forked from cleveragents/cleveragents-core
d59fa47fd0
Add the missing plan prompt CLI command and wire it through the local A2A facade into PlanLifecycleService so operator guidance can be injected into active execute-phase plans per the specification. The lifecycle flow now validates active state, records user-intervention decisions, and returns structured queue/decision metadata for rich and machine-readable output envelopes across all supported formats. Also stabilize flaky quality gates discovered while implementing #885 by hardening integration helper timeouts, relaxing an overly strict CLI-core timing assertion, defaulting test worker concurrency to serial for determinism, switching ASV to spawn launch mode for runner stability, and documenting the controlled transform sandbox exec/compile path for static security hooks. ISSUES CLOSED: #885 Co-authored-by: Brent E. Edwards <brent.edwards@cleverthis.com> Co-committed-by: Brent E. Edwards <brent.edwards@cleverthis.com>
182 lines
6.3 KiB
Python
182 lines
6.3 KiB
Python
"""Step definitions for the ``agents plan prompt`` CLI command."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
from unittest.mock import MagicMock, patch
|
|
|
|
from behave import given, then, when
|
|
from typer.testing import CliRunner
|
|
|
|
from cleveragents.a2a.facade import A2aLocalFacade
|
|
from cleveragents.a2a.models import A2aRequest
|
|
from cleveragents.cli.commands.plan import app as plan_app
|
|
from cleveragents.core.exceptions import PlanError
|
|
|
|
|
|
@when("I run plan command help")
|
|
def step_run_plan_help(context) -> None:
|
|
runner = CliRunner()
|
|
context.prompt_result = runner.invoke(plan_app, ["--help"])
|
|
|
|
|
|
@then("help output should include plan prompt command")
|
|
def step_help_includes_prompt(context) -> None:
|
|
assert context.prompt_result.exit_code == 0, context.prompt_result.output
|
|
assert "prompt" in context.prompt_result.output
|
|
|
|
|
|
@given("a mocked lifecycle service prompt response")
|
|
def step_mock_lifecycle_prompt_response(context) -> None:
|
|
context.prompt_service_mock = MagicMock()
|
|
context.prompt_service_mock.prompt_plan.return_value = {
|
|
"guidance_added": {
|
|
"plan": "01HXM8C2ZK4Q7C2B3F2R4VYV6J",
|
|
"guidance": "Use mocks for database tests",
|
|
"scope": "next execution step",
|
|
"phase": "execute",
|
|
"state_transition": "errored -> processing",
|
|
},
|
|
"decision_created": {
|
|
"type": "user_intervention",
|
|
"id": "01HXM9C5G7R2X8S3K4Z5Q8R6Y3",
|
|
"parent": "01HXM9A1C2Q7W3R5G8Z0P4Q1X9",
|
|
},
|
|
"queue": {"pending": 1, "applied": 0},
|
|
}
|
|
|
|
|
|
@given("lifecycle service rejects prompt for inactive plan phase")
|
|
def step_mock_lifecycle_prompt_reject(context) -> None:
|
|
context.prompt_service_mock = MagicMock()
|
|
context.prompt_service_mock.prompt_plan.side_effect = PlanError(
|
|
"Plan is not in active execute phase"
|
|
)
|
|
|
|
|
|
@when(
|
|
'I run plan prompt with plan id "{plan_id}" and guidance "{guidance}" in format "{fmt}"'
|
|
)
|
|
def step_run_plan_prompt(context, plan_id: str, guidance: str, fmt: str) -> None:
|
|
runner = CliRunner()
|
|
with patch(
|
|
"cleveragents.cli.commands.plan._get_lifecycle_service",
|
|
return_value=context.prompt_service_mock,
|
|
):
|
|
context.prompt_result = runner.invoke(
|
|
plan_app,
|
|
["prompt", plan_id, guidance, "--format", fmt],
|
|
)
|
|
|
|
|
|
@then("the prompt command should succeed")
|
|
def step_prompt_success(context) -> None:
|
|
assert context.prompt_result.exit_code == 0, context.prompt_result.output
|
|
|
|
|
|
@then("the prompt command should abort")
|
|
def step_prompt_abort(context) -> None:
|
|
assert context.prompt_result.exit_code != 0
|
|
|
|
|
|
@then(
|
|
