vis-agent 0.2.38.dev7565__tar.gz → 0.2.38.dev7570__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/PKG-INFO +1 -1
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/pyproject.toml +1 -1
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/evals/__init__.py +5 -3
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/evals/trajectory.py +8 -6
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_evals_trajectory.py +2 -2
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/.gitignore +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/LICENSE +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/README.md +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-arguments.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-cases.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-copy.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-execution.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-groups.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-read-session.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-reads.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-results.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-tables.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/diff.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/live-primitives.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/session-health.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/view.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/activity.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/agents.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/automations.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/common.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/config.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/content.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/council.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/diff.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/gateway.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/improve.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/provider.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/rooms.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/symbol.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/toggle.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/schema/view.json +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/examples/greeter/tests/test_greeter.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/hatch_build.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/__init__.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/_contracts.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/_outside.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/_wire.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/activity.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/__init__.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/_agent.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/_agents.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/_client.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/_council.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/_extensions.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/_local.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/evals/harbor.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/evals/stats.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/extension.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/extension_package.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/py.typed +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/rooms.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/views.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/conftest.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_activity.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_agent.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_agent_extension_lifecycle.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_agent_extensions.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_agents.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_api_guides.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_authoring.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_catalog.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_catalog_recipe.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_ci_report_recipe.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_client.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_client_extensions_engine.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_code_quality_recipe.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_contracts.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_council.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_declarations.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_engine.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_engine_docstrings.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_evals_harbor.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_evals_stats.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_extension_package.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_extension_reference.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_gateway_performance.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_goals.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_live_primitives.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_live_wait.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_local.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_method_typing.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_monitor_recipe.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_namespace.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_outside.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_providers.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_record_arguments.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_registration.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_repair_hooks.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_result_sequence.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_rooms.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_scoped_settings.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_sdk_bootstrap.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_sdk_guide.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_sdk_typing.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_structured_live.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_symbol_contract.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_testing.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_text_list_arguments.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_tools.py +0 -0
- {vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_views.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.5
|
|
2
2
|
Name: vis-agent
|
|
3
|
-
Version: 0.2.38.
|
|
3
|
+
Version: 0.2.38.dev7570
|
|
4
4
|
Summary: The Vis Python SDK — author extensions and control local or remote engines.
|
|
5
5
|
Project-URL: Homepage, https://vis.blockether.com
|
|
6
6
|
Project-URL: Documentation, https://vis.blockether.com
|
|
@@ -8,7 +8,7 @@ build-backend = "hatchling.build"
|
|
|
8
8
|
|
|
9
9
|
[project]
|
|
10
10
|
name = "vis-agent"
|
|
11
|
-
version = "0.2.38.
|
|
11
|
+
version = "0.2.38.dev7570"
|
|
12
12
|
description = "The Vis Python SDK — author extensions and control local or remote engines."
|
|
13
13
|
readme = "README.md"
|
|
14
14
|
requires-python = ">=3.11"
|
|
@@ -1,8 +1,10 @@
|
|
|
1
1
|
"""Evaluation tools: read Vis runs and Harbor jobs, and compare their results.
|
|
2
2
|
|
|
3
|
-
|
|
4
|
-
|
|
5
|
-
`
|
|
3
|
+
Read one Vis run with `load_trajectory` or `parse_trajectory`. Read the trials of a
|
|
4
|
+
Harbor job or round with `read_trials`, then count them with `case_results` or
|
|
5
|
+
`summarize`. Seal a finished round with `seal_round`. Compare paired results with
|
|
6
|
+
`paired_change` and `mcnemar_exact`. For installation and sessions, use the
|
|
7
|
+
[Python SDK guide](https://vis.blockether.com/python-sdk.html).
|
|
6
8
|
"""
|
|
7
9
|
|
|
8
10
|
from .harbor import (
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/evals/trajectory.py
RENAMED
|
@@ -486,8 +486,8 @@ class Trajectory:
|
|
|
486
486
|
"""One Vis run read from an ATIF trajectory.
