mirror of
https://github.com/usestrix/strix.git
synced 2026-08-24 20:02:39 +02:00
oyasumiandClaude Opus 4.8
ccbd8c7b58
Engine + integration: - Reviewer inspection now surfaces the real frozen source of an already-frozen workspace script/dependency instead of an empty string, so workspace-resident scripts resolve without a needless human defer. - Guard effectful static tools via an explicit, documented set plus the SDK's per-tool needs_approval signal; give the exec/stdin wrappers the same idempotency guard as their sibling wrappers. - Centralize DEFAULT_SAFETY_MODE and share one resume safety-mode rule between the CLI and runner so the two cannot drift; type InspectionContext.runner, reuse RUNTIME_STATE_DIR_NAME, and drop a dead workdir parameter and a write-only field. TUI approval experience: - Approve All drops the run into dangerous mode: it approves the pending call and turns review off for the rest of the run, with a standing "review off" status flag. - The status row shows the owning agent as paused while it waits on a decision. - Redesigned prompt: a risk + tool header, a collapsible command/reason preview that expands (e) and scrolls, and no internal digest, agent, or request ids. Full Python (1138) and Go suites, ruff, and mypy strix/ pass. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
719 lines
24 KiB
Python
719 lines
24 KiB
Python
from __future__ import annotations
|
|
|
|
import argparse
|
|
import asyncio
|
|
import os
|
|
from pathlib import Path
|
|
from types import SimpleNamespace
|
|
|
|
import pytest
|
|
|
|
from strix.config import apply_config_override, loader
|
|
from strix.config.settings import DEFAULT_MAX_TURNS
|
|
from strix.interface.tui.backend.controller import TuiController
|
|
|
|
|
|
def args() -> argparse.Namespace:
|
|
return argparse.Namespace(
|
|
needs_setup=True,
|
|
targets_info=[],
|
|
instruction=None,
|
|
scan_mode="deep",
|
|
max_budget_usd=None,
|
|
max_turns=DEFAULT_MAX_TURNS,
|
|
scope_mode="auto",
|
|
diff_base=None,
|
|
local_sources=[],
|
|
diff_scope={"active": False},
|
|
user_explicit_instruction=None,
|
|
run_name=None,
|
|
)
|
|
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def isolated_config(tmp_path: Path) -> None:
|
|
for key in (
|
|
"STRIX_LLM",
|
|
"OPENAI_API_KEY",
|
|
"ANTHROPIC_API_KEY",
|
|
"LLM_API_KEY",
|
|
"LLM_API_BASE",
|
|
"AZURE_API_KEY",
|
|
"AZURE_API_BASE",
|
|
"AZURE_API_VERSION",
|
|
):
|
|
os.environ.pop(key, None)
|
|
apply_config_override(tmp_path / "config.json")
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_setup_state_is_serializable() -> None:
|
|
controller = TuiController(args())
|
|
await controller.handle("setup.add_target", {"target": "https://example.com"})
|
|
await controller.handle("setup.set_instruction", {"instruction": "focus on auth"})
|
|
snapshot = controller.snapshot()
|
|
assert snapshot["targets"] == ["https://example.com"]
|
|
assert snapshot["instruction"] == "focus on auth"
|
|
assert snapshot["scan_state"] == "setup"
|
|
assert snapshot["scan_mode"] == "deep"
|
|
assert snapshot["max_budget_usd"] is None
|
|
assert snapshot["max_turns"] == 500
|
|
assert snapshot["scope_mode"] == "auto"
|
|
assert snapshot["diff_base"] is None
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_setup_instruction_starts_from_cli_and_can_be_cleared() -> None:
|
|
setup_args = args()
|
|
setup_args.instruction = " CLI instruction "
|
|
controller = TuiController(setup_args)
|
|
|
|
assert controller.snapshot()["instruction"] == "CLI instruction"
|
|
|
|
result = await controller.handle("setup.set_instruction", {"instruction": ""})
|
|
|
|
assert result == {"instruction": ""}
|
|
assert controller.snapshot()["instruction"] == ""
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_setup_controls_reject_changes_after_start() -> None:
|
|
controller = TuiController(args())
|
|
controller.setup_mode = False
|
|
controller.scan_started = True
|
|
|
|
with pytest.raises(RuntimeError, match="can no longer be changed"):
|
|
await controller.handle("setup.add_target", {"target": "https://example.com"})
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_large_target_list_reports_truncated_snapshot_count() -> None:
|
|
controller = TuiController(args())
|
|
|
|
for index in range(20):
|
|
await controller.handle("setup.add_target", {"target": f"https://target-{index}.example"})
|
|
added = await controller.handle("setup.add_target", {"target": "https://last.example"})
|
|
snapshot = controller.snapshot()
