mirror of
https://github.com/openclaw/openclaw.git
synced 2026-08-12 21:53:00 -06:00
Peter Steinberger
73bba03e4c
* refactor: canonicalize session delivery state * test: canonicalize reply persistence fixtures * test: canonicalize talk delivery fixtures * test: canonicalize voice session routes * test: canonicalize attachment delivery fixtures * test: migrate gateway delivery fixtures * fix: skip invalid session delivery rows * test: align delivery SDK surface gates * fix: preserve legacy delivery precedence * test: canonicalize heartbeat delivery fixtures * fix: preserve delivery route prompt identity * test: canonicalize session delivery fixtures * fix: preserve recoverable legacy delivery routes * fix: canonicalize remaining session state * fix: preserve canonical session classification * style: format delivery state changes * test: refresh plugin SDK delivery baseline * test: avoid mutating session fixture input * style: simplify delivery identity check * style: simplify delivery origin spread * fix: preserve fresh delivery route metadata * test: assert canonical surface route switch * fix: canonicalize doctor file-store imports * fix: preserve transitional delivery migration state * fix: satisfy canonical delivery CI gates * ci: scope GitHub App token permissions * test: infer canonical delivery projections * test: canonicalize ACP requester delivery fixtures * test: canonicalize harness rollback fixture * style: apply pinned formatter
4258 lines
151 KiB
TypeScript
4258 lines
151 KiB
TypeScript
/** Tests live model switching behavior in active agent command sessions. */
|
||
|
||
import { expectDefined } from "@openclaw/normalization-core";
|
||
import { afterAll, afterEach, beforeAll, beforeEach, describe, expect, it, vi } from "vitest";
|
||
import type { SessionEntry } from "../config/sessions.js";
|
||
import {
|
||
deliveryContextFromSession,
|
||
normalizeSessionDeliveryState,
|
||
} from "../utils/delivery-context.shared.js";
|
||
import {
|
||
buildTestAllowedModelSet,
|
||
type CommandSessionEntryFixture,
|
||
buildTestConfiguredModelCatalog,
|
||
buildTestModelAliasIndex,
|
||
createChannelModelRuntimeConfig,
|
||
createCommandSessionEntry,
|
||
createCommandSessionFixture,
|
||
createConfiguredModelCompatRuntimeConfig,
|
||
createTestModelVisibilityPolicy,
|
||
isTestModelKeyAllowed,
|
||
normalizeTestProviderId,
|
||
resolveTestConfiguredModelRef,
|
||
resolveTestDefaultModelForAgent,
|
||
resolveTestModelRefFromString,
|
||
} from "./agent-command.live-model-switch.test-helpers.js";
|
||
import {
|
||
INTERNAL_RUNTIME_CONTEXT_BEGIN,
|
||
INTERNAL_RUNTIME_CONTEXT_END,
|
||
} from "./internal-runtime-context.js";
|
||
import { LiveSessionModelSwitchError } from "./live-model-switch-error.js";
|
||
import {
|
||
createAgentRunDirectAbortError,
|
||
createAgentRunRestartAbortError,
|
||
} from "./run-termination.js";
|
||
|
||
const state = vi.hoisted(() => ({
|
||
defaultRuntimeConfig: {
|
||
agents: {
|
||
defaults: {
|
||
models: {
|
||
"anthropic/claude": {},
|
||
"openai/claude": {},
|
||
"openai/gpt-5.4": {},
|
||
},
|
||
},
|
||
},
|
||
},
|
||
runtimeConfigMock: undefined as unknown,
|
||
acpResolveSessionMock: vi.fn((..._args: unknown[]): unknown => null),
|
||
acpRunTurnMock: vi.fn((..._args: unknown[]): unknown => undefined),
|
||
buildAcpResultMock: vi.fn(),
|
||
createAcpVisibleTextAccumulatorMock: vi.fn(),
|
||
emitAcpLifecycleEndMock: vi.fn(),
|
||
emitAcpLifecycleErrorMock: vi.fn(),
|
||
persistCliTurnTranscriptMock: vi.fn(),
|
||
persistAcpTurnTranscriptMock: vi.fn(),
|
||
runCliTurnCompactionLifecycleMock: vi.fn(),
|
||
resolveAcpAgentPolicyErrorMock: vi.fn(),
|
||
resolveAcpDispatchPolicyErrorMock: vi.fn(),
|
||
resolveAcpExplicitTurnPolicyErrorMock: vi.fn(),
|
||
runWithModelFallbackMock: vi.fn(),
|
||
runAgentAttemptMock: vi.fn(),
|
||
resolveAgentSkillsFilterMock: vi.fn(
|
||
(_cfg?: unknown, _agentId?: string): string[] | undefined => undefined,
|
||
),
|
||
resolveEffectiveModelFallbacksMock: vi.fn().mockReturnValue(undefined),
|
||
hasLegacyAutoFallbackWithoutOriginMock: vi.fn((_entry: unknown) => false),
|
||
isModelSelectionLockedMock: vi.fn(
|
||
(entry: unknown) =>
|
||
(entry as { modelSelectionLocked?: boolean } | undefined)?.modelSelectionLocked === true,
|
||
),
|
||
applyModelOverrideToSessionEntryMock: vi.fn((_params: unknown) => ({ updated: false })),
|
||
repairProviderWrappedModelOverrideMock: vi.fn((_params: unknown) => ({ updated: false })),
|
||
resolveAutoFallbackPrimaryProbeMock: vi.fn((_params: unknown) => undefined as unknown),
|
||
resolveChannelModelOverrideMock: vi.fn((_params: unknown) => null as unknown),
|
||
assertLifecycleCurrentMock: vi.fn(),
|
||
emitAgentEventMock: vi.fn(),
|
||
registerAgentRunContextMock: vi.fn(),
|
||
clearAgentRunContextMock: vi.fn(),
|
||
loadSessionEntryMock: vi.fn(),
|
||
updateSessionStoreAfterAgentRunMock: vi.fn(),
|
||
deliverAgentCommandResultMock: vi.fn(),
|
||
resolveAgentDeliveryPlanMock: vi.fn(),
|
||
resolveAgentDeliveryPlanWithSessionRouteMock: vi.fn(),
|
||
resolveAgentOutboundTargetMock: vi.fn(),
|
||
resolveMessageChannelSelectionMock: vi.fn(),
|
||
createTrajectoryRuntimeRecorderMock: vi.fn(),
|
||
trajectoryRecordEventMock: vi.fn(),
|
||
trajectoryFlushMock: vi.fn(async () => undefined),
|
||
persistSessionEntryMock: vi.fn(async (..._args: unknown[]): Promise<unknown> => undefined),
|
||
clearSessionAuthProfileOverrideMock: vi.fn(),
|
||
isThinkingLevelSupportedMock: vi.fn((_args: unknown) => true),
|
||
resolveSupportedThinkingLevelMock: vi.fn(({ level }: { level?: string }) => level),
|
||
resolveThinkingDefaultMock: vi.fn((_args: unknown) => "low"),
|
||
loadManifestModelCatalogMock: vi.fn(() => []),
|
||
buildWorkspaceSkillSnapshotMock: vi.fn((..._args: unknown[]): unknown => ({
|
||
prompt: "",
|
||
skills: [],
|
||
resolvedSkills: [],
|
||
version: 0,
|
||
})),
|
||
prepareInternalSessionEffectsSessionMock: vi.fn(),
|
||
removeInternalSessionEffectsSessionMock: vi.fn(),
|
||
authProfileStoreMock: { profiles: {} } as { profiles: Record<string, unknown> },
|
||
sessionEntryMock: undefined as SessionEntry | undefined,
|
||
sessionStoreMock: undefined as unknown,
|
||
storePathMock: undefined as string | undefined,
|
||
resolvedSessionKeyMock: undefined as string | undefined,
|
||
trajectoryRecorderParamsMock: vi.fn(),
|
||
}));
|
||
|
||
vi.mock("./model-fallback.js", () => ({
|
||
runWithModelFallback: (params: unknown) => state.runWithModelFallbackMock(params),
|
||
}));
|
||
|
||
vi.mock("./command/attempt-execution.runtime.js", () => ({
|
||
buildAcpResult: (...args: unknown[]) => state.buildAcpResultMock(...args),
|
||
createAcpToolLifecycleTracker: () => ({
|
||
active: new Map(),
|
||
terminalToolCallIds: new Set(),
|
||
saturated: false,
|
||
}),
|
||
createAcpVisibleTextAccumulator: () => state.createAcpVisibleTextAccumulatorMock(),
|
||
emitAcpAssistantDelta: vi.fn(),
|
||
emitAcpLifecycleEnd: (...args: unknown[]) => state.emitAcpLifecycleEndMock(...args),
|
||
emitAcpLifecycleError: (...args: unknown[]) => state.emitAcpLifecycleErrorMock(...args),
|
||
emitAcpLifecycleStart: vi.fn(),
|
||
emitAcpRuntimeEvent: vi.fn(),
|
||
persistCliTurnTranscript: (...args: unknown[]) => state.persistCliTurnTranscriptMock(...args),
|
||
persistAcpTurnTranscript: (...args: unknown[]) => state.persistAcpTurnTranscriptMock(...args),
|
||
persistSessionEntry: vi.fn(),
|
||
prependInternalEventContext: (body: string) => body,
|
||
runAgentAttempt: (...args: unknown[]) => state.runAgentAttemptMock(...args),
|
||
sessionFileHasContent: vi.fn(async () => false),
|
||
}));
|
||
|
||
vi.mock("./command/attempt-execution.shared.js", async () => {
|
||
const actual = await vi.importActual<typeof import("./command/attempt-execution.shared.js")>(
|
||
"./command/attempt-execution.shared.js",
|
||
);
|
||
return {
|
||
...actual,
|
||
persistSessionEntry: (...args: unknown[]) => state.persistSessionEntryMock(...args),
|
||
};
|
||
});
|
||
|
||
vi.mock("./command/delivery.runtime.js", () => ({
|
||
deliverAgentCommandResult: (...args: unknown[]) => state.deliverAgentCommandResultMock(...args),
|
||
}));
|
||
|
||
vi.mock("./command/cli-compaction.js", () => ({
|
||
runCliTurnCompactionLifecycle: (...args: unknown[]) =>
|
||
state.runCliTurnCompactionLifecycleMock(...args),
|
||
}));
|
||
|
||
vi.mock("./command/run-context.js", () => ({
|
||
resolveAgentRunContext: (opts: {
|
||
accountId?: string;
|
||
channel?: string;
|
||
groupChannel?: string | null;
|
||
groupId?: string | null;
|
||
groupSpace?: string | null;
|
||
messageChannel?: string;
|
||
replyChannel?: string;
|
||
runContext?: {
|
||
accountId?: string;
|
||
currentChannelId?: string;
|
||
currentThreadTs?: string;
|
||
groupChannel?: string | null;
|
||
groupId?: string | null;
|
||
groupSpace?: string | null;
|
||
messageChannel?: string;
|
||
replyToMode?: "off" | "first" | "all" | "batched";
|
||
};
|
||
threadId?: string | number;
|
||
to?: string;
|
||
}) => ({
|
||
messageChannel:
|
||
opts.runContext?.messageChannel ?? opts.messageChannel ?? opts.replyChannel ?? opts.channel,
|
||
accountId: opts.runContext?.accountId ?? opts.accountId ?? "acct",
|
||
groupId: opts.runContext?.groupId ?? opts.groupId,
|
||
groupChannel: opts.runContext?.groupChannel ?? opts.groupChannel,
|
||
groupSpace: opts.runContext?.groupSpace ?? opts.groupSpace,
|
||
currentChannelId: undefined,
|
||
currentThreadTs:
|
||
opts.runContext?.currentThreadTs ??
|
||
(opts.threadId == null ? undefined : String(opts.threadId)),
|
||
replyToMode: opts.runContext?.replyToMode,
|
||
hasRepliedRef: { current: false },
|
||
}),
|
||
}));
|
||
|
||
vi.mock("./command/session-store.runtime.js", () => ({
|
||
loadSessionEntry: (...args: unknown[]) => state.loadSessionEntryMock(...args),
|
||
loadSessionEntryReadOnly: (...args: unknown[]) => state.loadSessionEntryMock(...args),
|
||
updateSessionStoreAfterAgentRun: (...args: unknown[]) =>
|
||
state.updateSessionStoreAfterAgentRunMock(...args),
|
||
}));
|
||
|
||
vi.mock("./command/session.js", () => ({
|
||
resolveSession: () => {
|
||
const sessionEntry: SessionEntry = state.sessionEntryMock ?? {
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
};
|
||
return {
|
||
sessionId: "session-1",
|
||
sessionKey: state.resolvedSessionKeyMock ?? "agent:main:main",
|
||
sessionEntry,
|
||
sessionStore: state.sessionStoreMock,
|
||
storePath: state.storePathMock,
|
||
isNewSession: false,
|
||
persistedThinking:
|
||
typeof sessionEntry.thinkingLevel === "string" ? sessionEntry.thinkingLevel : undefined,
|
||
persistedVerbose: undefined,
|
||
};
|
||
},
|
||
}));
|
||
|
||
vi.mock("./command/types.js", () => ({}));
|
||
|
||
// Recovery ownership has dedicated store-backed coverage. This command suite
|
||
// uses an intentionally synthetic session resolver with no durable store path.
|
||
vi.mock("./main-session-recovery-store.js", () => ({
|
||
claimMainSessionRecoveryOwner: vi.fn(async () => ({ kind: "not_required" })),
|
||
inspectMainSessionRecoveryRequired: vi.fn(async () => ({ kind: "not_required" })),
|
||
releaseMainSessionRecoveryOwner: vi.fn(async () => undefined),
|
||
validateMainSessionRecoveryOwner: vi.fn(async () => true),
|
||
}));
|
||
|
||
vi.mock("./harness/runtime-plugin.js", () => ({
|
||
ensureSelectedAgentHarnessPlugin: vi.fn(async () => undefined),
|
||
}));
|
||
|
||
// Harness selection has dedicated coverage; this command suite registers no auto harnesses.
