Merge every hooks source into one Claude Code ``hooks.json`` and wrap each command handler. Hook ids always use the source's own event names, so
src/skillevaluator/tier3/plugin_native.py:605
↓ 2 callers
Function
write
(name: str, body: str)
tests/test_quality_config.py:382
↓ 2 callers
Function
write
(name: str, description: str)
tests/harness/test_overlay.py:996
↓ 2 callers
Method
write
(self, text: str)
tests/test_tier3_progress.py:1236
↓ 2 callers
Function
write_attempt
Lay out one downloaded tier artifact as actions/download-artifact does for a pattern.
tests/ci/test_run_tiers.py:604
↓ 2 callers
Function
write_trial
(jobs, skill, arm="with", *, missing=False, steps=1)
tests/tier3/test_plugin_evidence.py:83
↓ 2 callers
Function
yaml_text
(value: dict)
src/skillevaluator/ci_onboarding.py:152
↓ 2 callers
Function
yn
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
za
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 2 callers
Function
zt
(t,e,i,s)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Aa
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
An
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Ao
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
At
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
B
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 1 callers
Function
Be
(t,e,i,s,n)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Bi
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Bs
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Bt
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Co
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
D
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:7
↓ 1 callers
Function
De
(t)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Di
(t,e,i)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Do
(t,e,i,s,n)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Ei
({start:t,end:e,count:i,loop:s,style:n})
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Fi
(t,e,i,s)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Fs
(t,e,i,s)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
↓ 1 callers
Function
Ge
(t,e)
src/skillevaluator/reporting/assets/chart.umd.min.js:13
← previous
next →
5,301–5,400 of 24,741, ranked by callers