|
|
|
|
assert added == {"target": "https://last.example", "total": 21}
|
|
assert snapshot["target_count"] == 21
|
|
# The snapshot only carries a bounded prefix of the list.
|
|
assert len(snapshot["targets"]) == 16
|
|
|
|
|
|
def test_state_populates_model_warning_for_non_frontier_model() -> None:
|
|
os.environ["STRIX_LLM"] = "openai/gpt-3.5-turbo"
|
|
loader._cached = None
|
|
|
|
warning = TuiController(args()).snapshot()["model_warning"]
|
|
|
|
assert "openai/gpt-3.5-turbo" in warning
|
|
assert "not a recommended frontier model" in warning
|
|
|
|
|
|
def test_setup_restores_prepared_cli_targets() -> None:
|
|
setup_args = args()
|
|
setup_args.targets_info = [
|
|
{"type": "web", "details": {}, "original": "https://example.com"},
|
|
{"type": "local_code", "details": {}, "original": "/workspace/source"},
|
|
]
|
|
|
|
controller = TuiController(setup_args)
|
|
|
|
assert controller.snapshot()["targets"] == ["https://example.com", "/workspace/source"]
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_start_validates_model_before_callback() -> None:
|
|
started = False
|
|
|
|
async def start(_verify: bool = True) -> None:
|
|
nonlocal started
|
|
started = True
|
|
|
|
controller = TuiController(args(), on_start=start)
|
|
await controller.handle("setup.add_target", {"target": "https://example.com"})
|
|
with pytest.raises(ValueError, match="No model configured"):
|
|
await controller.handle("setup.start", {})
|
|
assert started is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_start_launches_with_a_configured_model() -> None:
|
|
started = False
|
|
|
|
async def start(_verify: bool = True) -> None:
|
|
nonlocal started
|
|
started = True
|
|
|
|
os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
|
|
loader._cached = None
|
|
controller = TuiController(args(), on_start=start)
|
|
await controller.handle("setup.add_target", {"target": "https://example.com"})
|
|
|
|
result = await controller.handle("setup.start", {})
|
|
|
|
assert result == {"started": True}
|
|
assert started is True
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_start_without_target_requires_mount_consent() -> None:
|
|
started = False
|
|
|
|
async def start(_verify: bool = True) -> None:
|
|
nonlocal started
|
|
started = True
|
|
|
|
os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
|
|
os.environ["ANTHROPIC_API_KEY"] = "test-key"
|
|
loader._cached = None
|
|
controller = TuiController(args(), on_start=start)
|
|
|
|
# Mounting the working directory is never silent.
|
|
with pytest.raises(ValueError, match="No target set"):
|
|
await controller.handle("setup.start", {"verify": False})
|
|
assert started is False
|
|
assert controller.targets == []
|
|
assert controller.workspace_mount is None
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_target_less_start_enters_live_view_and_waits_for_the_mount() -> None:
|
|
"""Nothing is prepared until the live-view confirmation is answered."""