|
||
vi.mock("./harness/support.js", () => ({
|
||
resolveAutoAgentHarnessId: () => undefined,
|
||
}));
|
||
|
||
vi.mock("../acp/policy.js", () => ({
|
||
isAcpEnabledByPolicy: () => true,
|
||
resolveAcpAgentPolicyError: (...args: unknown[]) => state.resolveAcpAgentPolicyErrorMock(...args),
|
||
resolveAcpDispatchPolicyError: (...args: unknown[]) =>
|
||
state.resolveAcpDispatchPolicyErrorMock(...args),
|
||
resolveAcpExplicitTurnPolicyError: (...args: unknown[]) =>
|
||
state.resolveAcpExplicitTurnPolicyErrorMock(...args),
|
||
}));
|
||
|
||
vi.mock("../acp/runtime/errors.js", () => ({
|
||
toAcpRuntimeError: ({ error }: { error: unknown }) =>
|
||
error instanceof Error ? error : new Error(String(error)),
|
||
}));
|
||
|
||
vi.mock("@openclaw/acp-core/runtime/session-identifiers", () => ({
|
||
resolveAcpSessionCwd: () => "/tmp",
|
||
}));
|
||
|
||
vi.mock("../auto-reply/thinking.js", () => ({
|
||
formatThinkingLevels: () => "low, medium, high",
|
||
normalizeThinkLevel: (v?: string) => v || undefined,
|
||
normalizeVerboseLevel: (v?: string) => v || undefined,
|
||
isThinkingLevelSupported: (args: unknown) => state.isThinkingLevelSupportedMock(args),
|
||
resolveSupportedThinkingLevel: (args: { level?: string }) =>
|
||
state.resolveSupportedThinkingLevelMock(args),
|
||
supportsXHighThinking: () => false,
|
||
}));
|
||
|
||
vi.mock("../cli/command-format.js", () => ({
|
||
formatCliCommand: (cmd: string) => cmd,
|
||
}));
|
||
|
||
vi.mock("../cli/command-secret-gateway.js", () => ({
|
||
resolveCommandSecretRefsViaGateway: async (params: { config: unknown }) => ({
|
||
resolvedConfig: params.config,
|
||
diagnostics: [],
|
||
}),
|
||
}));
|
||
|
||
vi.mock("../cli/command-secret-targets.js", () => ({
|
||
getAgentRuntimeCommandSecretTargetIds: () => [],
|
||
}));
|
||
|
||
vi.mock("../cli/deps.js", () => ({
|
||
createDefaultDeps: () => ({}),
|
||
}));
|
||
|
||
vi.mock("../config/io.js", () => ({
|
||
getRuntimeConfig: () => state.runtimeConfigMock ?? state.defaultRuntimeConfig,
|
||
readConfigFileSnapshotForWrite: async () => ({
|
||
snapshot: { valid: false },
|
||
}),
|
||
}));
|
||
|
||
vi.mock("./agent-runtime-config.js", () => {
|
||
return {
|
||
resolveAgentRuntimeConfig: async () => ({
|
||
loadedRaw: state.runtimeConfigMock ?? state.defaultRuntimeConfig,
|
||
sourceConfig: state.runtimeConfigMock ?? state.defaultRuntimeConfig,
|
||
cfg: state.runtimeConfigMock ?? state.defaultRuntimeConfig,
|
||
}),
|
||
};
|
||
});
|
||
|
||
vi.mock("../plugins/plugin-metadata-snapshot.js", () => ({
|
||
isPluginMetadataSnapshotCompatible: () => false,
|
||
resolvePluginMetadataSnapshot: () => ({ plugins: [] }),
|
||
}));
|
||
|
||
vi.mock("../config/runtime-snapshot.js", () => ({
|
||
setRuntimeConfigSnapshot: vi.fn(),
|
||
}));
|
||
|
||
vi.mock("../config/sessions.js", () => ({
|
||
resolveAgentIdFromSessionKey: () => "default",
|
||
mergeSessionEntry: (a: unknown, b: unknown) => ({ ...(a as object), ...(b as object) }),
|
||
updateSessionStore: vi.fn(
|
||
async (_path: string, fn: (store: Record<string, unknown>) => unknown) => {
|
||
const store: Record<string, unknown> = {};
|
||
return fn(store);
|
||
},
|
||
),
|
||
}));
|
||
|
||
vi.mock("../config/sessions/transcript-resolve.runtime.js", () => ({
|
||
resolveSessionTranscriptFile: async (params: { sessionEntry?: SessionEntry }) => ({
|
||
sessionFile: params.sessionEntry?.sessionFile ?? "/tmp/session.jsonl",
|
||
sessionEntry: params.sessionEntry ?? { sessionId: "session-1", updatedAt: Date.now() },
|
||
}),
|
||
}));
|
||
|
||
vi.mock("./internal-session-effects.js", () => ({
|
||
prepareInternalSessionEffectsSession: (...args: unknown[]) =>
|
||
state.prepareInternalSessionEffectsSessionMock(...args),
|
||
removeInternalSessionEffectsSession: (...args: unknown[]) =>
|
||
state.removeInternalSessionEffectsSessionMock(...args),
|
||
resolveInternalSessionEffectsTarget: ({ agentId, runId, storePath }: Record<string, string>) => ({
|
||
agentId,
|
||
sessionId: `internal-${runId}`,
|
||
sessionKey: `agent:${agentId}:internal-session-effects:${runId}`,
|
||
storePath,
|
||
}),
|
||
}));
|
||
|
||
vi.mock("../infra/agent-events.js", () => ({
|
||
assertAgentRunLifecycleGenerationCurrent: (...args: unknown[]) =>
|
||
state.assertLifecycleCurrentMock(...args),
|
||
captureAgentRunLifecycleGeneration: () => "test-generation",
|
||
clearAgentRunContext: (...args: unknown[]) => state.clearAgentRunContextMock(...args),
|
||
emitAgentEvent: (...args: unknown[]) => state.emitAgentEventMock(...args),
|
||
getAgentEventLifecycleGeneration: () => "test-generation",
|
||
isAgentEventLifecycleGenerationCurrent: (generation: string) => generation === "test-generation",
|
||
onAgentEvent: vi.fn(),
|
||
registerAgentEventLifecycleRotationHandler: vi.fn(),
|
||
registerAgentRunContext: (...args: unknown[]) => state.registerAgentRunContextMock(...args),
|
||
withAgentRunLifecycleGeneration: (_generation: string, run: () => unknown) => run(),
|
||
}));
|
||
|
||
vi.mock("../infra/outbound/session-context.js", () => ({
|
||
buildOutboundSessionContext: () => ({}),
|
||
}));
|
||
|
||
vi.mock("../infra/outbound/agent-delivery.js", () => ({
|
||
resolveAgentDeliveryPlan: (...args: unknown[]) => state.resolveAgentDeliveryPlanMock(...args),
|
||
resolveAgentDeliveryPlanWithSessionRoute: (...args: unknown[]) =>
|
||
state.resolveAgentDeliveryPlanWithSessionRouteMock(...args),
|
||
resolveAgentOutboundTarget: (...args: unknown[]) => state.resolveAgentOutboundTargetMock(...args),
|
||
}));
|
||
|
||
vi.mock("../infra/outbound/channel-selection.js", () => ({
|
||
resolveMessageChannelSelection: (...args: unknown[]) =>
|
||
state.resolveMessageChannelSelectionMock(...args),
|
||
}));
|
||
|
||
vi.mock("../infra/skills-remote.js", () => ({
|
||
getRemoteSkillEligibility: () => ({ eligible: false }),
|
||
}));
|
||
|
||
vi.mock("../logging/subsystem.js", () => ({
|
||
createSubsystemLogger: () => {
|
||
const logger = {
|
||
info: vi.fn(),
|
||
warn: vi.fn(),
|
||
error: vi.fn(),
|
||
debug: vi.fn(),
|
||
trace: vi.fn(),
|
||
raw: vi.fn(),
|
||
child: vi.fn(() => logger),
|
||
};
|
||
return logger;
|
||
},
|
||
}));
|
||
|
||
afterAll(() => {
|
||
// This suite runs in a shared worker; do not leak its module-level logger
|
||
// mock into later files that verify real warning diagnostics.
|
||
vi.doUnmock("../logging/subsystem.js");
|
||
});
|
||
|
||
vi.mock("../channels/model-overrides.js", () => ({
|
||
resolveChannelModelOverride: (params: unknown) => state.resolveChannelModelOverrideMock(params),
|
||
}));
|
||
|
||
vi.mock("../routing/session-key.js", async () => {
|
||
const actual = await vi.importActual<typeof import("../routing/session-key.js")>(
|
||
"../routing/session-key.js",
|
||
);
|
||
return {
|
||
...actual,
|
||
normalizeAgentId: (id: string) => id,
|
||
normalizeMainKey: (key?: string | null) => key?.trim() || "main",
|
||
};
|
||
});
|
||
|
||
vi.mock("../runtime.js", () => ({
|
||
defaultRuntime: {
|
||
error: vi.fn(),
|
||
log: vi.fn(),
|
||
},
|
||
}));
|
||
|
||
vi.mock("../sessions/level-overrides.js", () => ({
|
||
applyVerboseOverride: vi.fn(),
|
||
}));
|
||
|
||
vi.mock("../sessions/model-overrides.js", () => ({
|
||
applyModelOverrideToSessionEntry: (params: unknown) =>
|
||
state.applyModelOverrideToSessionEntryMock(params),
|
||
isModelSelectionLocked: (entry: unknown) => state.isModelSelectionLockedMock(entry),
|
||
MODEL_SELECTION_LOCKED_MESSAGE: "Model selection is locked for this session.",
|
||
ModelSelectionLockedError: class ModelSelectionLockedError extends Error {
|
||
constructor() {
|
||
super("Model selection is locked for this session.");
|
||
this.name = "ModelSelectionLockedError";
|
||
}
|
||
},
|
||
repairProviderWrappedModelOverride: (params: unknown) =>
|
||
state.repairProviderWrappedModelOverrideMock(params),
|
||
}));
|
||
|
||
vi.mock("../sessions/send-policy.js", () => ({
|
||
resolveSendPolicy: () => "allow",
|
||
}));
|
||
|
||
vi.mock("../terminal/ansi.js", () => ({
|
||
sanitizeForLog: (s: string) => s,
|
||
}));
|
||
|
||
vi.mock("../trajectory/runtime.js", () => ({
|
||
createTrajectoryRuntimeRecorder: (params: unknown) => {
|
||
state.createTrajectoryRuntimeRecorderMock(params);
|
||
state.trajectoryRecorderParamsMock(params);
|
||
return {
|
||
enabled: true,
|
||
filePath: "/tmp/session.trajectory.jsonl",
|
||
recordEvent: (...args: unknown[]) => state.trajectoryRecordEventMock(...args),
|
||
flush: () => state.trajectoryFlushMock(),
|
||
};
|
||
},
|
||
}));
|
||
|
||
vi.mock("../utils/message-channel.js", () => ({
|
||
INTERNAL_MESSAGE_CHANNEL: "internal",
|
||
isDeliverableMessageChannel: (value: string) => value !== "internal",
|
||
normalizeMessageChannel: (value?: string | null) => value?.trim().toLowerCase() || undefined,
|
||
resolveMessageChannel: (...values: Array<string | null | undefined>) =>
|
||
values
|
||
.find((value) => value?.trim())
|
||
?.trim()
|
||
.toLowerCase(),
|
||
}));
|
||
|
||
vi.mock("./agent-scope.js", () => ({
|
||
clearAutoFallbackPrimaryProbeSelection: vi.fn(),
|
||
entryMatchesAutoFallbackPrimaryProbe: () => true,
|
||
hasLegacyAutoFallbackWithoutOrigin: (entry: unknown) =>
|
||
state.hasLegacyAutoFallbackWithoutOriginMock(entry),
|
||
hasSessionAutoModelFallbackProvenance: () => false,
|
||
listAgentEntries: () => [],
|
||
listAgentIds: () => ["default"],
|
||
markAutoFallbackPrimaryProbe: vi.fn(),
|
||
resolveAutoFallbackPrimaryProbe: (params: unknown) =>
|
||
state.resolveAutoFallbackPrimaryProbeMock(params),
|
||
resolveAgentConfig: () => undefined,
|
||
resolveAgentDir: () => "/tmp/agent",
|
||
resolveAgentEffectiveModelPrimary: (cfg: unknown) => {
|
||
const raw = (cfg as { agents?: { defaults?: { model?: string | { primary?: string } } } })
|
||
?.agents?.defaults?.model;
|
||
return typeof raw === "string" ? raw : raw?.primary;
|
||
},
|
||
resolveDefaultAgentId: () => "default",
|
||
resolveEffectiveModelFallbacks: state.resolveEffectiveModelFallbacksMock,
|
||
resolveSessionAgentIds: () => ({ defaultAgentId: "default", sessionAgentId: "default" }),
|
||
resolveSessionAgentId: () => "default",
|
||
resolveAgentSkillsFilter: () => undefined,
|
||
resolveAgentWorkspaceDir: () => "/tmp/workspace",
|
||
}));
|
||
|
||
vi.mock("./auth-profiles.js", () => ({
|
||
ensureAuthProfileStore: () => ({ profiles: {} }),
|
||
}));
|
||
|
||
vi.mock("./auth-profiles/store.js", () => ({
|
||
ensureAuthProfileStore: () => state.authProfileStoreMock,
|
||
}));
|
||
|
||
vi.mock("./auth-profiles/session-override.js", () => ({
|
||
clearSessionAuthProfileOverride: (...args: unknown[]) =>
|
||
state.clearSessionAuthProfileOverrideMock(...args),
|
||
}));
|
||
|
||
vi.mock("./defaults.js", () => ({
|
||
DEFAULT_MODEL: "claude",
|
||
DEFAULT_PROVIDER: "anthropic",
|
||
}));
|
||
|
||
// Exec eligibility is outside model-switch scope; avoid loading its policy graph.
|
||
vi.mock("./exec-defaults.js", () => ({
|
||
resolveNodeExecEligibility: () => ({ canExec: false }),
|
||
}));
|
||
|
||
vi.mock("./lanes.js", () => ({
|
||
AGENT_LANE_SUBAGENT: "subagent",
|
||
}));
|
||
|
||
vi.mock("./model-catalog.js", () => ({
|
||
loadManifestModelCatalog: state.loadManifestModelCatalogMock,
|
||
}));
|
||
|
||
vi.mock("./model-selection.js", () => ({
|
||
buildAllowedModelSet: buildTestAllowedModelSet,
|
||
createModelVisibilityPolicy: createTestModelVisibilityPolicy,
|
||
buildConfiguredModelCatalog: ({ cfg }: { cfg?: unknown }) => buildTestConfiguredModelCatalog(cfg),
|
||
isModelKeyAllowedBySet: isTestModelKeyAllowed,
|
||
buildModelAliasIndex: buildTestModelAliasIndex,
|
||
modelKey: (provider: string, model: string) => `${provider}/${model}`,
|
||
normalizeModelRef: (provider: string, model: string) => ({
|
||
provider: normalizeTestProviderId(provider),
|
||
model,
|
||
}),
|
||
normalizeProviderId: normalizeTestProviderId,
|
||
normalizeProviderIdForAuth: normalizeTestProviderId,
|
||
parseModelRef: (model: string, provider: string) => {
|
||
const slash = model.indexOf("/");
|
||
return slash > 0
|
||
? { provider: model.slice(0, slash), model: model.slice(slash + 1) }
|
||
: { provider, model };
|
||
},
|
||
resolveModelRefFromString: resolveTestModelRefFromString,
|
||
resolveConfiguredModelRef: resolveTestConfiguredModelRef,
|
||
resolveDefaultModelForAgent: resolveTestDefaultModelForAgent,
|
||
resolveThinkingDefault: (args: unknown) => state.resolveThinkingDefaultMock(args),
|
||
}));
|
||
|
||
vi.mock("./model-visibility-policy.js", () => ({
|
||
createModelVisibilityPolicy: createTestModelVisibilityPolicy,
|
||
}));
|
||
|
||
vi.mock("./provider-auth-aliases.js", () => ({
|
||
resolveProviderAuthAliasMap: () => ({}),
|
||
resolveProviderIdForAuth: (provider: string) =>
|
||
provider.trim().toLowerCase() === "codex-cli" ? "openai" : provider.trim().toLowerCase(),
|
||
}));
|
||
|
||
vi.mock("../skills/discovery/agent-filter.js", () => ({
|
||
resolveEffectiveAgentSkillFilter: (_cfg: unknown, agentId: string) =>
|
||
state.resolveAgentSkillsFilterMock(_cfg, agentId),
|
||
}));
|
||
|
||
vi.mock("../skills/runtime/remote.js", () => ({
|
||
getRemoteSkillEligibility: () => ({ eligible: false }),
|
||
}));
|
||
|
||
vi.mock("../skills/runtime/session-snapshot.js", () => ({
|
||
resolveReusableWorkspaceSkillSnapshot: (params: {
|
||
workspaceDir: string;
|
||
existingSnapshot?: { resolvedSkills?: unknown };
|
||
skillFilter?: string[];
|
||
}) => {
|
||
if (params.skillFilter !== undefined && params.skillFilter.length === 0) {
|
||
return {
|
||
snapshot: {
|
||
prompt: "",
|
||
skills: [],
|
||
resolvedSkills: [],
|
||
skillFilter: params.skillFilter,
|
||
version: 0,
|
||
},
|
||
shouldRefresh: !params.existingSnapshot,
|
||
snapshotVersion: 0,
|
||
};
|
||
}
|
||
if (params.existingSnapshot?.resolvedSkills !== undefined) {
|
||
return {
|
||
snapshot: params.existingSnapshot,
|
||
shouldRefresh: false,
|
||
snapshotVersion: 0,
|
||
};
|
||
}
|
||
const rebuilt = state.buildWorkspaceSkillSnapshotMock(params.workspaceDir, params) as {
|
||
resolvedSkills?: unknown;
|
||
};
|
||
return {
|
||
snapshot: params.existingSnapshot
|
||
? { ...params.existingSnapshot, resolvedSkills: rebuilt?.resolvedSkills }
|
||
: rebuilt,
|
||
shouldRefresh: !params.existingSnapshot,
|
||
snapshotVersion: 0,
|
||
};
|
||
},
|
||
}));
|
||
|
||
vi.mock("./spawned-context.js", () => ({
|
||
normalizeSpawnedRunMetadata: (meta: unknown) => meta ?? {},
|
||
}));
|
||
|
||
vi.mock("./timeout.js", () => ({
|
||
resolveAgentTimeoutMs: ({ overrideSeconds }: { overrideSeconds?: number | null }) =>
|
||
typeof overrideSeconds === "number" && Number.isFinite(overrideSeconds)
|
||
? overrideSeconds === 0
|
||
? 2_147_483_647
|
||
: Math.max(overrideSeconds * 1000, 1)
|
||
: 30_000,
|
||
}));
|
||
|
||
vi.mock("./workspace.js", () => ({
|
||
ensureAgentWorkspace: async () => ({ dir: "/tmp/workspace" }),
|
||
}));
|
||
|
||
vi.mock("../acp/control-plane/manager.js", () => ({
|
||
getAcpSessionManager: () => ({
|
||
resolveSession: (...args: unknown[]) => state.acpResolveSessionMock(...args),
|
||
runTurn: (...args: unknown[]) => state.acpRunTurnMock(...args),
|
||
}),
|
||
}));
|
||
|
||
let agentCommand: typeof import("./agent-command.js").agentCommand;
|
||
let agentCommandTesting: typeof import("./agent-command.js").testing;
|
||
|
||
beforeAll(async () => {
|
||
const mod = await import("./agent-command.js");
|
||
agentCommand ??= mod.agentCommand;
|
||
agentCommandTesting ??= mod.testing;
|
||
});
|
||
|
||
type FallbackRunnerParams = {
|
||
provider: string;
|
||
model: string;
|
||
sessionId?: string;
|
||
fallbacksOverride?: string[];
|
||
resolveAgentHarnessRuntimeOverride?: (provider: string, model: string) => string | undefined;
|
||
run: (provider: string, model: string) => Promise<unknown>;
|
||
onFallbackStep?: (step: Record<string, unknown>) => void | Promise<void>;
|
||
classifyResult?: (params: {
|
||
provider: string;
|
||
model: string;
|
||
result: unknown;
|
||
attempt: number;
|
||
total: number;
|
||
}) => unknown;
|
||
};
|
||
|
||
type ModelSwitchOptions = ConstructorParameters<typeof LiveSessionModelSwitchError>[0];
|
||
|
||
function makeSuccessResult(provider: string, model: string) {
|
||
return {
|