|
|
487
487
|
|
|
488
488
|
The typed fields hold what evaluations check. `document` keeps the whole input,
|
|
489
|
-
also the keys that this reader does not know. Child runs
|
|
490
|
-
|
|
489
|
+
also the keys that this reader does not know. Child runs are in
|
|
490
|
+
`subagent_trajectories`.
|
|
491
491
|
"""
|
|
492
492
|
|
|
493
493
|
schema_version: str
|
|
@@ -497,7 +497,7 @@ class Trajectory:
|
|
|
497
497
|
steps: tuple[Step, ...]
|
|
498
498
|
final_metrics: TrajectoryMetrics | None
|
|
499
499
|
extra: Mapping[str, Any]
|
|
500
|
-
|
|
500
|
+
subagent_trajectories: tuple[Trajectory, ...]
|
|
501
501
|
document: Mapping[str, Any]
|
|
502
502
|
|
|
503
503
|
@property
|
|
@@ -603,7 +603,7 @@ class Trajectory:
|
|
|
603
603
|
operation: Keep only calls of this function, for example `grep`.
|
|
604
604
|
state: Keep only calls with this outcome, for example `failed`.
|
|
605
605
|
|
|
606
|
-
The calls of
|
|
606
|
+
The calls of child runs are in `subagent_trajectories`, not here.
|
|
607
607
|
"""
|
|
608
608
|
return tuple(
|
|
609
609
|
call
|
|
@@ -633,7 +633,9 @@ class Trajectory:
|
|
|
633
633
|
self.final_metrics.to_dict() if self.final_metrics is not None else None
|
|
634
634
|
),
|
|
635
635
|
"steps": [step.to_dict() for step in self.steps],
|
|
636
|
-
"
|
|
636
|
+
"subagent_trajectories": [
|
|
637
|
+
child.to_dict() for child in self.subagent_trajectories
|
|
638
|
+
],
|
|
637
639
|
}
|
|
638
640
|
|
|
639
641
|
|
|
@@ -739,7 +741,7 @@ def _trajectory(
|
|
|
739
741
|
steps=tuple(_step(step, index, strict) for index, step in enumerate(steps)),
|
|
740
742
|
final_metrics=TrajectoryMetrics._from_atif(document.get("final_metrics")),
|
|
741
743
|
extra=_map(document.get("extra")),
|
|
742
|
-
|
|
744
|
+
subagent_trajectories=tuple(
|
|
743
745
|
_trajectory(child, strict, version)
|
|
744
746
|
for child in (children if isinstance(children, tuple) else ())
|
|
745
747
|
if isinstance(child, Mapping)
|
|
@@ -302,9 +302,9 @@ def test_a_step_without_metrics_or_blocks_has_empty_values():
|
|
|
302
302
|
assert final.function_calls == () and final.is_final
|
|
303
303
|
|
|
304
304
|
|
|
305
|
-
def
|
|
305
|
+
def test_child_runs_are_separate_trajectories():
|
|
306
306
|
run = parse_trajectory(_document())
|
|
307
|
-
(child,) = run.
|
|
307
|
+
(child,) = run.subagent_trajectories
|
|
308
308
|
assert (child.session_id, child.schema_version, child.status) == (
|
|
309
309
|
"child-1",
|
|
310
310
|
"ATIF-v1.8",
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-arguments.json
RENAMED
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-cases.json
RENAMED
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-execution.json
RENAMED
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-groups.json
RENAMED
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-read-session.json
RENAMED
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-reads.json
RENAMED
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-results.json
RENAMED
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/activity-tables.json
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/live-primitives.json
RENAMED
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/contracts/fixtures/session-health.json
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/examples/greeter/tests/test_greeter.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/engine/_extensions.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/src/blockether/vis/extension_package.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_agent_extension_lifecycle.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{vis_agent-0.2.38.dev7565 → vis_agent-0.2.38.dev7570}/tests/test_client_extensions_engine.py
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|