'lifecycle prompt should be called with plan id "{plan_id}" and guidance "{guidance}"'
|
|
)
|
|
def step_prompt_called(context, plan_id: str, guidance: str) -> None:
|
|
context.prompt_service_mock.prompt_plan.assert_called_once_with(plan_id, guidance)
|
|
|
|
|
|
def _json_output(result_output: str) -> dict[str, object]:
|
|
return json.loads(result_output.strip())
|
|
|
|
|
|
@then('prompt output envelope should contain command "{command}"')
|
|
def step_prompt_envelope_command(context, command: str) -> None:
|
|
payload = _json_output(context.prompt_result.output)
|
|
assert payload.get("command") == command
|
|
|
|
|
|
@then('prompt output envelope should contain status "{status}"')
|
|
def step_prompt_envelope_status(context, status: str) -> None:
|
|
payload = _json_output(context.prompt_result.output)
|
|
assert payload.get("status") == status
|
|
|
|
|
|
@then("prompt output data should include queued guidance")
|
|
def step_prompt_envelope_data(context) -> None:
|
|
payload = _json_output(context.prompt_result.output)
|
|
data = payload.get("data")
|
|
assert isinstance(data, dict)
|
|
guidance_added = data.get("guidance_added")
|
|
assert isinstance(guidance_added, dict)
|
|
assert guidance_added.get("guidance") == "Use mocks for database tests"
|
|
queue = data.get("queue")
|
|
assert isinstance(queue, dict)
|
|
assert queue.get("pending") == 1
|
|
|
|
|
|
@then('prompt output should include guidance text "{guidance}"')
|
|
def step_prompt_output_contains_guidance(context, guidance: str) -> None:
|
|
normalized_output = " ".join(context.prompt_result.output.split())
|
|
normalized_guidance = " ".join(guidance.split())
|
|
assert normalized_guidance in normalized_output
|
|
|
|
|
|
@then("prompt output should mention inactive execute phase")
|
|
def step_prompt_output_inactive_phase(context) -> None:
|
|
assert "active execute phase" in context.prompt_result.output.lower()
|
|
|
|
|
|
@given("an A2A facade with a lifecycle prompt service")
|
|
def step_facade_with_prompt_service(context) -> None:
|
|
class _PromptLifecycleService:
|
|
def prompt_plan(self, plan_id: str, guidance: str) -> dict[str, object]:
|
|
return {
|
|
"guidance_added": {
|
|
"plan": plan_id,
|
|
"guidance": guidance,
|
|
"scope": "next execution step",
|
|
"phase": "execute",
|
|
"state_transition": "errored -> processing",
|
|
},
|
|
"decision_created": {
|
|
"type": "user_intervention",
|
|
"id": "01HXM9C5G7R2X8S3K4Z5Q8R6Y3",
|
|
"parent": "01HXM9A1C2Q7W3R5G8Z0P4Q1X9",
|
|
},
|
|
"queue": {"pending": 1, "applied": 0},
|
|
}
|
|
|
|
context.prompt_facade = A2aLocalFacade(
|
|
{"plan_lifecycle_service": _PromptLifecycleService()}
|
|
)
|
|
|
|
|
|
@when(
|
|
'I dispatch _cleveragents plan prompt for plan "{plan_id}" and guidance "{guidance}"'
|
|
)
|
|
def step_dispatch_facade_prompt(context, plan_id: str, guidance: str) -> None:
|
|
request = A2aRequest(
|
|
operation="_cleveragents/plan/prompt",
|
|
params={"plan_id": plan_id, "guidance": guidance},
|
|
)
|
|
context.facade_prompt_response = context.prompt_facade.dispatch(request)
|
|
|
|
|
|
@then("facade prompt response should not be a stub")
|
|
def step_facade_prompt_not_stub(context) -> None:
|
|
data = context.facade_prompt_response.data
|
|
assert data.get("stub") is not True
|
|
|
|
|
|
@then('facade prompt response should contain plan id "{plan_id}"')
|
|
def step_facade_prompt_plan(context, plan_id: str) -> None:
|
|
data = context.facade_prompt_response.data
|
|
assert data.get("plan_id") == plan_id
|
|
|
|
|
|
@then('facade prompt response should contain guidance "{guidance}"')
|
|
def step_facade_prompt_guidance(context, guidance: str) -> None:
|
|
data = context.facade_prompt_response.data
|
|
assert data.get("guidance") == guidance
|