||
payloads: [{ text: "ok" }],
|
||
meta: {
|
||
durationMs: 100,
|
||
aborted: false,
|
||
stopReason: "end_turn",
|
||
agentMeta: { provider, model },
|
||
},
|
||
};
|
||
}
|
||
|
||
function makeEmptyResult(provider: string, model: string) {
|
||
return {
|
||
payloads: [],
|
||
meta: {
|
||
durationMs: 30_000,
|
||
aborted: false,
|
||
stopReason: "end_turn",
|
||
agentHarnessResultClassification: "empty",
|
||
agentMeta: { provider, model },
|
||
},
|
||
};
|
||
}
|
||
|
||
function setupModelSwitchRetry(switchOptions: ModelSwitchOptions) {
|
||
let invocation = 0;
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
invocation += 1;
|
||
if (invocation === 1) {
|
||
throw new LiveSessionModelSwitchError(switchOptions);
|
||
}
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
}
|
||
|
||
function setupSingleAttemptFallback() {
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
}
|
||
|
||
function setupSuccessfulAttempt(provider = "openai", model = "gpt-5.4"): void {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult(provider, model));
|
||
}
|
||
|
||
function setupAcpSession(): void {
|
||
state.acpResolveSessionMock.mockReturnValue({
|
||
kind: "ready",
|
||
meta: { agent: "claude", cwd: "/tmp/workspace" },
|
||
});
|
||
}
|
||
|
||
function requireRecord(value: unknown, label: string): Record<string, unknown> {
|
||
if (!value || typeof value !== "object") {
|
||
throw new Error(`expected ${label} to be an object`);
|
||
}
|
||
return value as Record<string, unknown>;
|
||
}
|
||
|
||
function requireArray(value: unknown, label: string): unknown[] {
|
||
if (!Array.isArray(value)) {
|
||
throw new Error(`expected ${label} to be an array`);
|
||
}
|
||
return value;
|
||
}
|
||
|
||
function mockCallArg(mock: ReturnType<typeof vi.fn>, callIndex = 0, argIndex = 0): unknown {
|
||
const call = mock.mock.calls[callIndex] as unknown[] | undefined;
|
||
if (!call) {
|
||
throw new Error(`expected mock call ${callIndex}`);
|
||
}
|
||
return call[argIndex];
|
||
}
|
||
|
||
function expectRecordFields(value: unknown, expected: Record<string, unknown>): void {
|
||
const actual = requireRecord(value, "record");
|
||
for (const [key, expectedValue] of Object.entries(expected)) {
|
||
expect(actual[key]).toEqual(expectedValue);
|
||
}
|
||
}
|
||
|
||
async function runBasicAgentCommand() {
|
||
await agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
});
|
||
}
|
||
|
||
function runDiscordDelivery(overrides: Partial<Parameters<typeof agentCommand>[0]> = {}) {
|
||
return agentCommand({
|
||
message: "hello",
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
accountId: "main",
|
||
deliver: true,
|
||
...overrides,
|
||
});
|
||
}
|
||
|
||
function runInternalModelCommand(runId: string) {
|
||
return agentCommand({
|
||
message: "probe",
|
||
to: "+1234567890",
|
||
runId,
|
||
modelRun: true,
|
||
promptMode: "none",
|
||
sessionEffects: "internal",
|
||
});
|
||
}
|
||
|
||
function setupStoredSession(
|
||
overrides: CommandSessionEntryFixture = {},
|
||
storePath = "/tmp/openclaw-sessions.json",
|
||
sessionKey = "agent:main:main",
|
||
): { entry: SessionEntry; store: Record<string, SessionEntry> } {
|
||
const fixture = createCommandSessionFixture(overrides, sessionKey);
|
||
state.sessionEntryMock = fixture.entry;
|
||
state.sessionStoreMock = fixture.store;
|
||
state.storePathMock = storePath;
|
||
return fixture;
|
||
}
|
||
|
||
function setupBareStoredSession(
|
||
overrides: CommandSessionEntryFixture = {},
|
||
storePath = "/tmp/openclaw-sessions.json",
|
||
sessionKey = "agent:main:main",
|
||
): { entry: SessionEntry; store: Record<string, SessionEntry> } {
|
||
const entry = createCommandSessionEntry(overrides);
|
||
const store = { [sessionKey]: entry };
|
||
state.sessionEntryMock = entry;
|
||
state.sessionStoreMock = store;
|
||
state.storePathMock = storePath;
|
||
return { entry, store };
|
||
}
|
||
|
||
function expectFallbackOverrideCalls(first: boolean, second: boolean) {
|
||
expect(state.resolveEffectiveModelFallbacksMock).toHaveBeenCalledTimes(2);
|
||
expectRecordFields(mockCallArg(state.resolveEffectiveModelFallbacksMock, 0), {
|
||
hasSessionModelOverride: first,
|
||
});
|
||
expectRecordFields(mockCallArg(state.resolveEffectiveModelFallbacksMock, 1), {
|
||
hasSessionModelOverride: second,
|
||
});
|
||
}
|
||
|
||
describe("agentCommand – LiveSessionModelSwitchError retry", () => {
|
||
beforeEach(() => {
|
||
vi.clearAllMocks();
|
||
state.acpResolveSessionMock.mockReturnValue(null);
|
||
state.resolveAcpAgentPolicyErrorMock.mockReturnValue(null);
|
||
state.resolveAcpDispatchPolicyErrorMock.mockReturnValue(null);
|
||
state.resolveAcpExplicitTurnPolicyErrorMock.mockReturnValue(null);
|
||
state.runtimeConfigMock = undefined;
|
||
delete (state.defaultRuntimeConfig.agents as { list?: unknown }).list;
|
||
state.isThinkingLevelSupportedMock.mockReturnValue(true);
|
||
state.resolveSupportedThinkingLevelMock.mockImplementation(
|
||
({ level }: { level?: string }) => level,
|
||
);
|
||
state.resolveThinkingDefaultMock.mockReturnValue("low");
|
||
state.resolveAgentSkillsFilterMock.mockReturnValue(undefined);
|
||
state.loadManifestModelCatalogMock.mockReturnValue([]);
|
||
state.hasLegacyAutoFallbackWithoutOriginMock.mockReturnValue(false);
|
||
state.isModelSelectionLockedMock.mockImplementation(
|
||
(entry: unknown) =>
|
||
(entry as { modelSelectionLocked?: boolean } | undefined)?.modelSelectionLocked === true,
|
||
);
|
||
state.applyModelOverrideToSessionEntryMock.mockReturnValue({ updated: false });
|
||
state.repairProviderWrappedModelOverrideMock.mockReturnValue({ updated: false });
|
||
state.resolveAutoFallbackPrimaryProbeMock.mockReturnValue(undefined);
|
||
state.resolveChannelModelOverrideMock.mockImplementation((params: unknown) => {
|
||
const input = params as {
|
||
cfg?: { channels?: { modelByChannel?: Record<string, Record<string, string>> } };
|
||
channel?: string;
|
||
groupId?: string;
|
||
parentSessionKey?: string;
|
||
};
|
||
const channel = input.channel?.trim().toLowerCase();
|
||
const entries = channel ? input.cfg?.channels?.modelByChannel?.[channel] : undefined;
|
||
if (!entries) {
|
||
return null;
|
||
}
|
||
const direct = input.groupId ? entries[input.groupId] : undefined;
|
||
if (direct) {
|
||
return { channel, model: direct, matchKey: input.groupId };
|
||
}
|
||
const parentChannel = input.parentSessionKey?.match(/:channel:([^:]+)/u)?.[1];
|
||
const parent = parentChannel ? entries[parentChannel] : undefined;
|
||
return parent ? { channel, model: parent, matchKey: parentChannel } : null;
|
||
});
|
||
state.acpRunTurnMock.mockImplementation(async (params: unknown) => {
|
||
const onEvent = (params as { onEvent?: (event: unknown) => void }).onEvent;
|
||
onEvent?.({ type: "text_delta", stream: "output", text: "done" });
|
||
onEvent?.({ type: "done", stopReason: "end_turn" });
|
||
});
|
||
state.createAcpVisibleTextAccumulatorMock.mockImplementation(() => {
|
||
let text = "";
|
||
return {
|
||
consume(chunk: string) {
|
||
text += chunk;
|
||
return { text, delta: chunk };
|
||
},
|
||
finalizeRaw: () => text,
|
||
finalize: () => text,
|
||
};
|
||
});
|
||
state.buildAcpResultMock.mockImplementation((params: { payloadText?: string }) => ({
|
||
payloads: params.payloadText ? [{ text: params.payloadText }] : [],
|
||
meta: { durationMs: 0, stopReason: "end_turn" },
|
||
}));
|
||
state.persistCliTurnTranscriptMock.mockImplementation(
|
||
async (params: { sessionEntry?: unknown }) => ({
|
||
kind: "persisted",
|
||
sessionEntry: params.sessionEntry,
|
||
}),
|
||
);
|
||
state.persistAcpTurnTranscriptMock.mockImplementation(
|
||
async (params: { sessionEntry?: unknown }) => ({
|
||
kind: "persisted",
|
||
sessionEntry: params.sessionEntry,
|
||
}),
|
||
);
|
||
state.runCliTurnCompactionLifecycleMock.mockImplementation(
|
||
async (params: { sessionEntry?: unknown }) => params.sessionEntry,
|
||
);
|
||
state.authProfileStoreMock = { profiles: {} };
|
||
state.sessionEntryMock = undefined;
|
||
state.sessionStoreMock = undefined;
|
||
state.storePathMock = undefined;
|
||
state.resolvedSessionKeyMock = undefined;
|
||
state.persistSessionEntryMock.mockImplementation(async (...args: unknown[]) => {
|
||
const params = args[0] as {
|
||
sessionStore?: Record<string, unknown>;
|
||
sessionKey?: string;
|
||
entry?: unknown;
|
||
shouldPersist?: (entry: unknown) => boolean;
|
||
};
|
||
const current =
|
||
params.sessionStore && params.sessionKey
|
||
? params.sessionStore[params.sessionKey]
|
||
: undefined;
|
||
if (params.shouldPersist && !params.shouldPersist(current)) {
|
||
if (current === undefined && params.sessionStore && params.sessionKey) {
|
||
delete params.sessionStore[params.sessionKey];
|
||
}
|
||
return current;
|
||
}
|
||
if (params.sessionStore && params.sessionKey && params.entry) {
|
||
params.sessionStore[params.sessionKey] = params.entry;
|
||
return params.entry;
|
||
}
|
||
return current;
|
||
});
|
||
state.buildWorkspaceSkillSnapshotMock.mockReturnValue({
|
||
prompt: "",
|
||
skills: [],
|
||
resolvedSkills: [],
|
||
version: 0,
|
||
});
|
||
state.deliverAgentCommandResultMock.mockResolvedValue(undefined);
|
||
state.resolveAgentOutboundTargetMock.mockImplementation(
|
||
(params: { plan?: { resolvedTo?: string }; targetMode?: string }) => ({
|
||
resolvedTarget: null,
|
||
resolvedTo: params.plan?.resolvedTo,
|
||
targetMode: params.targetMode ?? "implicit",
|
||
}),
|
||
);
|
||
state.resolveMessageChannelSelectionMock.mockRejectedValue(new Error("channel required"));
|
||
state.loadSessionEntryMock.mockReset().mockImplementation((params: { sessionKey?: string }) => {
|
||
const sessionKey = params.sessionKey ?? state.resolvedSessionKeyMock ?? "agent:main:main";
|
||
return (state.sessionStoreMock as Record<string, SessionEntry> | undefined)?.[sessionKey];
|
||
});
|
||
state.resolveAgentDeliveryPlanMock.mockImplementation(
|
||
(params: {
|
||
accountId?: string;
|
||
explicitThreadId?: string | number;
|
||
explicitTo?: string;
|
||
requestedChannel?: string;
|
||
sessionEntry?: SessionEntry;
|
||
}) => {
|
||
const context = deliveryContextFromSession(params.sessionEntry);
|
||
const channel = params.requestedChannel ?? context?.channel ?? "internal";
|
||
const to = params.explicitTo ?? context?.to;
|
||
const accountId = params.accountId ?? context?.accountId;
|
||
const threadId = params.explicitThreadId ?? context?.threadId;
|
||
return {
|
||
baseDelivery: {},
|
||
resolvedChannel: channel,
|
||
resolvedTo: to,
|
||
resolvedAccountId: accountId,
|
||
resolvedThreadId: threadId,
|
||
deliveryTargetMode: params.explicitTo ? "explicit" : to ? "implicit" : undefined,
|
||
};
|
||
},
|
||
);
|
||
state.resolveAgentDeliveryPlanWithSessionRouteMock.mockImplementation((params: unknown) =>
|
||
state.resolveAgentDeliveryPlanMock(params),
|
||
);
|
||
state.updateSessionStoreAfterAgentRunMock.mockResolvedValue(undefined);
|
||
state.trajectoryFlushMock.mockResolvedValue(undefined);
|
||
state.prepareInternalSessionEffectsSessionMock.mockResolvedValue({
|
||
agentId: "default",
|
||
sessionId: "internal-session",
|
||
sessionKey: "agent:default:internal-session-effects:run",
|
||
storePath: "/tmp/openclaw-session-store.json",
|
||
sessionFile: "sqlite:default:internal-session:/tmp/openclaw-session-store.json",
|
||
sessionEntry: { sessionId: "internal-session", updatedAt: 1 },
|
||
});
|
||
state.removeInternalSessionEffectsSessionMock.mockResolvedValue(undefined);
|
||
});
|
||
|
||
afterEach(() => {
|
||
vi.restoreAllMocks();
|
||
});
|
||
|
||
it("retries with the switched provider/model when LiveSessionModelSwitchError is thrown", async () => {
|
||
setupModelSwitchRetry({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
});
|
||
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(state.runWithModelFallbackMock).toHaveBeenCalledTimes(2);
|
||
|
||
const secondCall = mockCallArg(state.runWithModelFallbackMock, 1) as FallbackRunnerParams;
|
||
expect(secondCall.provider).toBe("openai");
|
||
expect(secondCall.model).toBe("gpt-5.4");
|
||
expect(secondCall.sessionId).toBe("session-1");
|
||
|
||
const lifecycleEndCalls = state.emitAgentEventMock.mock.calls.filter((call: unknown[]) => {
|
||
const arg = call[0] as { stream?: string; data?: { phase?: string } };
|
||
return arg?.stream === "lifecycle" && arg?.data?.phase === "end";
|
||
});
|
||
expect(lifecycleEndCalls.length).toBeGreaterThanOrEqual(1);
|
||
const lifecycleFinishingCalls = state.emitAgentEventMock.mock.calls.filter(
|
||
(call: unknown[]) => {
|
||
const arg = call[0] as { stream?: string; data?: { phase?: string } };
|
||
return arg?.stream === "lifecycle" && arg?.data?.phase === "finishing";
|
||
},
|
||
);
|
||
expect(lifecycleFinishingCalls.length).toBeGreaterThanOrEqual(1);
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
deferTerminalLifecycle: true,
|
||
});
|
||
const firstFinishingIndex = state.emitAgentEventMock.mock.calls.findIndex((call: unknown[]) => {
|
||
const arg = call[0] as { stream?: string; data?: { phase?: string } };
|
||
return arg?.stream === "lifecycle" && arg?.data?.phase === "finishing";
|
||
});
|
||
const lastEndIndex = state.emitAgentEventMock.mock.calls.findLastIndex((call: unknown[]) => {
|
||
const arg = call[0] as { stream?: string; data?: { phase?: string } };
|
||
return arg?.stream === "lifecycle" && arg?.data?.phase === "end";
|
||
});
|
||
expect(state.deliverAgentCommandResultMock).toHaveBeenCalledTimes(1);
|
||
const deliveryOrder = state.deliverAgentCommandResultMock.mock.invocationCallOrder[0] ?? 0;
|
||
expect(
|
||
state.emitAgentEventMock.mock.invocationCallOrder[firstFinishingIndex] ?? 0,
|
||
).toBeLessThan(deliveryOrder);
|
||
expect(deliveryOrder).toBeLessThan(
|
||
state.emitAgentEventMock.mock.invocationCallOrder[lastEndIndex] ?? 0,
|
||
);
|
||
});
|
||
|
||
it("forwards the auth profile bound to the configured default model", async () => {
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
model: { primary: "anthropic/claude@anthropic:verified" },
|
||
models: { "anthropic/claude": {} },
|
||
},
|
||
},
|
||
};
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
authProfileOverride: "anthropic:stale-auto",
|
||
authProfileOverrideSource: "auto",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
providerOverride: "anthropic",
|
||
modelOverride: "claude",
|
||
configuredAuthProfileId: "anthropic:verified",
|
||
});
|
||
});
|
||
|
||
it("retries a same-model switch with the runtime carried by the error", async () => {
|
||
const sessionEntry: SessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
agentRuntimeOverride: "openclaw",
|
||
};
|
||
state.sessionEntryMock = sessionEntry;
|
||
state.sessionStoreMock = { "agent:main:main": sessionEntry };
|
||
state.storePathMock = "/tmp/openclaw-sessions.json";
|
||
setupModelSwitchRetry({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
agentRuntimeOverride: "codex",
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const retry = mockCallArg(state.runWithModelFallbackMock, 1) as FallbackRunnerParams;
|
||
expect(retry.resolveAgentHarnessRuntimeOverride?.("openai", "gpt-5.4")).toBe("codex");
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
providerOverride: "openai",
|
||
modelOverride: "gpt-5.4",
|
||
agentHarnessRuntimeOverride: "codex",
|
||
});
|
||
});
|
||
|
||
it("rejects live model switches for locked sessions without retrying", async () => {
|
||
setupModelSwitchRetry({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
});
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
modelSelectionLocked: true,
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
state.isModelSelectionLockedMock.mockReturnValue(true);
|
||
|
||
await expect(runBasicAgentCommand()).rejects.toMatchObject({
|
||
name: "ModelSelectionLockedError",
|
||
message: "Model selection is locked for this session.",
|
||
});
|
||
|
||
expect(state.runWithModelFallbackMock).toHaveBeenCalledTimes(1);
|
||
expect(state.runAgentAttemptMock).not.toHaveBeenCalled();
|
||
expect(state.trajectoryFlushMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it("pins catalog-adopted direct runs before fallback preflight", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