|
|
started = False
|
|
|
|
async def start(_verify: bool = True) -> None:
|
|
nonlocal started
|
|
started = True
|
|
|
|
os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
|
|
os.environ["ANTHROPIC_API_KEY"] = "test-key"
|
|
loader._cached = None
|
|
controller = TuiController(args(), on_start=start)
|
|
|
|
result = await controller.handle("setup.start", {"verify": False, "mount_working_dir": True})
|
|
|
|
assert result == {"started": True}
|
|
# The live view is up so the prompt can be shown there, but the scan has not
|
|
# been prepared and nothing is mounted yet.
|
|
assert started is False
|
|
assert controller.setup_mode is False
|
|
assert controller.scan_state == "preparing"
|
|
assert controller.pending_workspace_mount == str(Path.cwd())
|
|
assert controller.workspace_mount is None
|
|
assert controller.snapshot()["pending_mount"] == str(Path.cwd())
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_confirming_the_mount_starts_the_scan_without_a_target() -> None:
|
|
started = False
|
|
seen_verify: bool | None = None
|
|
|
|
async def start(verify: bool = True) -> None:
|
|
nonlocal started, seen_verify
|
|
started = True
|
|
seen_verify = verify
|
|
|
|
os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
|
|
os.environ["ANTHROPIC_API_KEY"] = "test-key"
|
|
loader._cached = None
|
|
controller = TuiController(args(), on_start=start)
|
|
await controller.handle("setup.start", {"verify": False, "mount_working_dir": True})
|
|
|
|
result = await controller.handle("setup.confirm_mount", {"approved": True})
|
|
|
|
assert result == {"approved": True}
|
|
assert started is True
|
|
# Launched optimistically, and mounted as a workspace: the scan genuinely
|
|
# has no target, so the instruction is the only source of truth.
|
|
assert seen_verify is False
|
|
assert controller.workspace_mount == str(Path.cwd())
|
|
assert controller.targets == []
|
|
assert controller.scan_state == "running"
|
|
assert controller.snapshot()["pending_mount"] == ""
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_declining_the_mount_returns_to_the_start_screen() -> None:
|
|
started = False
|
|
|
|
async def start(_verify: bool = True) -> None:
|
|
nonlocal started
|
|
started = True
|
|
|
|
os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
|
|
os.environ["ANTHROPIC_API_KEY"] = "test-key"
|
|
loader._cached = None
|
|
controller = TuiController(args(), on_start=start)
|
|
await controller.handle("setup.start", {"verify": False, "mount_working_dir": True})
|
|
|
|
result = await controller.handle("setup.confirm_mount", {"approved": False})
|
|
|
|
assert result == {"approved": False}
|
|
# Nothing was prepared, so the session goes back to the start screen and can
|
|
# be launched again.
|
|
assert started is False
|
|
assert controller.workspace_mount is None
|
|
assert controller.pending_workspace_mount is None
|
|
assert controller.setup_mode is True
|
|
assert controller.scan_started is False
|
|
assert controller.scan_state == "setup"
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_confirm_mount_requires_a_pending_request() -> None:
|
|
controller = TuiController(args())
|
|
|
|
with pytest.raises(RuntimeError, match="No mount confirmation is pending"):
|
|
await controller.handle("setup.confirm_mount", {"approved": True})
|
|
|
|
|
|
def test_snapshot_exposes_working_directory() -> None:
|
|
controller = TuiController(args())
|
|
|
|
assert controller.snapshot()["working_dir"] == str(Path.cwd())
|
|
assert controller.snapshot()["pending_mount"] == ""
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_start_forwards_verify_flag_by_default() -> None:
|
|
seen_verify: bool | None = None
|
|
|
|
async def start(verify: bool = True) -> None:
|
|
nonlocal seen_verify
|
|
seen_verify = verify
|
|
|
|
os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
|
|
os.environ["ANTHROPIC_API_KEY"] = "test-key"
|
|
loader._cached = None
|
|
controller = TuiController(args(), on_start=start)
|
|
await controller.handle("setup.add_target", {"target": "https://example.com"})
