Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
Python agentagentic-aiagentic-frameworkagentic-workflowaiautomationclaudedeepseekgenaigptllmlow-codemcpnextjsno-codeopenaipythonskillsworkflow
import pytest
from services.enterprise.base import EnterpriseRequest
from services.enterprise.rbac_service import RBACService
from tests.helpers.unit.rbac_fakes import RBACTransport
def test_membership_gateway_sends_identity_language_and_assignment(monkeypatch: pytest.MonkeyPatch) -> None:
transport = RBACTransport(response={"account_id": "member", "roles": []})
monkeypatch.setattr(EnterpriseRequest, "send_inner_rbac_request", transport.send)
result = RBACService.MemberRoles.replace("workspace", "actor", "member", ["role"], language="zh")
assert result.account_id == "member"
sent = transport.only_request
assert (sent.method, sent.endpoint) == ("PUT", "/rbac/members/rbac-roles")
assert (sent.tenant_id, sent.account_id) == ("workspace", "actor")
assert sent.params == {"account_id": "member", "language": "zh"}
assert sent.json == {"role_ids": ["role"]}
def test_permission_gateway_excludes_absent_resource_selectors(monkeypatch: pytest.MonkeyPatch) -> None:
transport = RBACTransport(response={})
monkeypatch.setattr(EnterpriseRequest, "send_inner_rbac_request", transport.send)
RBACService.MyPermissions.get("workspace", "actor", app_id=None, dataset_id=None, agent_id="agent", language=None)
assert transport.only_request.endpoint == "/rbac/my-permissions"
assert transport.only_request.params == {"agent_id": "agent"}