models: state.defaultRuntimeConfig.agents.defaults.models,
|
||
},
|
||
},
|
||
};
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
agentHarnessId: "codex",
|
||
modelSelectionLocked: true,
|
||
pluginExtensions: {
|
||
codex: {
|
||
supervision: {
|
||
sourceThreadId: "019f-codex-thread",
|
||
modelLocked: true,
|
||
},
|
||
},
|
||
},
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
state.isModelSelectionLockedMock.mockReturnValue(true);
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.resolveAgentHarnessRuntimeOverride?.("anthropic", "claude")).toBe(
|
||
"codex",
|
||
);
|
||
expect(fallbackParams.fallbacksOverride).toEqual([]);
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
providerOverride: "anthropic",
|
||
modelOverride: "claude",
|
||
agentHarnessRuntimeOverride: "codex",
|
||
sessionEntry: expect.objectContaining({
|
||
agentHarnessId: "codex",
|
||
modelSelectionLocked: true,
|
||
}),
|
||
});
|
||
});
|
||
|
||
it("skips legacy override repair when continuing an ordinary locked harness session", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.resolvedSessionKeyMock = "agent:main:plugin-owned";
|
||
state.hasLegacyAutoFallbackWithoutOriginMock.mockReturnValue(true);
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
agentHarnessId: "codex",
|
||
modelSelectionLocked: true,
|
||
providerOverride: "openai",
|
||
modelOverride: "stale-fallback-model",
|
||
modelOverrideSource: "auto",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(state.applyModelOverrideToSessionEntryMock).not.toHaveBeenCalled();
|
||
expect(state.repairProviderWrappedModelOverrideMock).not.toHaveBeenCalled();
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
agentHarnessRuntimeOverride: "codex",
|
||
sessionEntry: expect.objectContaining({
|
||
agentHarnessId: "codex",
|
||
modelSelectionLocked: true,
|
||
providerOverride: "openai",
|
||
modelOverride: "stale-fallback-model",
|
||
modelOverrideSource: "auto",
|
||
}),
|
||
});
|
||
});
|
||
|
||
it("keeps the fast mode cutoff timestamp across live model switch retries", async () => {
|
||
let invocation = 0;
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
invocation++;
|
||
const result = await params.run(params.provider, params.model);
|
||
if (invocation === 1) {
|
||
throw new LiveSessionModelSwitchError({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
});
|
||
}
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const firstAttempt = mockCallArg(state.runAgentAttemptMock, 0) as {
|
||
fastModeStartedAtMs?: number;
|
||
};
|
||
const secondAttempt = mockCallArg(state.runAgentAttemptMock, 1) as {
|
||
fastModeStartedAtMs?: number;
|
||
};
|
||
expect(firstAttempt.fastModeStartedAtMs).toBe(secondAttempt.fastModeStartedAtMs);
|
||
});
|
||
|
||
it("reuses durable user-turn proof across live model switch retries", async () => {
|
||
let fallbackInvocation = 0;
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
fallbackInvocation += 1;
|
||
const result = await params.run(params.provider, params.model);
|
||
if (fallbackInvocation === 1) {
|
||
throw new LiveSessionModelSwitchError({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
});
|
||
}
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: unknown) => {
|
||
const attempt = attemptParams as {
|
||
userTurnTranscriptRecorder?: {
|
||
markRuntimePersisted: (message: { role: "user"; content: string }) => void;
|
||
};
|
||
};
|
||
if (state.runAgentAttemptMock.mock.calls.length === 1) {
|
||
attempt.userTurnTranscriptRecorder?.markRuntimePersisted({
|
||
role: "user",
|
||
content: "hello",
|
||
});
|
||
}
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const firstAttempt = mockCallArg(state.runAgentAttemptMock, 0) as {
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
userTurnTranscriptRecorder?: unknown;
|
||
};
|
||
const secondAttempt = mockCallArg(state.runAgentAttemptMock, 1) as {
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
userTurnTranscriptRecorder?: unknown;
|
||
};
|
||
expect(secondAttempt.userTurnTranscriptRecorder).toBe(firstAttempt.userTurnTranscriptRecorder);
|
||
expect(firstAttempt.suppressPromptPersistenceOnRetry).toBe(false);
|
||
expect(secondAttempt.suppressPromptPersistenceOnRetry).toBe(true);
|
||
});
|
||
|
||
it("uses an embedded queue rebound generation for terminal lifecycle and cleanup", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: unknown) => {
|
||
(
|
||
attemptParams as {
|
||
onLifecycleGenerationChanged?: (lifecycleGeneration: string) => void;
|
||
}
|
||
).onLifecycleGenerationChanged?.("post-restart-generation");
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const lifecycleEnd = state.emitAgentEventMock.mock.calls
|
||
.map(
|
||
(call) =>
|
||
call[0] as {
|
||
stream?: string;
|
||
data?: { phase?: string };
|
||
lifecycleGeneration?: string;
|
||
},
|
||
)
|
||
.find((event) => event.stream === "lifecycle" && event.data?.phase === "end");
|
||
expect(lifecycleEnd?.lifecycleGeneration).toBe("post-restart-generation");
|
||
expect(state.clearAgentRunContextMock).toHaveBeenCalledWith(
|
||
expect.any(String),
|
||
"post-restart-generation",
|
||
);
|
||
});
|
||
|
||
it("uses an embedded queue rebound generation for cleanup when the attempt fails", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: unknown) => {
|
||
(
|
||
attemptParams as {
|
||
onLifecycleGenerationChanged?: (lifecycleGeneration: string) => void;
|
||
}
|
||
).onLifecycleGenerationChanged?.("post-restart-generation");
|
||
throw new Error("attempt failed after queue rebound");
|
||
});
|
||
|
||
await expect(runBasicAgentCommand()).rejects.toThrow("attempt failed after queue rebound");
|
||
|
||
expect(state.clearAgentRunContextMock).toHaveBeenCalledWith(
|
||
expect.any(String),
|
||
"post-restart-generation",
|
||
);
|
||
});
|
||
|
||
it("preserves restart ownership when an aborted attempt resolves normally", async () => {
|
||
setupSingleAttemptFallback();
|
||
const controller = new AbortController();
|
||
state.runAgentAttemptMock.mockImplementation(async () => {
|
||
controller.abort(createAgentRunRestartAbortError());
|
||
return {
|
||
payloads: [],
|
||
meta: {
|
||
durationMs: 100,
|
||
aborted: true,
|
||
stopReason: "end_turn",
|
||
agentMeta: { provider: "anthropic", model: "claude" },
|
||
},
|
||
};
|
||
});
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
abortSignal: controller.signal,
|
||
}),
|
||
).rejects.toThrow("agent run aborted for restart");
|
||
|
||
const lifecycleEvents = state.emitAgentEventMock.mock.calls
|
||
.map((call) => call[0] as { stream?: string; data?: Record<string, unknown> })
|
||
.filter((event) => event.stream === "lifecycle");
|
||
expect(lifecycleEvents).toEqual(
|
||
expect.arrayContaining([
|
||
expect.objectContaining({
|
||
data: expect.objectContaining({
|
||
phase: "error",
|
||
aborted: true,
|
||
stopReason: "restart",
|
||
}),
|
||
}),
|
||
]),
|
||
);
|
||
expect(state.deliverAgentCommandResultMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("preserves restart ownership when an aborted ACP turn resolves normally", async () => {
|
||
setupAcpSession();
|
||
const controller = new AbortController();
|
||
controller.abort(createAgentRunRestartAbortError());
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
sessionKey: "agent:main:main",
|
||
abortSignal: controller.signal,
|
||
}),
|
||
).rejects.toThrow("agent run aborted for restart");
|
||
|
||
expect(state.emitAcpLifecycleEndMock).not.toHaveBeenCalled();
|
||
expect(state.deliverAgentCommandResultMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("suppresses ACP delivery when restart begins during transcript persistence", async () => {
|
||
setupAcpSession();
|
||
const controller = new AbortController();
|
||
state.persistAcpTurnTranscriptMock.mockImplementation(
|
||
async (params: { sessionEntry?: unknown }) => {
|
||
controller.abort(createAgentRunRestartAbortError());
|
||
return { kind: "persisted", sessionEntry: params.sessionEntry };
|
||
},
|
||
);
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
sessionKey: "agent:main:main",
|
||
abortSignal: controller.signal,
|
||
}),
|
||
).rejects.toThrow("agent run aborted for restart");
|
||
|
||
expect(state.emitAcpLifecycleEndMock).not.toHaveBeenCalled();
|
||
expect(state.emitAcpLifecycleErrorMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
runId: "session-1",
|
||
sessionKey: "agent:main:main",
|
||
}),
|
||
);
|
||
const lifecycleError = state.emitAcpLifecycleErrorMock.mock.calls[0]?.[0] as
|
||
| { abortSignal?: AbortSignal }
|
||
| undefined;
|
||
expect(lifecycleError?.abortSignal?.aborted).toBe(true);
|
||
expect(lifecycleError?.abortSignal?.reason).toBe(controller.signal.reason);
|
||
expect(state.persistAcpTurnTranscriptMock).toHaveBeenCalledTimes(1);
|
||
expect(state.buildAcpResultMock).not.toHaveBeenCalled();
|
||
expect(state.deliverAgentCommandResultMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("threads lifecycle ownership into ACP delivery", async () => {
|
||
setupAcpSession();
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
sessionKey: "agent:main:main",
|
||
});
|
||
|
||
const deliveryParams = requireRecord(
|
||
mockCallArg(state.deliverAgentCommandResultMock),
|
||
"ACP delivery params",
|
||
);
|
||
expect(deliveryParams.assertDeliveryCurrent).toBeTypeOf("function");
|
||
(deliveryParams.assertDeliveryCurrent as () => void)();
|
||
expect(state.assertLifecycleCurrentMock).toHaveBeenLastCalledWith("test-generation");
|
||
});
|
||
|
||
it("persists structured transcript media for ACP turns", async () => {
|
||
setupAcpSession();
|
||
|
||
await agentCommand({
|
||
message: "[media attached: media://inbound/image-1]",
|
||
transcriptMessage: "",
|
||
transcriptMedia: [{ path: "/media/inbound/image-1.png", contentType: "image/png" }],
|
||
sessionKey: "agent:main:main",
|
||
});
|
||
|
||
expect(state.persistAcpTurnTranscriptMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
transcriptBody: "",
|
||
userInput: {
|
||
text: "",
|
||
media: [{ path: "/media/inbound/image-1.png", contentType: "image/png" }],
|
||
},
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("keeps the initial session touch for local runs", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupStoredSession();
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const touchWrites = state.persistSessionEntryMock.mock.calls.filter((call) => {
|
||
const entry = (call[0] as { entry?: Record<string, unknown> } | undefined)?.entry;
|
||
return entry?.lastInteractionAt !== undefined;
|
||
});
|
||
expect(touchWrites).toHaveLength(1);
|
||
expect(state.updateSessionStoreAfterAgentRunMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it("threads lifecycle ownership into normal delivery", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupStoredSession();
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const deliveryParams = requireRecord(
|
||
mockCallArg(state.deliverAgentCommandResultMock),
|
||
"delivery params",
|
||
);
|
||
expect(deliveryParams.assertDeliveryCurrent).toBeTypeOf("function");
|
||
(deliveryParams.assertDeliveryCurrent as () => void)();
|
||
expect(state.assertLifecycleCurrentMock).toHaveBeenLastCalledWith("test-generation");
|
||
});
|
||
|
||
it("passes explicit timeout overrides into agent attempts", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
timeout: "600",
|
||
});
|
||
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
timeoutMs: 600_000,
|
||
runTimeoutOverrideMs: 600_000,
|
||
});
|
||
});
|
||
|
||
it("clamps unsupported explicit thinking for subagent spawns instead of throwing", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude-fable-5"));
|
||
state.resolvedSessionKeyMock = "agent:planner:subagent:00000000-0000-4000-8000-000000000000";
|
||
state.isThinkingLevelSupportedMock.mockReturnValue(false);
|
||
state.resolveSupportedThinkingLevelMock.mockReturnValue("high");
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
sessionKey: state.resolvedSessionKeyMock,
|
||
thinking: "xhigh",
|
||
lane: "subagent",
|
||
});
|
||
|
||
expect(state.resolveSupportedThinkingLevelMock).toHaveBeenCalled();
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
resolvedThinkLevel: "high",
|
||
});
|
||
});
|
||
|
||
it.each([
|
||
{
|
||
name: "rejects unsupported explicit thinking for interactive subagent-key runs",
|
||
sessionKey: "agent:planner:subagent:00000000-0000-4000-8000-000000000000",
|
||
lane: undefined,
|
||
},
|
||
{
|
||
name: "rejects unsupported explicit thinking for non-subagent sessions on the subagent lane",
|
||
sessionKey: "agent:main:main",
|
||
lane: "subagent" as const,
|
||
},
|
||
])("$name", async ({ sessionKey, lane }) => {
|
||
setupSuccessfulAttempt("anthropic", "claude-fable-5");
|
||
state.resolvedSessionKeyMock = sessionKey;
|
||
state.isThinkingLevelSupportedMock.mockReturnValue(false);
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
sessionKey: state.resolvedSessionKeyMock,
|
||
thinking: "xhigh",
|
||
lane,
|
||
}),
|
||
).rejects.toThrow(/is not supported/u);
|
||
expect(state.runAgentAttemptMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("rejects unsupported explicit thinking for direct interactive runs", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude-fable-5"));
|
||
state.resolvedSessionKeyMock = "agent:main:main";
|
||
state.isThinkingLevelSupportedMock.mockReturnValue(false);
|
||
const { store } = setupStoredSession({ thinkingLevel: "low" });
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
thinking: "ultra",
|
||
}),
|
||
).rejects.toThrow(/is not supported/u);
|
||
expect(state.runAgentAttemptMock).not.toHaveBeenCalled();
|
||
expect(store["agent:main:main"]?.thinkingLevel).toBe("low");
|
||
expect(state.persistSessionEntryMock).not.toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
entry: expect.objectContaining({ thinkingLevel: "ultra" }),
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("skips the initial session touch after gateway ingress already persisted activity", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupStoredSession();
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
skipInitialSessionTouch: true,
|
||
});
|
||
|
||
const touchWrites = state.persistSessionEntryMock.mock.calls.filter((call) => {
|
||
const entry = (call[0] as { entry?: Record<string, unknown> } | undefined)?.entry;
|
||
return entry?.lastInteractionAt !== undefined;
|
||
});
|
||
expect(touchWrites).toHaveLength(0);
|
||
expect(state.updateSessionStoreAfterAgentRunMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it.each([
|
||
{
|
||
name: "uses channel model override as the initial run model for channel-backed sessions",
|
||
sessionEntry: createCommandSessionFixture({ channel: "discord", groupId: "channel-123" })
|
||
.entry,
|
||
command: undefined,
|
||
legacyFallback: false,
|
||
expectResolverContext: true,
|
||
expectAttemptModel: true,
|
||
},
|
||
{
|
||
name: "uses current run channel context when persisted session metadata is absent",
|
||
sessionEntry: undefined,
|
||
command: {
|
||
message: "hello",
|
||
channel: "discord",
|
||
groupId: "channel-123",
|
||
to: "discord:channel:channel-123",
|
||
},
|
||
legacyFallback: false,
|
||
expectResolverContext: false,
|
||
expectAttemptModel: false,
|
||
},
|
||
{
|
||
name: "keeps persisted channel model override when current run context is internal",
|
||
sessionEntry: createCommandSessionFixture({ channel: "discord", groupId: "channel-123" })
|
||
.entry,
|
||
command: {
|
||
message: "hello",
|
||
channel: "internal",
|
||
messageChannel: "internal",
|
||
to: "internal",
|
||
},
|
||
legacyFallback: false,
|
||
expectResolverContext: true,
|
||
expectAttemptModel: false,
|
||
},
|
||
{
|
||
name: "uses channel model override after ignoring stale legacy fallback overrides",
|
||
sessionEntry: createCommandSessionFixture({
|
||
channel: "discord",
|
||
groupId: "channel-123",
|
||
providerOverride: "anthropic",
|
||
modelOverride: "stale-fallback-model",
|
||
}).entry,
|
||
command: undefined,
|
||
legacyFallback: true,
|
||
expectResolverContext: false,
|
||
expectAttemptModel: false,
|
||
},
|
||
])(