|
|
|
|
# A named target keeps the upfront model check.
|
|
await controller.handle("setup.start", {})
|
|
|
|
assert seen_verify is True
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_start_rejects_concurrent_and_repeated_submissions() -> None:
|
|
entered = asyncio.Event()
|
|
release = asyncio.Event()
|
|
|
|
async def start(_verify: bool = True) -> None:
|
|
entered.set()
|
|
await release.wait()
|
|
|
|
os.environ["STRIX_LLM"] = "anthropic/claude-sonnet-4"
|
|
os.environ["ANTHROPIC_API_KEY"] = "test-key"
|
|
loader._cached = None
|
|
controller = TuiController(args(), on_start=start)
|
|
await controller.handle("setup.add_target", {"target": "https://example.com"})
|
|
|
|
first_start = asyncio.create_task(controller.handle("setup.start", {}))
|
|
await entered.wait()
|
|
with pytest.raises(RuntimeError, match="already starting or running"):
|
|
await controller.handle("setup.start", {})
|
|
release.set()
|
|
await first_start
|
|
with pytest.raises(RuntimeError, match="already starting or running"):
|
|
await controller.handle("setup.start", {})
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
@pytest.mark.parametrize("status", ["completed", "failed", "crashed", "stopped"])
|
|
async def test_stop_rejects_terminal_agents(status: str) -> None:
|
|
class Coordinator:
|
|
def __init__(self) -> None:
|
|
self.calls: list[str] = []
|
|
|
|
async def cancel_descendants_graceful(self, agent_id: str) -> list[str]:
|
|
self.calls.append(agent_id)
|
|
return [agent_id]
|
|
|
|
coordinator = Coordinator()
|
|
controller = TuiController(args(), coordinator=coordinator)
|
|
controller.set_runtime(scan_loop=asyncio.get_running_loop())
|
|
controller.live_view.upsert_agent("agent-1", name="Agent", status=status)
|
|
|
|
with pytest.raises(RuntimeError, match=f"cannot be stopped while {status}"):
|
|
await controller.handle("agent.stop", {"agent_id": "agent-1"})
|
|
|
|
assert coordinator.calls == []
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
@pytest.mark.parametrize("status", ["running", "waiting", "budget_paused"])
|
|
async def test_stop_allows_active_agents(status: str) -> None:
|
|
class Coordinator:
|
|
def __init__(self) -> None:
|
|
self.calls: list[str] = []
|
|
|
|
async def cancel_descendants_graceful(self, agent_id: str) -> list[str]:
|
|
self.calls.append(agent_id)
|
|
return [agent_id]
|
|
|
|
coordinator = Coordinator()
|
|
controller = TuiController(args(), coordinator=coordinator)
|
|
controller.set_runtime(scan_loop=asyncio.get_running_loop())
|
|
controller.live_view.upsert_agent("agent-1", name="Agent", status=status)
|
|
|
|
result = await controller.handle("agent.stop", {"agent_id": "agent-1"})
|
|
|
|
assert result == {"stopped": True}
|
|
assert coordinator.calls == ["agent-1"]
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_stopping_agent_denies_pending_approvals_for_its_subtree() -> None:
|
|
class Coordinator:
|
|
async def cancel_descendants_graceful(self, agent_id: str) -> list[str]:
|
|
return ["agent-child", agent_id]
|
|
|
|
controller = TuiController(args(), coordinator=Coordinator())
|
|
controller.set_runtime(scan_loop=asyncio.get_running_loop())
|
|
controller.live_view.upsert_agent("agent-1", name="Agent", status="running")
|
|
approvals = [
|
|
asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{
|
|
"request_id": f"approval-{agent_id}",
|
|
"agent_id": agent_id,
|
|
"action": "Run action",
|
|
"reason": "Ambiguous effect",
|
|
}
|
|
)
|
|
)
|
|
for agent_id in ("agent-1", "agent-child")
|
|
]
|
|
await asyncio.sleep(0)
|
|
|
|
await controller.handle("agent.stop", {"agent_id": "agent-1"})
|
|
|
|
assert await asyncio.gather(*approvals) == ["cancelled", "cancelled"]
|
|
assert controller.snapshot()["pending_approvals"] == []