|
||
"$name",
|
||
async ({
|
||
sessionEntry,
|
||
command,
|
||
legacyFallback,
|
||
expectResolverContext,
|
||
expectAttemptModel,
|
||
}) => {
|
||
setupSuccessfulAttempt("openai", "channel-model");
|
||
state.runtimeConfigMock = createChannelModelRuntimeConfig();
|
||
state.sessionEntryMock = sessionEntry;
|
||
state.hasLegacyAutoFallbackWithoutOriginMock.mockReturnValue(legacyFallback);
|
||
|
||
await (command ? agentCommand(command) : runBasicAgentCommand());
|
||
|
||
if (expectResolverContext) {
|
||
expect(mockCallArg(state.resolveChannelModelOverrideMock)).toMatchObject({
|
||
channel: "discord",
|
||
groupId: "channel-123",
|
||
});
|
||
}
|
||
if (expectAttemptModel) {
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
providerOverride: "openai",
|
||
modelOverride: "channel-model",
|
||
});
|
||
}
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.provider).toBe("openai");
|
||
expect(fallbackParams.model).toBe("channel-model");
|
||
},
|
||
);
|
||
|
||
it("uses a concurrent user override adopted during legacy fallback repair", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.applyModelOverrideToSessionEntryMock.mockImplementation((params: unknown) => {
|
||
const { entry } = params as { entry: SessionEntry };
|
||
delete entry.providerOverride;
|
||
delete entry.modelOverride;
|
||
delete entry.modelOverrideSource;
|
||
return { updated: true };
|
||
});
|
||
state.hasLegacyAutoFallbackWithoutOriginMock.mockImplementation(
|
||
(entry: unknown) =>
|
||
(entry as SessionEntry | undefined)?.modelOverride === "stale-fallback-model",
|
||
);
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
model: "anthropic/default-model",
|
||
models: {
|
||
"anthropic/default-model": {},
|
||
"anthropic/stale-fallback-model": {},
|
||
"google/gemini-3-pro": {},
|
||
},
|
||
},
|
||
},
|
||
};
|
||
const sessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
providerOverride: "anthropic",
|
||
modelOverride: "stale-fallback-model",
|
||
modelOverrideSource: "auto",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
} satisfies SessionEntry;
|
||
state.sessionEntryMock = sessionEntry;
|
||
state.sessionStoreMock = { "agent:main:main": sessionEntry };
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
state.persistSessionEntryMock.mockImplementation(async (...args: unknown[]) => {
|
||
const params = args[0] as { entry?: SessionEntry };
|
||
if (params.entry?.modelOverride === "stale-fallback-model") {
|
||
return params.entry;
|
||
}
|
||
return {
|
||
...sessionEntry,
|
||
updatedAt: 2,
|
||
providerOverride: "google",
|
||
modelOverride: "gemini-3-pro",
|
||
modelOverrideSource: "user",
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("google", "gemini-3-pro"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.provider).toBe("google");
|
||
expect(fallbackParams.model).toBe("gemini-3-pro");
|
||
});
|
||
|
||
it("probes the channel primary when a session is pinned to an auto fallback", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.resolveAutoFallbackPrimaryProbeMock.mockReturnValue({
|
||
provider: "openai",
|
||
model: "channel-model",
|
||
fallbackProvider: "anthropic",
|
||
fallbackModel: "fallback-model",
|
||
});
|
||
state.runtimeConfigMock = createChannelModelRuntimeConfig({
|
||
additionalModels: { "anthropic/fallback-model": {} },
|
||
});
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
channel: "discord",
|
||
groupId: "channel-123",
|
||
providerOverride: "anthropic",
|
||
modelOverride: "fallback-model",
|
||
modelOverrideSource: "auto",
|
||
modelOverrideFallbackOriginProvider: "openai",
|
||
modelOverrideFallbackOriginModel: "channel-model",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "channel-model"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(mockCallArg(state.resolveAutoFallbackPrimaryProbeMock), {
|
||
primaryProvider: "openai",
|
||
primaryModel: "channel-model",
|
||
});
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.provider).toBe("openai");
|
||
expect(fallbackParams.model).toBe("channel-model");
|
||
});
|
||
|
||
it("uses current threaded session key for parent channel model overrides", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runtimeConfigMock = createChannelModelRuntimeConfig({
|
||
channel: "slack",
|
||
matchKey: "general",
|
||
model: "openai/parent-channel-model",
|
||
});
|
||
state.resolvedSessionKeyMock = "agent:main:slack:channel:general:thread:thread-1";
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
channel: "slack",
|
||
groupId: "thread-1",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(
|
||
makeSuccessResult("openai", "parent-channel-model"),
|
||
);
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.provider).toBe("openai");
|
||
expect(fallbackParams.model).toBe("parent-channel-model");
|
||
});
|
||
|
||
it("keeps stored session model overrides ahead of channel model overrides", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runtimeConfigMock = createChannelModelRuntimeConfig({
|
||
additionalModels: { "anthropic/stored-model": {} },
|
||
});
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
channel: "discord",
|
||
groupId: "channel-123",
|
||
providerOverride: "anthropic",
|
||
modelOverride: "stored-model",
|
||
modelOverrideSource: "user",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "stored-model"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.provider).toBe("anthropic");
|
||
expect(fallbackParams.model).toBe("stored-model");
|
||
});
|
||
|
||
it("keeps explicit run model overrides ahead of channel model overrides", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runtimeConfigMock = createChannelModelRuntimeConfig({
|
||
additionalModels: { "openai/explicit-model": {} },
|
||
});
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
channel: "discord",
|
||
groupId: "channel-123",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "explicit-model"));
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
model: "openai/explicit-model",
|
||
allowModelOverride: true,
|
||
});
|
||
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.provider).toBe("openai");
|
||
expect(fallbackParams.model).toBe("explicit-model");
|
||
});
|
||
|
||
it("uses rotated session identity for all post-run session persistence", async () => {
|
||
setupSingleAttemptFallback();
|
||
setupStoredSession();
|
||
const rotatedEntry: SessionEntry = {
|
||
sessionId: "rotated-session",
|
||
sessionFile: "/tmp/rotated-session.jsonl",
|
||
updatedAt: 2,
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
};
|
||
const result = makeSuccessResult("openai", "gpt-5.4") as ReturnType<
|
||
typeof makeSuccessResult
|
||
> & {
|
||
meta: Record<string, unknown> & { agentMeta: Record<string, unknown> };
|
||
};
|
||
result.meta.executionTrace = {
|
||
runner: "embedded",
|
||
fallbackUsed: false,
|
||
winnerProvider: "openai",
|
||
winnerModel: "gpt-5.4",
|
||
};
|
||
result.meta.finalAssistantVisibleText = "ok";
|
||
result.meta.agentMeta = {
|
||
...result.meta.agentMeta,
|
||
sessionId: "rotated-session",
|
||
sessionFile: "/tmp/rotated-session.jsonl",
|
||
};
|
||
state.runAgentAttemptMock.mockResolvedValue(result);
|
||
state.updateSessionStoreAfterAgentRunMock.mockImplementation(async () => {
|
||
state.sessionStoreMock = { "agent:main:main": rotatedEntry };
|
||
});
|
||
state.persistCliTurnTranscriptMock.mockResolvedValue({
|
||
kind: "persisted",
|
||
sessionEntry: rotatedEntry,
|
||
});
|
||
state.runCliTurnCompactionLifecycleMock.mockResolvedValue(rotatedEntry);
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(mockCallArg(state.updateSessionStoreAfterAgentRunMock), {
|
||
sessionId: "rotated-session",
|
||
});
|
||
expectRecordFields(mockCallArg(state.persistCliTurnTranscriptMock), {
|
||
sessionId: "rotated-session",
|
||
sessionKey: "agent:main:main",
|
||
});
|
||
expectRecordFields(mockCallArg(state.runCliTurnCompactionLifecycleMock), {
|
||
sessionId: "rotated-session",
|
||
sessionKey: "agent:main:main",
|
||
});
|
||
expectRecordFields(mockCallArg(state.deliverAgentCommandResultMock), {
|
||
expectedSessionIdForFreshDelivery: "rotated-session",
|
||
});
|
||
});
|
||
|
||
it("skips post-run persistence after the session is deleted", async () => {
|
||
setupSingleAttemptFallback();
|
||
setupStoredSession();
|
||
const result = makeSuccessResult("openai", "gpt-5.4") as ReturnType<
|
||
typeof makeSuccessResult
|
||
> & {
|
||
meta: Record<string, unknown> & { executionTrace: Record<string, unknown> };
|
||
};
|
||
result.meta.executionTrace = {
|
||
runner: "cli",
|
||
fallbackUsed: false,
|
||
winnerProvider: "openai",
|
||
winnerModel: "gpt-5.4",
|
||
};
|
||
state.runAgentAttemptMock.mockResolvedValue(result);
|
||
state.persistCliTurnTranscriptMock.mockResolvedValue({
|
||
kind: "session-rebound",
|
||
sessionEntry: undefined,
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(state.persistCliTurnTranscriptMock).toHaveBeenCalledTimes(1);
|
||
expect(state.runCliTurnCompactionLifecycleMock).not.toHaveBeenCalled();
|
||
expect(state.deliverAgentCommandResultMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it("preserves restart recovery ownership when delivery fails after a session rebound", async () => {
|
||
setupSingleAttemptFallback();
|
||
setupStoredSession();
|
||
const sessionStore = state.sessionStoreMock as Record<string, SessionEntry>;
|
||
const claimedEntry = expectDefined(
|
||
sessionStore["agent:main:main"],
|
||
"preclaimed restart recovery session",
|
||
);
|
||
claimedEntry.restartRecoveryDeliveryRunId = "session-1";
|
||
claimedEntry.restartRecoveryDeliveryContext = {
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
accountId: "main",
|
||
};
|
||
const result = makeSuccessResult("openai", "gpt-5.4") as ReturnType<
|
||
typeof makeSuccessResult
|
||
> & {
|
||
meta: Record<string, unknown> & { executionTrace: Record<string, unknown> };
|
||
};
|
||
result.meta.executionTrace = {
|
||
runner: "cli",
|
||
fallbackUsed: false,
|
||
winnerProvider: "openai",
|
||
winnerModel: "gpt-5.4",
|
||
};
|
||
state.runAgentAttemptMock.mockResolvedValue(result);
|
||
state.persistCliTurnTranscriptMock.mockResolvedValue({
|
||
kind: "session-rebound",
|
||
sessionEntry: undefined,
|
||
});
|
||
state.deliverAgentCommandResultMock.mockRejectedValue(new Error("delivery failed"));
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
accountId: "main",
|
||
deliver: true,
|
||
}),
|
||
).rejects.toThrow("delivery failed");
|
||
|
||
expect(sessionStore["agent:main:main"]?.restartRecoveryDeliveryRunId).toBe("session-1");
|
||
});
|
||
|
||
it.each([
|
||
{
|
||
name: "persists only the CLI assistant reply after the runner persists the current user turn",
|
||
runner: "cli",
|
||
canonicalUserRecorder: true,
|
||
embeddedAssistantGapFill: false,
|
||
},
|
||
{
|
||
name: "persists the full embedded turn when no canonical user recorder exists",
|
||
runner: "embedded",
|
||
canonicalUserRecorder: false,
|
||
embeddedAssistantGapFill: true,
|
||
},
|
||
])("$name", async ({ runner, canonicalUserRecorder, embeddedAssistantGapFill }) => {
|
||
type AttemptCall = {
|
||
onUserMessagePersisted?: () => void;
|
||
};
|
||
setupSingleAttemptFallback();
|
||
setupStoredSession();
|
||
const result = makeSuccessResult("openai", "gpt-5.4") as ReturnType<
|
||
typeof makeSuccessResult
|
||
> & {
|
||
meta: Record<string, unknown> & { executionTrace: Record<string, unknown> };
|
||
};
|
||
result.meta.executionTrace = {
|
||
runner,
|
||
fallbackUsed: false,
|
||
winnerProvider: "openai",
|
||
winnerModel: "gpt-5.4",
|
||
};
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: AttemptCall) => {
|
||
attemptParams.onUserMessagePersisted?.();
|
||
return result;
|
||
});
|
||
state.persistCliTurnTranscriptMock.mockResolvedValue({
|
||
kind: "persisted",
|
||
sessionEntry: state.sessionEntryMock,
|
||
});
|
||
|
||
await (canonicalUserRecorder
|
||
? agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
userTurnTranscriptRecorder: {} as NonNullable<
|
||
Parameters<typeof agentCommand>[0]["userTurnTranscriptRecorder"]
|
||
>,
|
||
})
|
||
: runBasicAgentCommand());
|
||
|
||
expectRecordFields(mockCallArg(state.persistCliTurnTranscriptMock), {
|
||
embeddedAssistantGapFill,
|
||
});
|
||
});
|
||
|
||
it("does not treat backend CLI session id as OpenClaw session identity", async () => {
|
||
setupSingleAttemptFallback();
|
||
setupStoredSession();
|
||
const result = makeSuccessResult("openai", "gpt-5.4") as ReturnType<
|
||
typeof makeSuccessResult
|
||
> & {
|
||
meta: Record<string, unknown> & { agentMeta: Record<string, unknown> };
|
||
};
|
||
result.meta.agentMeta = {
|
||
...result.meta.agentMeta,
|
||
sessionId: "backend-cli-session",
|
||
};
|
||
state.runAgentAttemptMock.mockResolvedValue(result);
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(mockCallArg(state.updateSessionStoreAfterAgentRunMock), {
|
||
sessionId: "session-1",
|
||
});
|
||
expectRecordFields(mockCallArg(state.deliverAgentCommandResultMock), {
|
||
expectedSessionIdForFreshDelivery: "session-1",
|
||
});
|
||
});
|
||
|
||
it("scopes explicit-agent sentinel store keys before command routing", () => {
|
||
expect(
|
||
agentCommandTesting.resolveExplicitAgentCommandSessionKey({
|
||
rawExplicitSessionKey: "global",
|
||
agentIdOverride: "work",
|
||
cfg: {},
|
||
}),
|
||
).toBe("agent:work:global");
|
||
expect(
|
||
agentCommandTesting.resolveExplicitAgentCommandSessionKey({
|
||
rawExplicitSessionKey: "main",
|
||
agentIdOverride: "work",
|
||
cfg: {},
|
||
}),
|
||
).toBe("agent:work:main");
|
||
});
|
||
|
||
it("persists explicit overrides even when ingress skips the initial touch", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupStoredSession();
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
thinking: "medium",
|
||
skipInitialSessionTouch: true,
|
||
});
|
||
|
||
const touchWrite = state.persistSessionEntryMock.mock.calls.find((call) => {
|
||
const entry = (call[0] as { entry?: Record<string, unknown> } | undefined)?.entry;
|
||
return entry?.thinkingLevel === "medium";
|
||
})?.[0] as { entry?: Record<string, unknown> } | undefined;
|
||
expect(touchWrite?.entry?.lastInteractionAt).toBeDefined();
|
||
expect(state.updateSessionStoreAfterAgentRunMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it("forwards an explicit OpenClaw runtime override into fallback and attempt execution", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
model: { primary: "openai/gpt-5.4" },
|
||
models: { "openai/gpt-5.4": {} },
|
||
},
|
||
},
|
||
};
|
||
setupStoredSession({ agentRuntimeOverride: "openclaw", agentHarnessId: "codex" });
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.resolveAgentHarnessRuntimeOverride?.("openai", "gpt-5.4")).toBe(
|
||
"openclaw",
|
||
);
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
providerOverride: "openai",
|
||
modelOverride: "gpt-5.4",
|
||
agentHarnessRuntimeOverride: "openclaw",
|
||
});
|
||
});
|
||
|
||
it("does not persist turn-local thinking fallback over a stored session override", async () => {
|
||
setupSingleAttemptFallback();
|
||
const { entry: sessionEntry, store: sessionStore } = setupStoredSession({
|
||
thinkingLevel: "high",
|
||
});
|
||
state.isThinkingLevelSupportedMock.mockReturnValue(false);
|
||
state.resolveSupportedThinkingLevelMock.mockReturnValue("off");
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
resolvedThinkLevel: "off",
|
||
});
|
||
expect(sessionEntry.thinkingLevel).toBe("high");
|
||
expect(sessionStore["agent:main:main"]?.thinkingLevel).toBe("high");
|
||
expect(state.persistSessionEntryMock).not.toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
entry: expect.objectContaining({ thinkingLevel: "off" }),
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("revalidates immutable Ultra for each model fallback without persisting the remap", async () => {
|
||
const { entry: sessionEntry } = setupStoredSession({ thinkingLevel: "ultra" });
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
model: { primary: "openai/gpt-5.6-luna" },
|
||
models: {
|
||
"openai/gpt-5.6-luna": { agentRuntime: { id: "codex" } },
|
||
"openai/gpt-5.6-sol": { agentRuntime: { id: "codex" } },
|
||
},
|
||
},
|
||
},
|
||
};
|
||
state.isThinkingLevelSupportedMock.mockImplementation((args: unknown) => {
|
||
const { model, level } = args as { model?: string; level?: string };
|
||
return model !== "gpt-5.6-luna" || level !== "ultra";
|
||
});
|
||
state.resolveSupportedThinkingLevelMock.mockImplementation(
|
||
({ level, model }: { level?: string; model?: string }) =>
|
||
model === "gpt-5.6-luna" && level === "ultra" ? "max" : level,
|
||
);
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
await params.run(params.provider, params.model);
|
||
const result = await params.run("openai", "gpt-5.6-sol");
|
||
return {
|
||
result,
|
||
provider: "openai",
|
||
model: "gpt-5.6-sol",
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(
|
||
async (params: { providerOverride: string; modelOverride: string }) =>
|
||
makeSuccessResult(params.providerOverride, params.modelOverride),
|
||
);
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock, 0), {
|
||
modelOverride: "gpt-5.6-luna",
|
||
resolvedThinkLevel: "max",
|
||
});
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock, 1), {
|
||
modelOverride: "gpt-5.6-sol",
|
||
resolvedThinkLevel: "ultra",
|
||
});
|
||
expect(state.resolveSupportedThinkingLevelMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
provider: "openai",
|
||
model: "gpt-5.6-luna",
|
||
level: "ultra",
|
||
agentRuntime: "codex",
|
||
}),
|
||
);
|
||
expect(sessionEntry.thinkingLevel).toBe("ultra");
|
||
expect(state.persistSessionEntryMock).not.toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
entry: expect.objectContaining({ thinkingLevel: "max" }),
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("recomputes a model-derived thinking default for each fallback candidate", async () => {
|
||
setupStoredSession();
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
model: { primary: "openai/gpt-5.6-sol" },
|
||
models: {
|
||
"openai/gpt-5.6-sol": { agentRuntime: { id: "codex" } },
|
||
"openai/gpt-5.6-terra": { agentRuntime: { id: "codex" } },
|
||
},
|
||
},
|
||
},
|
||
};
|
||
state.resolveThinkingDefaultMock.mockImplementation((args: unknown) => {
|
||
const { model } = args as { model?: string };
|
||
return model === "gpt-5.6-terra" ? "medium" : "low";
|
||
});
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
await params.run(params.provider, params.model);
|
||
const result = await params.run("openai", "gpt-5.6-terra");
|
||
return {
|
||
result,
|
||
provider: "openai",
|
||
model: "gpt-5.6-terra",
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(
|
||
async (params: { providerOverride: string; modelOverride: string }) =>
|
||
makeSuccessResult(params.providerOverride, params.modelOverride),
|
||
);
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock, 0), {
|
||
modelOverride: "gpt-5.6-sol",
|
||
resolvedThinkLevel: "low",
|
||
});
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock, 1), {
|
||
modelOverride: "gpt-5.6-terra",
|
||
resolvedThinkLevel: "medium",
|
||
});
|
||
});
|
||
|
||
it("persists and clears current run delivery context for restart recovery", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession();
|
||
state.deliverAgentCommandResultMock.mockResolvedValue({ deliverySucceeded: true });
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
accountId: "main",
|
||
threadId: "reply-1",
|
||
deliver: true,
|
||
});
|
||
|
||
const persistedContexts = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoveryDeliveryContext;
|
||
});
|
||
expect(persistedContexts).toContainEqual({
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
accountId: "main",
|
||
threadId: "reply-1",
|
||
});
|
||
const cleanupParams = state.persistSessionEntryMock.mock.calls.at(-1)?.[0] as
|
||
| { sessionStore?: Record<string, SessionEntry> }
|
||
| undefined;
|
||
const stored = cleanupParams?.sessionStore?.["agent:main:main"];
|
||
expect(stored?.restartRecoveryDeliveryContext).toBeUndefined();
|
||
});
|
||
|
||
it("records generated-media delivery runs as durable terminal sources", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession();
|
||
state.deliverAgentCommandResultMock.mockImplementation(async (params: unknown) => {
|
||
const onDeliveryResult = (params as { onDeliveryResult?: (result: unknown) => void })
|
||
.onDeliveryResult;
|
||
const deliveryResult = {
|
||
payloads: [{ isReasoning: true }, { text: "ready", mediaUrls: ["/tmp/payload.png"] }],
|
||
meta: {},
|
||
deliverySucceeded: true,
|
||
deliveryStatus: {
|
||
status: "partial_failed",
|
||
payloadOutcomes: [
|
||
{ index: 0, status: "suppressed" },
|
||
{ index: 1, status: "failed", sentBeforeError: false },
|
||
],
|
||
},
|
||
messagingToolSentTargets: [
|
||
{
|
||
provider: "discord",
|
||
to: "discord:dm:123",
|
||
threadId: 42,
|
||
mediaUrls: ["/tmp/proof.png"],
|
||
},
|
||
],
|
||
};
|
||
onDeliveryResult?.(deliveryResult);
|
||
return deliveryResult;
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "generated image ready",
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
deliver: false,
|
||
runId: "image:task-1:agent-loop",
|
||
sourceReplyDeliveryMode: "automatic",
|
||
disableMessageTool: true,
|
||
forceRestartSafeTools: true,
|
||
internalDeliveryMediaUrls: ["/tmp/payload.png"],
|
||
inputProvenance: {
|
||
kind: "inter_session",
|
||
sourceChannel: "webchat",
|
||
sourceTool: "image_generate",
|
||
},
|
||
});
|
||
|
||
const persistedSourceRunIds = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoveryDeliverySourceRunId;
|
||
});
|
||
expect(persistedSourceRunIds).toContain("image:task-1:agent-loop");
|
||
expect(
|
||
state.persistSessionEntryMock.mock.calls.some((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoverySourceReplyDeliveryMode === "automatic";
|
||
}),
|
||
).toBe(true);
|
||
const cleanupParams = state.persistSessionEntryMock.mock.calls.at(-1)?.[0] as
|
||
| { sessionStore?: Record<string, SessionEntry> }
|
||
| undefined;
|
||
const stored = cleanupParams?.sessionStore?.["agent:main:main"];
|
||
expect(stored?.restartRecoveryDeliverySourceRunId).toBeUndefined();
|
||
expect(stored?.restartRecoverySourceReplyDeliveryMode).toBeUndefined();
|
||
expect(stored?.restartRecoveryTerminalRunIds).toEqual(["image:task-1:agent-loop"]);
|
||
expect(stored?.restartRecoveryTerminalDeliveryEvidence).toEqual([
|
||
{
|
||
runId: "image:task-1:agent-loop",
|
||
captured: true,
|
||
payloads: [{ visible: false }, { mediaUrls: ["/tmp/payload.png"], visible: true }],
|
||
deliveryStatus: {
|
||
status: "partial_failed",
|
||
payloadOutcomes: [
|
||
{ index: 0, status: "suppressed" },
|
||
{ index: 1, status: "failed", sentBeforeError: false },
|
||
],
|
||
},
|
||
messagingToolSentTargets: [
|
||
{
|
||
provider: "discord",
|
||
to: "discord:dm:123",
|
||
threadId: "42",
|
||
mediaUrls: ["/tmp/proof.png"],
|
||
visible: true,
|
||
},
|
||
],
|
||
restartUnsafeSideEffectsDetected: true,
|
||
},
|
||
]);
|
||
});
|
||
|
||
it("does not make an unconstrained message-tool completion replayable", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession();
|
||
|
||
await agentCommand({
|
||
message: "generated image ready",
|
||
sessionKey: "agent:main:main",
|
||
deliver: false,
|
||
runId: "image:unsafe-message-tool:agent-loop",
|
||
sourceReplyDeliveryMode: "message_tool_only",
|
||
inputProvenance: {
|
||
kind: "inter_session",
|
||
sourceChannel: "webchat",
|
||
sourceTool: "image_generate",
|
||
},
|
||
});
|
||
|
||
expect(
|
||
state.persistSessionEntryMock.mock.calls.some((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoveryDeliverySourceRunId !== undefined;
|
||
}),
|
||
).toBe(false);
|
||
});
|
||
|
||
it("constrains recovery delivery to host-owned media before persistence and send", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue({
|
||
...makeSuccessResult("openai", "gpt-5.4"),
|
||
payloads: [{ text: "ready", mediaUrls: ["/tmp/already-delivered.png"] }],
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "deliver only missing generated media",
|
||
channel: "discord",
|
||
to: "channel:123",
|
||
deliver: true,
|
||
sourceReplyDeliveryMode: "automatic",
|
||
disableMessageTool: true,
|
||
forceRestartSafeTools: true,
|
||
internalDeliveryMediaUrls: ["/tmp/missing.png"],
|
||
runId: "image:task-policy:agent-loop",
|
||
inputProvenance: {
|
||
kind: "inter_session",
|
||
sourceChannel: "webchat",
|
||
sourceTool: "image_generate",
|
||
},
|
||
});
|
||
|
||
const deliveryParams = requireRecord(
|
||
mockCallArg(state.deliverAgentCommandResultMock),
|
||
"delivery params",
|
||
);
|
||
expect(requireRecord(deliveryParams.result, "delivery result").payloads).toEqual([
|
||
{ text: "ready" },
|
||
{ mediaUrls: ["/tmp/missing.png"], trustedLocalMedia: true },
|
||
]);
|
||
expect(deliveryParams.payloads).toEqual([
|
||
{ text: "ready" },
|
||
{ mediaUrls: ["/tmp/missing.png"], trustedLocalMedia: true },
|
||
]);
|
||
expect(
|
||
state.persistSessionEntryMock.mock.calls.some((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return (
|
||
params.entry?.restartRecoveryDeliveryMediaUrls?.[0] === "/tmp/missing.png" &&
|
||
params.entry.restartRecoveryDisableMessageTool === true &&
|
||
params.entry.restartRecoveryForceSafeTools === true
|
||
);
|
||
}),
|
||
).toBe(true);
|
||
});
|
||
|
||
it("restores the exact generated-media policy for a preclaimed recovery run", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue({
|
||
...makeSuccessResult("openai", "gpt-5.4"),
|
||
payloads: [{ text: "ready", mediaUrls: ["/tmp/model-selected.png"] }],
|
||
});
|
||
setupBareStoredSession({
|
||
restartRecoveryDeliveryRunId: "recovery-run",
|
||
restartRecoveryDeliverySourceRunId: "image:task-policy:agent-loop",
|
||
restartRecoveryDeliveryContext: { channel: "discord", to: "channel:123" },
|
||
restartRecoveryDeliveryMediaUrls: ["/tmp/missing.png"],
|
||
restartRecoveryDisableMessageTool: true,
|
||
restartRecoverySuppressTextDelivery: true,
|
||
restartRecoverySourceReplyDeliveryMode: "automatic",
|
||
restartRecoveryForceSafeTools: true,
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "continue generated media delivery",
|
||
sessionKey: "agent:main:main",
|
||
channel: "discord",
|
||
to: "channel:123",
|
||
deliver: true,
|
||
runId: "recovery-run",
|
||
});
|
||
|
||
const deliveryParams = requireRecord(
|
||
mockCallArg(state.deliverAgentCommandResultMock),
|
||
"delivery params",
|
||
);
|
||
expect(requireRecord(deliveryParams.result, "delivery result").payloads).toEqual([
|
||
{ mediaUrls: ["/tmp/missing.png"], trustedLocalMedia: true },
|
||
]);
|
||
const persistedPolicies = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return {
|
||
mediaUrls: params.entry?.restartRecoveryDeliveryMediaUrls,
|
||
disableMessageTool: params.entry?.restartRecoveryDisableMessageTool,
|
||
forceSafeTools: params.entry?.restartRecoveryForceSafeTools,
|
||
suppressText: params.entry?.restartRecoverySuppressTextDelivery,
|
||
};
|
||
});
|
||
expect(persistedPolicies).toContainEqual({
|
||
mediaUrls: undefined,
|
||
disableMessageTool: undefined,
|
||
forceSafeTools: undefined,
|
||
suppressText: undefined,
|
||
});
|
||
});
|
||
|
||
it("rejects host-owned media constraints without the scoped recovery policy", async () => {
|
||
await expect(
|
||
agentCommand({
|
||
message: "unsafe delivery constraint",
|
||
sessionKey: "agent:main:main",
|
||
internalDeliveryMediaUrls: ["/tmp/proof.png"],
|
||
}),
|
||
).rejects.toThrow(
|
||
"internal delivery media constraints require automatic delivery with restart-safe tools and no message tool",
|
||
);
|
||
|
||
expect(state.runAgentAttemptMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("retains and clears a preclaimed transcript-only recovery run", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession({
|
||
restartRecoveryDeliveryRunId: "recovery-run",
|
||
restartRecoveryDeliverySourceRunId: "control-ui-run",
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "continue after restart",
|
||
sessionKey: "agent:main:main",
|
||
deliver: false,
|
||
runId: "recovery-run",
|
||
});
|
||
|
||
const persistedClaims = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return {
|
||
context: params.entry?.restartRecoveryDeliveryContext,
|
||
runId: params.entry?.restartRecoveryDeliveryRunId,
|
||
sourceRunId: params.entry?.restartRecoveryDeliverySourceRunId,
|
||
};
|
||
});
|
||
expect(persistedClaims).toContainEqual({
|
||
context: undefined,
|
||
runId: "recovery-run",
|
||
sourceRunId: "control-ui-run",
|
||
});
|
||
expect(persistedClaims.at(-1)).toEqual({
|
||
context: undefined,
|
||
runId: undefined,
|
||
sourceRunId: undefined,
|
||
});
|
||
const cleanupParams = state.persistSessionEntryMock.mock.calls.at(-1)?.[0] as
|
||
| {
|
||
sessionStore?: Record<string, SessionEntry>;
|
||
}
|
||
| undefined;
|
||
const stored = cleanupParams?.sessionStore?.["agent:main:main"];
|
||
expect(stored?.restartRecoveryDeliveryContext).toBeUndefined();
|
||
expect(stored?.restartRecoveryDeliveryRunId).toBeUndefined();
|
||
expect(stored?.restartRecoveryDeliverySourceRunId).toBeUndefined();
|
||
expect(stored?.restartRecoveryTerminalRunIds).toEqual(["control-ui-run", "recovery-run"]);
|
||
});
|
||
|
||
it("refreshes delivery session entries through the session accessor", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession();
|
||
const freshEntry = createCommandSessionEntry({
|
||
sessionId: "session-1",
|
||
updatedAt: 2,
|
||
deliveryContext: {
|
||
channel: "discord",
|
||
to: "discord:dm:sqlite",
|
||
accountId: "main",
|
||
},
|
||
});
|
||
state.loadSessionEntryMock.mockReturnValue(freshEntry);
|
||
state.deliverAgentCommandResultMock.mockImplementation(async (params: unknown) => {
|
||
const resolver = (
|
||
params as {
|
||
resolveFreshSessionEntryForDelivery?: () => Promise<SessionEntry | undefined>;
|
||
}
|
||
).resolveFreshSessionEntryForDelivery;
|
||
const resolved = await resolver?.();
|
||
expect(resolved).toEqual(freshEntry);
|
||
return { deliverySucceeded: true };
|
||
});
|
||
|
||
await runDiscordDelivery();
|
||
|
||
expect(state.loadSessionEntryMock).toHaveBeenCalledWith({
|
||
storePath: "/tmp/openclaw-sessions.json",
|
||
sessionKey: "agent:main:main",
|
||
readConsistency: "latest",
|
||
clone: false,
|
||
});
|
||
});
|
||
|
||
it("preserves parsed explicit target threads for restart recovery", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession();
|
||
state.deliverAgentCommandResultMock.mockResolvedValue({ deliverySucceeded: true });
|
||
state.resolveAgentDeliveryPlanMock.mockReturnValueOnce({
|
||
baseDelivery: {
|
||
mode: "explicit",
|
||
threadId: "thread-1",
|
||
threadIdSource: "explicit",
|
||
},
|
||
resolvedChannel: "discord",
|
||
resolvedTo: "discord:channel:general",
|
||
resolvedAccountId: "main",
|
||
resolvedThreadId: "thread-1",
|
||
deliveryTargetMode: "explicit",
|
||
});
|
||
|
||
await runDiscordDelivery({ to: "discord:channel:general/thread:thread-1" });
|
||
|
||
const persistedContexts = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoveryDeliveryContext;
|
||
});
|
||
expect(persistedContexts).toContainEqual({
|
||
channel: "discord",
|
||
to: "discord:channel:general",
|
||
accountId: "main",
|
||
threadId: "thread-1",
|
||
});
|
||
});
|
||
|
||
it("does not inherit a stale thread when restart recovery uses an explicit target", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession({
|
||
lastThreadId: "stale-thread",
|
||
});
|
||
state.deliverAgentCommandResultMock.mockResolvedValue({ deliverySucceeded: true });
|
||
|
||
await runDiscordDelivery();
|
||
|
||
const persistedContexts = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoveryDeliveryContext;
|
||
});
|
||
expect(persistedContexts).toContainEqual({
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
accountId: "main",
|
||
});
|
||
});
|
||
|
||
it("persists implicit session delivery route for restart recovery", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession({
|
||
delivery: normalizeSessionDeliveryState({
|
||
context: {
|
||
channel: "discord",
|
||
to: "discord:channel:general",
|
||
accountId: "main",
|
||
threadId: "thread-1",
|
||
},
|
||
}),
|
||
});
|
||
state.deliverAgentCommandResultMock.mockResolvedValue({ deliverySucceeded: true });