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_stop_handles_coordinator_rejection_after_stale_active_projection() -> None:
|
|
class Coordinator:
|
|
async def cancel_descendants_graceful(self, _agent_id: str) -> list[str]:
|
|
return []
|
|
|
|
controller = TuiController(args(), coordinator=Coordinator())
|
|
controller.set_runtime(scan_loop=asyncio.get_running_loop())
|
|
controller.live_view.upsert_agent("agent-1", name="Agent", status="running")
|
|
|
|
with pytest.raises(RuntimeError, match="no longer active"):
|
|
await controller.handle("agent.stop", {"agent_id": "agent-1"})
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_unknown_command_is_rejected() -> None:
|
|
controller = TuiController(args())
|
|
with pytest.raises(ValueError, match="Unknown command"):
|
|
await controller.handle("nope", {})
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_safety_approvals_are_all_visible_and_resolve_independently() -> None:
|
|
controller = TuiController(args())
|
|
first = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{
|
|
"request_id": "approval-1",
|
|
"agent_id": "agent-1",
|
|
"action": "Run exploit",
|
|
"reason": "Mutates state",
|
|
}
|
|
)
|
|
)
|
|
second = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
SimpleNamespace(
|
|
request_id="approval-2",
|
|
agent_id="agent-2",
|
|
action="Write a file",
|
|
reason="Changes the workspace",
|
|
)
|
|
)
|
|
)
|
|
await asyncio.sleep(0)
|
|
|
|
assert controller.snapshot()["pending_approvals"] == [
|
|
{
|
|
"request_id": "approval-1",
|
|
"action": "Run exploit",
|
|
"reason": "Mutates state",
|
|
"agent_id": "agent-1",
|
|
"tool_name": "",
|
|
"digest": "",
|
|
"risk": "",
|
|
},
|
|
{
|
|
"request_id": "approval-2",
|
|
"action": "Write a file",
|
|
"reason": "Changes the workspace",
|
|
"agent_id": "agent-2",
|
|
"tool_name": "",
|
|
"digest": "",
|
|
"risk": "",
|
|
},
|
|
]
|
|
with pytest.raises(ValueError, match="duplicate safety approval request_id"):
|
|
await controller.safety_approval_callback(
|
|
{
|
|
"request_id": "approval-1",
|
|
"agent_id": "agent-1",
|
|
"action": "Duplicate",
|
|
"reason": "Duplicate",
|
|
}
|
|
)
|
|
assert await controller.handle(
|
|
"safety.resolve", {"request_id": "approval-2", "approved": False}
|
|
) == {"request_id": "approval-2", "approved": False, "approve_all": False}
|
|
assert await second is False
|
|
assert [item["request_id"] for item in controller.snapshot()["pending_approvals"]] == [
|
|
"approval-1"
|
|
]
|
|
|
|
assert await controller.handle(
|
|
"safety.resolve", {"request_id": "approval-1", "approved": True}
|
|
) == {"request_id": "approval-1", "approved": True, "approve_all": False}
|
|
assert await first is True
|
|
with pytest.raises(RuntimeError, match="stale or unknown"):
|
|
await controller.handle("safety.resolve", {"request_id": "approval-2", "approved": False})
|
|
assert controller.snapshot()["pending_approvals"] == []
|
|
|
|
|
|
class _RecordingRuntime:
|
|
def __init__(self) -> None:
|
|
self.mode = "guarded"
|
|
|
|
def disable(self) -> None:
|
|
self.mode = "off"
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_approve_all_disables_review_and_releases_the_queue() -> None:
|
|
controller = TuiController(args())
|
|
runtime = _RecordingRuntime()
|
|
controller.register_safety_runtime(runtime)
|
|
first = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{"request_id": "a-1", "agent_id": "agent-1", "action": "Run", "reason": "x"}
|
|
)
|
|
)
|
|
second = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{"request_id": "a-2", "agent_id": "agent-2", "action": "Write", "reason": "y"}
|
|
)
|
|
)
|
|
await asyncio.sleep(0)
|
|
assert len(controller.snapshot()["pending_approvals"]) == 2
|
|
|
|
result = await controller.handle(
|
|
"safety.resolve", {"request_id": "a-1", "approved": True, "approve_all": True}
|
|
)