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
sessionKey: "agent:main:main",
|
||
deliver: true,
|
||
});
|
||
|
||
const persistedContexts = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoveryDeliveryContext;
|
||
});
|
||
expect(persistedContexts).toContainEqual({
|
||
channel: "discord",
|
||
to: "discord:channel:general",
|
||
accountId: "main",
|
||
threadId: "thread-1",
|
||
});
|
||
expect(state.resolveAgentDeliveryPlanMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
explicitTo: undefined,
|
||
requestedChannel: undefined,
|
||
sessionEntry: expect.objectContaining({
|
||
delivery: expect.objectContaining({
|
||
context: expect.objectContaining({ to: "discord:channel:general" }),
|
||
}),
|
||
}),
|
||
wantsDelivery: true,
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("persists default target delivery route for restart recovery", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession();
|
||
state.deliverAgentCommandResultMock.mockResolvedValue({ deliverySucceeded: true });
|
||
state.resolveMessageChannelSelectionMock.mockResolvedValue({
|
||
channel: "discord",
|
||
configured: ["discord"],
|
||
source: "single-configured",
|
||
});
|
||
state.resolveAgentOutboundTargetMock.mockReturnValue({
|
||
resolvedTarget: { ok: true, to: "discord:channel:default" },
|
||
resolvedTo: "discord:channel:default",
|
||
targetMode: "implicit",
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
sessionKey: "agent:main:main",
|
||
deliver: true,
|
||
});
|
||
|
||
const persistedContexts = state.persistSessionEntryMock.mock.calls.map((call) => {
|
||
const params = call[0] as { entry?: SessionEntry };
|
||
return params.entry?.restartRecoveryDeliveryContext;
|
||
});
|
||
expect(persistedContexts).toContainEqual({
|
||
channel: "discord",
|
||
to: "discord:channel:default",
|
||
});
|
||
});
|
||
|
||
it.each([
|
||
{
|
||
name: "does not overwrite another active run's restart recovery context",
|
||
attemptResult: makeEmptyResult("openai", "gpt-5.4"),
|
||
},
|
||
{
|
||
name: "does not clear another active run's restart recovery context",
|
||
attemptResult: makeEmptyResult("openai", "gpt-5.4"),
|
||
},
|
||
])("$name", async ({ attemptResult }) => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(attemptResult);
|
||
const staleEntry = createCommandSessionEntry();
|
||
const laterRunEntry = createCommandSessionEntry({
|
||
updatedAt: 2,
|
||
restartRecoveryDeliveryContext: {
|
||
channel: "discord",
|
||
to: "discord:dm:456",
|
||
accountId: "main",
|
||
},
|
||
restartRecoveryDeliveryRunId: "later-run",
|
||
});
|
||
const sessionStore = { "agent:main:main": laterRunEntry };
|
||
state.sessionEntryMock = staleEntry;
|
||
state.sessionStoreMock = sessionStore;
|
||
state.storePathMock = "/tmp/openclaw-sessions.json";
|
||
|
||
await runDiscordDelivery({ sessionKey: "agent:main:main", runId: "stale-run" });
|
||
|
||
expect(sessionStore["agent:main:main"]?.restartRecoveryDeliveryContext).toEqual(
|
||
laterRunEntry.restartRecoveryDeliveryContext,
|
||
);
|
||
expect(sessionStore["agent:main:main"]?.restartRecoveryDeliveryRunId).toBe("later-run");
|
||
});
|
||
|
||
it("keeps current run delivery context when restart marker wins the cleanup race", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
const { store: sessionStore } = setupBareStoredSession();
|
||
state.deliverAgentCommandResultMock.mockImplementation(async () => {
|
||
const current = sessionStore["agent:main:main"];
|
||
if (current) {
|
||
current.abortedLastRun = true;
|
||
}
|
||
return { deliverySucceeded: false };
|
||
});
|
||
|
||
await runDiscordDelivery();
|
||
|
||
expect(sessionStore["agent:main:main"]?.abortedLastRun).toBe(true);
|
||
expect(state.persistSessionEntryMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
entry: expect.objectContaining({
|
||
restartRecoveryDeliveryContext: {
|
||
channel: "discord",
|
||
to: "discord:dm:123",
|
||
accountId: "main",
|
||
},
|
||
restartRecoveryDeliveryRunId: "session-1",
|
||
}),
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("does not recreate a deleted session entry during restart recovery cleanup", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeEmptyResult("openai", "gpt-5.4"));
|
||
const { store: sessionStore } = setupBareStoredSession();
|
||
state.deliverAgentCommandResultMock.mockImplementation(async () => {
|
||
delete sessionStore["agent:main:main"];
|
||
return { deliverySucceeded: true };
|
||
});
|
||
|
||
await runDiscordDelivery();
|
||
|
||
expect(sessionStore["agent:main:main"]).toBeUndefined();
|
||
});
|
||
|
||
it("does not clear restart recovery context from a rotated session entry", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeEmptyResult("openai", "gpt-5.4"));
|
||
const { store: sessionStore } = setupBareStoredSession();
|
||
const rotatedEntry = createCommandSessionEntry({
|
||
sessionId: "session-2",
|
||
updatedAt: 2,
|
||
restartRecoveryDeliveryContext: {
|
||
channel: "discord",
|
||
to: "discord:dm:456",
|
||
accountId: "main",
|
||
},
|
||
});
|
||
state.deliverAgentCommandResultMock.mockImplementation(async () => {
|
||
sessionStore["agent:main:main"] = rotatedEntry;
|
||
return { deliverySucceeded: true };
|
||
});
|
||
|
||
await runDiscordDelivery();
|
||
|
||
expect(sessionStore["agent:main:main"]).toEqual(rotatedEntry);
|
||
});
|
||
|
||
it("does not clear restart recovery context from another active run in the same session", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeEmptyResult("openai", "gpt-5.4"));
|
||
const { store: sessionStore } = setupBareStoredSession();
|
||
const laterRunEntry = createCommandSessionEntry({
|
||
updatedAt: 2,
|
||
restartRecoveryDeliveryContext: {
|
||
channel: "discord",
|
||
to: "discord:dm:456",
|
||
accountId: "main",
|
||
},
|
||
restartRecoveryDeliveryRunId: "later-run",
|
||
});
|
||
state.deliverAgentCommandResultMock.mockImplementation(async () => {
|
||
sessionStore["agent:main:main"] = laterRunEntry;
|
||
return { deliverySucceeded: false };
|
||
});
|
||
|
||
await runDiscordDelivery();
|
||
|
||
expect(sessionStore["agent:main:main"]).toEqual(laterRunEntry);
|
||
});
|
||
|
||
it("stores and delivers with the prepared canonical current-run target", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
setupBareStoredSession();
|
||
state.deliverAgentCommandResultMock.mockResolvedValue({ deliverySucceeded: false });
|
||
state.resolveAgentDeliveryPlanWithSessionRouteMock.mockResolvedValueOnce({
|
||
baseDelivery: {},
|
||
resolvedChannel: "discord",
|
||
resolvedTo: "channel:1524410080953634829",
|
||
resolvedAccountId: "main",
|
||
deliveryTargetMode: "explicit",
|
||
});
|
||
|
||
await runDiscordDelivery({ to: "channel:general" });
|
||
|
||
const pendingEntries = state.persistSessionEntryMock.mock.calls
|
||
.map((call) => (call[0] as { entry?: SessionEntry }).entry)
|
||
.filter((entry): entry is SessionEntry => entry?.pendingFinalDelivery === true);
|
||
expect(pendingEntries).toContainEqual(
|
||
expect.objectContaining({
|
||
pendingFinalDeliveryText: "ok",
|
||
pendingFinalDeliveryContext: {
|
||
channel: "discord",
|
||
to: "channel:1524410080953634829",
|
||
accountId: "main",
|
||
},
|
||
}),
|
||
);
|
||
expect(state.deliverAgentCommandResultMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
opts: expect.objectContaining({
|
||
replyChannel: "discord",
|
||
replyTo: "channel:1524410080953634829",
|
||
replyAccountId: "main",
|
||
deliveryTargetMode: "explicit",
|
||
}),
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("rejects a strict delivery target before the model run", async () => {
|
||
setupSingleAttemptFallback();
|
||
const targetError = new Error('Unknown Discord target "channel:missing"');
|
||
state.resolveAgentDeliveryPlanWithSessionRouteMock.mockResolvedValueOnce({
|
||
baseDelivery: {},
|
||
resolvedChannel: "discord",
|
||
resolvedTo: "channel:missing",
|
||
deliveryTargetMode: "explicit",
|
||
targetResolutionError: targetError,
|
||
});
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
channel: "discord",
|
||
to: "channel:missing",
|
||
deliver: true,
|
||
}),
|
||
).rejects.toBe(targetError);
|
||
|
||
expect(state.runAgentAttemptMock).not.toHaveBeenCalled();
|
||
expect(state.deliverAgentCommandResultMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("downgrades a rejected best-effort target to session-only before the model run", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
state.resolveAgentDeliveryPlanWithSessionRouteMock.mockResolvedValueOnce({
|
||
baseDelivery: {},
|
||
resolvedChannel: "discord",
|
||
resolvedTo: "channel:missing",
|
||
deliveryTargetMode: "explicit",
|
||
targetResolutionError: new Error('Unknown Discord target "channel:missing"'),
|
||
});
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
channel: "discord",
|
||
to: "channel:missing",
|
||
deliver: true,
|
||
bestEffortDeliver: true,
|
||
}),
|
||
).resolves.toBeUndefined();
|
||
|
||
expect(state.runAgentAttemptMock).toHaveBeenCalled();
|
||
expect(state.deliverAgentCommandResultMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({ opts: expect.objectContaining({ deliver: false }) }),
|
||
);
|
||
});
|
||
|
||
it("clears stale flag-only pending final delivery when there is no final payload", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeEmptyResult("openai", "gpt-5.4"));
|
||
|
||
setupBareStoredSession({
|
||
pendingFinalDelivery: true,
|
||
pendingFinalDeliveryCreatedAt: 2,
|
||
pendingFinalDeliveryLastAttemptAt: 3,
|
||
pendingFinalDeliveryAttemptCount: 4,
|
||
pendingFinalDeliveryLastError: "previous failure",
|
||
pendingFinalDeliveryContext: { channel: "tui" },
|
||
pendingFinalDeliveryIntentId: "intent-1",
|
||
});
|
||
state.deliverAgentCommandResultMock.mockResolvedValue(undefined);
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
channel: "whatsapp",
|
||
to: "+1234567890",
|
||
deliver: true,
|
||
});
|
||
|
||
expect(state.persistSessionEntryMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
entry: expect.objectContaining({
|
||
pendingFinalDelivery: undefined,
|
||
pendingFinalDeliveryText: undefined,
|
||
pendingFinalDeliveryCreatedAt: undefined,
|
||
pendingFinalDeliveryLastAttemptAt: undefined,
|
||
pendingFinalDeliveryAttemptCount: undefined,
|
||
pendingFinalDeliveryLastError: undefined,
|
||
pendingFinalDeliveryContext: undefined,
|
||
pendingFinalDeliveryIntentId: undefined,
|
||
}),
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("passes SQLite transcript markers to visible agent attempts", async () => {
|
||
setupSingleAttemptFallback();
|
||
const visibleEntry: SessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
sessionFile: "sqlite:default:session-1:/tmp/openclaw-session-store.json",
|
||
};
|
||
const sessionStore: Record<string, SessionEntry> = { "agent:main:main": visibleEntry };
|
||
state.sessionEntryMock = visibleEntry;
|
||
state.sessionStoreMock = sessionStore;
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
state.loadSessionEntryMock.mockReturnValue(visibleEntry);
|
||
const attemptCalls: Array<{ sessionFile?: string; sessionEntry?: SessionEntry }> = [];
|
||
state.runAgentAttemptMock.mockImplementation(async (params) => {
|
||
attemptCalls.push(params as { sessionFile?: string; sessionEntry?: SessionEntry });
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "visible run",
|
||
to: "+1234567890",
|
||
});
|
||
|
||
expect(attemptCalls).toHaveLength(1);
|
||
expect(attemptCalls[0]?.sessionFile).toBe(
|
||
"sqlite:default:session-1:/tmp/openclaw-session-store.json",
|
||
);
|
||
});
|
||
|
||
it("keeps internal session-effect CLI runs out of visible session state", async () => {
|
||
setupSingleAttemptFallback();
|
||
const visibleEntry: SessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: 1,
|
||
sessionFile: "/tmp/session.jsonl",
|
||
providerOverride: "anthropic",
|
||
modelOverride: "claude",
|
||
modelOverrideSource: "user",
|
||
skillsSnapshot: { prompt: "visible", skills: [{ name: "existing" }], version: 1 },
|
||
};
|
||
const sessionStore: Record<string, SessionEntry> = { "agent:main:main": visibleEntry };
|
||
state.sessionEntryMock = visibleEntry;
|
||
state.sessionStoreMock = sessionStore;
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
state.loadSessionEntryMock.mockReturnValue(visibleEntry);
|
||
const attemptCalls: Array<{ sessionFile?: string; sessionEntry?: SessionEntry }> = [];
|
||
state.runAgentAttemptMock.mockImplementation(async (params) => {
|
||
attemptCalls.push(params as { sessionFile?: string; sessionEntry?: SessionEntry });
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "internal resume",
|
||
to: "+1234567890",
|
||
sessionEffects: "internal",
|
||
suppressPromptPersistence: true,
|
||
});
|
||
|
||
expect(state.prepareInternalSessionEffectsSessionMock).toHaveBeenCalledWith({
|
||
agentId: "default",
|
||
cwd: "/tmp/workspace",
|
||
runId: "session-1",
|
||
source: {
|
||
agentId: "default",
|
||
sessionId: "session-1",
|
||
sessionKey: "agent:main:main",
|
||
storePath: "/tmp/openclaw-session-store.json",
|
||
},
|
||
storePath: "/tmp/openclaw-session-store.json",
|
||
});
|
||
expect(attemptCalls).toHaveLength(1);
|
||
expect(attemptCalls[0]?.sessionFile).toBe(
|
||
"sqlite:default:internal-session:/tmp/openclaw-session-store.json",
|
||
);
|
||
expect(attemptCalls[0]?.sessionEntry).toStrictEqual(visibleEntry);
|
||
expect(state.trajectoryRecorderParamsMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
sessionFile: "sqlite:default:internal-session:/tmp/openclaw-session-store.json",
|
||
}),
|
||
);
|
||
expect(state.persistSessionEntryMock).not.toHaveBeenCalled();
|
||
expect(state.updateSessionStoreAfterAgentRunMock).not.toHaveBeenCalled();
|
||
expect(sessionStore["agent:main:main"]).toBe(visibleEntry);
|
||
expect(state.registerAgentRunContextMock).toHaveBeenCalledWith(
|
||
expect.any(String),
|
||
expect.objectContaining({
|
||
sessionKey: "agent:main:main",
|
||
sessionId: "session-1",
|
||
isControlUiVisible: false,
|
||
}),
|
||
);
|
||
expect(state.removeInternalSessionEffectsSessionMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it.each([
|
||
{
|
||
name: "rejects session-id-resolved model runs for harness-owned sessions",
|
||
sessionKey: "agent:main:harness:codex:supervision:native-thread",
|
||
},
|
||
{
|
||
name: "rejects one-shot model runs for locked harness sessions with ordinary keys",
|
||
sessionKey: "agent:main:plugin-owned",
|
||
},
|
||
])("$name", async ({ sessionKey }) => {
|
||
state.resolvedSessionKeyMock = sessionKey;
|
||
state.sessionEntryMock = createCommandSessionEntry({
|
||
agentHarnessId: "codex",
|
||
modelSelectionLocked: true,
|
||
});
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "probe",
|
||
sessionId: "session-1",
|
||
modelRun: true,
|
||
promptMode: "none",
|
||
sessionEffects: "internal",
|
||
}),
|
||
).rejects.toThrow("Agent harness-owned sessions cannot be used for one-shot model runs.");
|
||
expect(state.runAgentAttemptMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("continues a grandfathered unlocked harness-prefixed session as an ordinary run", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.resolvedSessionKeyMock = "agent:main:harness:notes";
|
||
state.sessionEntryMock = {
|
||
agentHarnessId: "codex",
|
||
modelSelectionLocked: false,
|
||
sessionId: "session-1",
|
||
sessionFile: "/tmp/legacy-session.jsonl",
|
||
updatedAt: 1,
|
||
};
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(state.runAgentAttemptMock).toHaveBeenCalledTimes(1);
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock), {
|
||
agentHarnessRuntimeOverride: undefined,
|
||
sessionEntry: expect.objectContaining({
|
||
agentHarnessId: "codex",
|
||
modelSelectionLocked: false,
|
||
sessionId: "session-1",
|
||
}),
|
||
});
|
||
});
|
||
|
||
it("rejects a locked harness row with the wrong owner before transcript or model work", async () => {
|
||
state.resolvedSessionKeyMock = "agent:main:harness:codex:supervision:native-thread";
|
||
state.sessionEntryMock = {
|
||
agentHarnessId: "other",
|
||
modelSelectionLocked: true,
|
||
sessionId: "session-1",
|
||
sessionFile: "/tmp/native-session.jsonl",