|
|
|
|
assert result == {"request_id": "a-1", "approved": True, "approve_all": True}
|
|
# The chosen call is approved and every other queued call is released as approved.
|
|
assert await first is True
|
|
assert await second is True
|
|
# Review is switched off for the rest of the run and the queue is cleared.
|
|
assert runtime.mode == "off"
|
|
assert controller.snapshot()["pending_approvals"] == []
|
|
# A review already past the runtime's mode check is auto-approved, not queued.
|
|
later = await controller.safety_approval_callback(
|
|
{"request_id": "a-3", "agent_id": "agent-1", "action": "Later", "reason": "z"}
|
|
)
|
|
assert later is True
|
|
assert controller.snapshot()["pending_approvals"] == []
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_approve_all_is_ignored_when_the_answer_is_deny() -> None:
|
|
controller = TuiController(args())
|
|
runtime = _RecordingRuntime()
|
|
controller.register_safety_runtime(runtime)
|
|
pending = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{"request_id": "a-1", "agent_id": "agent-1", "action": "Run", "reason": "x"}
|
|
)
|
|
)
|
|
await asyncio.sleep(0)
|
|
|
|
result = await controller.handle(
|
|
"safety.resolve", {"request_id": "a-1", "approved": False, "approve_all": True}
|
|
)
|
|
|
|
assert result == {"request_id": "a-1", "approved": False, "approve_all": False}
|
|
assert await pending is False
|
|
# A denial must never flip the run into dangerous mode.
|
|
assert runtime.mode == "guarded"
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_safety_approval_validates_response_and_sanitizes_display() -> None:
|
|
controller = TuiController(args())
|
|
pending = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{
|
|
"request_id": "approval-safe",
|
|
"agent_id": "agent-safe",
|
|
"action": "run\x1b]52;c;Y2xpcA==\x07 command\x85",
|
|
"reason": "needs\x1b[31m review\x1b[0m\x7f",
|
|
}
|
|
)
|
|
)
|
|
await asyncio.sleep(0)
|
|
|
|
assert controller.snapshot()["pending_approvals"] == [
|
|
{
|
|
"request_id": "approval-safe",
|
|
"action": "run command",
|
|
"reason": "needs review",
|
|
"agent_id": "agent-safe",
|
|
"tool_name": "",
|
|
"digest": "",
|
|
"risk": "",
|
|
}
|
|
]
|
|
with pytest.raises(TypeError, match="approved must be a boolean"):
|
|
await controller.handle(
|
|
"safety.resolve", {"request_id": "approval-safe", "approved": "yes"}
|
|
)
|
|
with pytest.raises(ValueError, match="request_id must be a non-empty string"):
|
|
await controller.handle("safety.resolve", {"request_id": "", "approved": False})
|
|
|
|
await controller.handle("safety.resolve", {"request_id": "approval-safe", "approved": False})
|
|
assert await pending is False
|
|
|
|
assert (
|
|
await controller.safety_approval_callback(
|
|
{"request_id": "approval-long", "action": "x" * 513, "reason": "Too long"}
|
|
)
|
|
is False
|
|
)
|
|
assert controller.snapshot()["pending_approvals"] == []
|
|
|
|
with pytest.raises(ValueError, match="agent_id must be a non-empty string"):
|
|
await controller.safety_approval_callback(
|
|
{"request_id": "approval-ownerless", "action": "Action", "reason": "Reason"}
|
|
)
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_cancelled_safety_request_is_removed_and_reveals_next() -> None:
|
|
controller = TuiController(args())
|
|
first = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{
|
|
"request_id": "approval-1",
|
|
"agent_id": "agent-1",
|
|
"action": "First",
|
|
"reason": "First reason",
|
|
}
|
|
)
|
|
)
|
|
second = asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{
|
|
"request_id": "approval-2",
|
|
"agent_id": "agent-2",
|
|
"action": "Second",
|
|
"reason": "Second reason",
|
|
}
|
|
)
|
|
)
|
|
await asyncio.sleep(0)
|
|
|
|
first.cancel()
|
|
with pytest.raises(asyncio.CancelledError):
|
|
await first
|
|
|
|
assert controller.snapshot()["pending_approvals"][0]["request_id"] == "approval-2"
|
|
await controller.handle("safety.resolve", {"request_id": "approval-2", "approved": False})
|
|
assert await second is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_quit_denies_all_pending_and_future_safety_approvals() -> None:
|
|
controller = TuiController(args())
|
|
requests = [
|
|
asyncio.create_task(
|
|
controller.safety_approval_callback(
|
|
{
|
|
"request_id": f"approval-{index}",
|
|
"agent_id": f"agent-{index}",
|
|
"action": "Action",
|
|
"reason": "Reason",
|
|
}
|
|
)
|
|
)
|
|
for index in range(2)
|
|
]
|
|
await asyncio.sleep(0)
|
|
|
|
await controller.handle("app.quit", {})
|
|
|
|
assert await asyncio.gather(*requests) == ["cancelled", "cancelled"]
|
|
assert controller.snapshot()["pending_approvals"] == []
|
|
assert (
|
|
await controller.safety_approval_callback(
|
|
{
|
|
"request_id": "approval-late",
|
|
"agent_id": "agent-late",
|
|
"action": "Late",
|
|
"reason": "Late reason",
|
|
}
|
|
)
|
|
== "cancelled"
|
|
)
|
|
|
|
|
|
def test_messages_are_sanitized_and_agents_are_collection_only() -> None:
|
|
controller = TuiController(args())
|
|
controller.add_message("replace\x1b]52;c;Y2xpcA==\x07 key\x85")
|
|
for index in range(40):
|
|
controller.live_view.upsert_agent(f"agent-{index}", name=f"Agent {index}")
|
|
|
|
snapshot = controller.snapshot()
|
|
|
|
assert "agents" not in snapshot
|
|
assert [message["text"] for message in snapshot["messages"]] == ["replace key"]
|
|
assert len(controller.collection("agents")) == 40
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_existing_viewer_is_reopened_and_closed(
|
|
monkeypatch: pytest.MonkeyPatch,
|
|
) -> None:
|
|
opened: list[str] = []
|
|
|
|
class ViewerServer:
|
|
shutdown_called = False
|
|
close_called = False
|
|
|
|
def shutdown(self) -> None:
|
|
self.shutdown_called = True
|
|
|
|
def server_close(self) -> None:
|
|
self.close_called = True
|
|
|
|
controller = TuiController(args())
|
|
controller.viewer_status = "running"
|
|
controller.viewer_url = "http://127.0.0.1:1234/?token=test"
|
|
server = ViewerServer()
|
|
controller._viewer_httpd = server
|
|
monkeypatch.setattr("strix.interface.tui.backend.controller.webbrowser.open", opened.append)
|
|
|
|
result = await controller.handle("viewer.open", {})
|
|
controller.close_viewer()
|
|
|
|
assert result == {"status": "running", "url": controller.viewer_url}
|
|
assert opened == [controller.viewer_url]
|
|
assert server.shutdown_called is True
|
|
assert server.close_called is True
|