|
||
updatedAt: 1,
|
||
};
|
||
|
||
await expect(agentCommand({ message: "continue", sessionId: "session-1" })).rejects.toThrow(
|
||
"Session key namespace is reserved for agent harness-owned sessions.",
|
||
);
|
||
expect(state.prepareInternalSessionEffectsSessionMock).not.toHaveBeenCalled();
|
||
expect(state.runAgentAttemptMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("removes the one-shot internal model-run SQLite session after success", async () => {
|
||
setupSingleAttemptFallback();
|
||
const result = makeSuccessResult("openai", "gpt-5.4");
|
||
state.runAgentAttemptMock.mockResolvedValue({
|
||
...result,
|
||
meta: {
|
||
...result.meta,
|
||
executionTrace: {
|
||
runner: "embedded",
|
||
fallbackUsed: false,
|
||
winnerProvider: "openai",
|
||
winnerModel: "gpt-5.4",
|
||
},
|
||
finalAssistantVisibleText: "ok",
|
||
agentMeta: {
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
sessionId: "rotated-model-run-session",
|
||
sessionFile: "sqlite:default:rotated-model-run-session:/tmp/openclaw-session-store.json",
|
||
},
|
||
},
|
||
});
|
||
|
||
await runInternalModelCommand("model-run-success");
|
||
|
||
expect(state.createTrajectoryRuntimeRecorderMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
sessionFile: "sqlite:default:internal-session:/tmp/openclaw-session-store.json",
|
||
}),
|
||
);
|
||
expect(state.persistCliTurnTranscriptMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
sessionId: "rotated-model-run-session",
|
||
sessionKey: "agent:default:internal-session-effects:run",
|
||
storePath: "/tmp/openclaw-session-store.json",
|
||
}),
|
||
);
|
||
expect(state.removeInternalSessionEffectsSessionMock).toHaveBeenCalledWith({
|
||
agentId: "default",
|
||
sessionId: "rotated-model-run-session",
|
||
sessionKey: "agent:default:internal-session-effects:run",
|
||
storePath: "/tmp/openclaw-session-store.json",
|
||
sessionFile: "sqlite:default:internal-session:/tmp/openclaw-session-store.json",
|
||
sessionEntry: { sessionId: "internal-session", updatedAt: 1 },
|
||
});
|
||
const deliveryOrder = state.deliverAgentCommandResultMock.mock.invocationCallOrder[0] ?? 0;
|
||
const cleanupOrder =
|
||
state.removeInternalSessionEffectsSessionMock.mock.invocationCallOrder[0] ?? 0;
|
||
expect(deliveryOrder).toBeLessThan(cleanupOrder);
|
||
});
|
||
|
||
it("removes the one-shot internal model-run SQLite session after provider failure", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockRejectedValueOnce(new Error("probe failed"));
|
||
|
||
await expect(runInternalModelCommand("model-run-failure")).rejects.toThrow("probe failed");
|
||
|
||
expect(state.removeInternalSessionEffectsSessionMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({
|
||
sessionId: "internal-session",
|
||
sessionKey: "agent:default:internal-session-effects:run",
|
||
storePath: "/tmp/openclaw-session-store.json",
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("cleans the deterministic model-run session when preparation fails", async () => {
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
state.loadSessionEntryMock.mockReturnValue({ sessionId: "session-1", updatedAt: 1 });
|
||
state.prepareInternalSessionEffectsSessionMock.mockRejectedValueOnce(
|
||
new Error("session preparation failed"),
|
||
);
|
||
|
||
await expect(runInternalModelCommand("model-run-prepare-failure")).rejects.toThrow(
|
||
"session preparation failed",
|
||
);
|
||
|
||
expect(state.removeInternalSessionEffectsSessionMock).toHaveBeenCalledWith({
|
||
agentId: "default",
|
||
sessionId: "internal-model-run-prepare-failure",
|
||
sessionKey: "agent:default:internal-session-effects:model-run-prepare-failure",
|
||
storePath: "/tmp/openclaw-session-store.json",
|
||
});
|
||
});
|
||
|
||
it("does not replace a completed model-run result with a SQLite cleanup failure", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
state.loadSessionEntryMock.mockReturnValue({ sessionId: "session-1", updatedAt: 1 });
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
state.removeInternalSessionEffectsSessionMock.mockRejectedValue(
|
||
new Error("database is locked"),
|
||
);
|
||
|
||
await expect(runInternalModelCommand("model-run-cleanup-failure")).resolves.toBeUndefined();
|
||
|
||
expect(state.removeInternalSessionEffectsSessionMock).toHaveBeenCalled();
|
||
});
|
||
|
||
it("does not duplicate finishing lifecycle when an attempt already emitted finishing", async () => {
|
||
setupModelSwitchRetry({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: unknown) => {
|
||
state.emitAgentEventMock({
|
||
runId: "run-live-switch",
|
||
stream: "lifecycle",
|
||
data: { phase: "finishing" },
|
||
});
|
||
(attemptParams as { onAgentEvent?: (evt: unknown) => void }).onAgentEvent?.({
|
||
stream: "lifecycle",
|
||
data: { phase: "finishing" },
|
||
});
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const lifecycleFinishingCalls = state.emitAgentEventMock.mock.calls.filter(
|
||
(call: unknown[]) => {
|
||
const arg = call[0] as { stream?: string; data?: { phase?: string } };
|
||
return arg?.stream === "lifecycle" && arg?.data?.phase === "finishing";
|
||
},
|
||
);
|
||
expect(lifecycleFinishingCalls).toHaveLength(1);
|
||
});
|
||
|
||
it.each([
|
||
{
|
||
name: "validates explicit thinking against configured model compat without an allowlist",
|
||
allowlisted: false,
|
||
},
|
||
{
|
||
name: "validates explicit thinking against allowlisted configured model compat when manifest catalog is empty",
|
||
allowlisted: true,
|
||
},
|
||
])("$name", async ({ allowlisted }) => {
|
||
state.runtimeConfigMock = createConfiguredModelCompatRuntimeConfig(allowlisted);
|
||
if (allowlisted) {
|
||
state.loadManifestModelCatalogMock.mockReturnValue([]);
|
||
}
|
||
setupSuccessfulAttempt("gmn", "gpt-5.4");
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
thinking: "xhigh",
|
||
});
|
||
|
||
if (allowlisted) {
|
||
expect(state.loadManifestModelCatalogMock).toHaveBeenCalledTimes(1);
|
||
}
|
||
const thinkingArgs = requireRecord(
|
||
mockCallArg(state.isThinkingLevelSupportedMock),
|
||
"thinking args",
|
||
);
|
||
expect(thinkingArgs.provider).toBe("gmn");
|
||
expect(thinkingArgs.model).toBe("gpt-5.4");
|
||
expect(thinkingArgs.level).toBe("xhigh");
|
||
const catalog = requireArray(thinkingArgs.catalog, "thinking catalog");
|
||
expectRecordFields(catalog[0], {
|
||
provider: "gmn",
|
||
id: "gpt-5.4",
|
||
compat: { supportedReasoningEfforts: ["low", "medium", "high", "xhigh"] },
|
||
});
|
||
});
|
||
|
||
it("resolves explicit model aliases before thinking validation", async () => {
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
model: { primary: "openai/gpt-5.4" },
|
||
models: {
|
||
"openai/*": {},
|
||
"codex/gpt-5.5": {
|
||
alias: "code",
|
||
},
|
||
},
|
||
},
|
||
},
|
||
models: {
|
||
providers: {
|
||
codex: {
|
||
models: [
|
||
{
|
||
id: "gpt-5.5",
|
||
name: "GPT 5.5 Codex",
|
||
reasoning: true,
|
||
compat: { supportedReasoningEfforts: ["low", "medium", "high", "xhigh"] },
|
||
},
|
||
],
|
||
},
|
||
},
|
||
},
|
||
};
|
||
state.loadManifestModelCatalogMock.mockReturnValue([]);
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("codex", "gpt-5.5"));
|
||
|
||
await agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
model: "code",
|
||
thinking: "xhigh",
|
||
allowModelOverride: true,
|
||
});
|
||
|
||
const fallbackParams = mockCallArg(state.runWithModelFallbackMock) as FallbackRunnerParams;
|
||
expect(fallbackParams.provider).toBe("codex");
|
||
expect(fallbackParams.model).toBe("gpt-5.5");
|
||
const thinkingArgs = requireRecord(
|
||
mockCallArg(state.isThinkingLevelSupportedMock),
|
||
"thinking args",
|
||
);
|
||
expect(thinkingArgs.provider).toBe("codex");
|
||
expect(thinkingArgs.model).toBe("gpt-5.5");
|
||
expect(thinkingArgs.level).toBe("xhigh");
|
||
});
|
||
|
||
it("records fallback steps to the session trajectory runtime", async () => {
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
await params.onFallbackStep?.({
|
||
fallbackStepType: "fallback_step",
|
||
fallbackStepFromModel: "ollama/llama3",
|
||
fallbackStepToModel: "openai/gpt-5.4",
|
||
fallbackStepFromFailureReason: "overloaded",
|
||
fallbackStepChainPosition: 1,
|
||
fallbackStepFinalOutcome: "next_fallback",
|
||
});
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(state.trajectoryRecordEventMock).toHaveBeenCalledTimes(1);
|
||
expect(mockCallArg(state.trajectoryRecordEventMock, 0, 0)).toBe("model.fallback_step");
|
||
expectRecordFields(mockCallArg(state.trajectoryRecordEventMock, 0, 1), {
|
||
fallbackStepType: "fallback_step",
|
||
fallbackStepFromModel: "ollama/llama3",
|
||
fallbackStepToModel: "openai/gpt-5.4",
|
||
fallbackStepFromFailureReason: "overloaded",
|
||
fallbackStepChainPosition: 1,
|
||
fallbackStepFinalOutcome: "next_fallback",
|
||
});
|
||
expect(state.trajectoryFlushMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it("suppresses duplicate user persistence only after the current turn has flushed", async () => {
|
||
type AttemptCall = {
|
||
onUserMessagePersisted?: () => void;
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
};
|
||
const attemptCalls: AttemptCall[] = [];
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const first = await params.run(params.provider, params.model);
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [first],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: AttemptCall) => {
|
||
const firstAttempt = attemptCalls.length === 0;
|
||
attemptCalls.push(attemptParams);
|
||
if (firstAttempt) {
|
||
if (!attemptParams.onUserMessagePersisted) {
|
||
throw new Error("expected retry persistence callback on first attempt");
|
||
}
|
||
attemptParams.onUserMessagePersisted();
|
||
} else {
|
||
attemptParams.onUserMessagePersisted?.();
|
||
}
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(attemptCalls).toHaveLength(2);
|
||
expect(attemptCalls[0]?.suppressPromptPersistenceOnRetry).not.toBe(true);
|
||
expect(attemptCalls[1]?.suppressPromptPersistenceOnRetry).toBe(true);
|
||
});
|
||
|
||
it("keeps a hook-blocked user turn suppressed across model fallback", async () => {
|
||
type AttemptCall = {
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
userTurnTranscriptRecorder?: {
|
||
markBlocked: () => void;
|
||
};
|
||
};
|
||
const attemptCalls: AttemptCall[] = [];
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const first = await params.run(params.provider, params.model);
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [first],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: AttemptCall) => {
|
||
attemptCalls.push(attemptParams);
|
||
if (attemptCalls.length === 1) {
|
||
attemptParams.userTurnTranscriptRecorder?.markBlocked();
|
||
}
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(attemptCalls).toHaveLength(2);
|
||
expect(attemptCalls[1]?.userTurnTranscriptRecorder).toBe(
|
||
attemptCalls[0]?.userTurnTranscriptRecorder,
|
||
);
|
||
expect(attemptCalls[0]?.suppressPromptPersistenceOnRetry).toBe(false);
|
||
expect(attemptCalls[1]?.suppressPromptPersistenceOnRetry).toBe(true);
|
||
});
|
||
|
||
it("suppresses prompt persistence for internal handoffs on every fallback attempt", async () => {
|
||
type AttemptCall = {
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
};
|
||
const attemptCalls: AttemptCall[] = [];
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const first = await params.run(params.provider, params.model);
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [first],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(async (attemptParams: AttemptCall) => {
|
||
attemptCalls.push(attemptParams);
|
||
const result = makeSuccessResult("openai", "gpt-5.4") as ReturnType<
|
||
typeof makeSuccessResult
|
||
> & {
|
||
meta: Record<string, unknown> & { executionTrace?: Record<string, unknown> };
|
||
};
|
||
result.meta.executionTrace = {
|
||
runner: "cli",
|
||
fallbackUsed: false,
|
||
winnerProvider: "openai",
|
||
winnerModel: "gpt-5.4",
|
||
};
|
||
return result;
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "internal handoff",
|
||
to: "+1234567890",
|
||
suppressPromptPersistence: true,
|
||
});
|
||
|
||
expect(attemptCalls).toHaveLength(2);
|
||
expect(attemptCalls[0]?.suppressPromptPersistenceOnRetry).toBe(true);
|
||
expect(attemptCalls[1]?.suppressPromptPersistenceOnRetry).toBe(true);
|
||
expectRecordFields(mockCallArg(state.persistCliTurnTranscriptMock), {
|
||
skipUserTurn: true,
|
||
});
|
||
});
|
||
|
||
it("preserves an explicit empty transcript message as user-turn omission", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await agentCommand({
|
||
message: "synthetic announce prompt",
|
||
transcriptMessage: "",
|
||
to: "+1234567890",
|
||
});
|
||
|
||
const attempt = mockCallArg(state.runAgentAttemptMock) as {
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
userTurnTranscriptRecorder?: { message?: unknown };
|
||
};
|
||
expect(attempt.suppressPromptPersistenceOnRetry).toBe(true);
|
||
expect(attempt.userTurnTranscriptRecorder?.message).toBeUndefined();
|
||
});
|
||
|
||
it("uses a tracker-only recorder for text plus image turns", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await agentCommand({
|
||
message: "inspect this image",
|
||
transcriptMessage: "canonical image caption",
|
||
images: [{ type: "image", data: "aGVsbG8=", mimeType: "image/png" }],
|
||
to: "+1234567890",
|
||
});
|
||
|
||
const attempt = mockCallArg(state.runAgentAttemptMock) as {
|
||
transcriptBody?: string;
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
userTurnTranscriptRecorder?: { message?: unknown };
|
||
};
|
||
expect(attempt.transcriptBody).toBe("canonical image caption");
|
||
expect(attempt.suppressPromptPersistenceOnRetry).toBe(false);
|
||
expect(attempt.userTurnTranscriptRecorder?.message).toMatchObject({
|
||
role: "user",
|
||
content: "canonical image caption",
|
||
});
|
||
});
|
||
|
||
it("persists structured transcript media without a caption", async () => {
|
||
setupSingleAttemptFallback();
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await agentCommand({
|
||
message: "[media attached: media://inbound/image-1]",
|
||
transcriptMessage: "",
|
||
transcriptMedia: [{ path: "/media/inbound/image-1.png", contentType: "image/png" }],
|
||
images: [{ type: "image", data: "aGVsbG8=", mimeType: "image/png" }],
|
||
to: "+1234567890",
|
||
});
|
||
|
||
const attempt = mockCallArg(state.runAgentAttemptMock) as {
|
||
suppressPromptPersistenceOnRetry?: boolean;
|
||
userTurnTranscriptRecorder?: { message?: unknown };
|
||
};
|
||
expect(attempt.suppressPromptPersistenceOnRetry).toBe(false);
|
||
expect(attempt.userTurnTranscriptRecorder?.message).toMatchObject({
|
||
role: "user",
|
||
content: "",
|
||
MediaPath: "/media/inbound/image-1.png",
|
||
MediaPaths: ["/media/inbound/image-1.png"],
|
||
MediaType: "image/png",
|
||
MediaTypes: ["image/png"],
|
||
});
|
||
});
|
||
|
||
it("propagates non-switch errors without retrying and emits lifecycle error", async () => {
|
||
state.runWithModelFallbackMock.mockRejectedValueOnce(new Error("provider down"));
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
}),
|
||
).rejects.toThrow("provider down");
|
||
|
||
expect(state.runWithModelFallbackMock).toHaveBeenCalledTimes(1);
|
||
|
||
const lifecycleErrorCalls = state.emitAgentEventMock.mock.calls.filter((call: unknown[]) => {
|
||
const arg = call[0] as { stream?: string; data?: { phase?: string } };
|
||
return arg?.stream === "lifecycle" && arg?.data?.phase === "error";
|
||
});
|
||
expect(lifecycleErrorCalls.length).toBeGreaterThanOrEqual(1);
|
||
});
|
||
|
||
it("marks lifecycle errors aborted when cancellation reaches post-turn handling", async () => {
|
||
const abortController = new AbortController();
|
||
state.runWithModelFallbackMock.mockImplementationOnce(async () => {
|
||
abortController.abort();
|
||
throw new Error("request aborted");
|
||
});
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
abortSignal: abortController.signal,
|
||
}),
|
||
).rejects.toThrow("request aborted");
|
||
|
||
expect(
|
||
state.emitAgentEventMock.mock.calls.some(([event]) => {
|
||
const candidate = event as {
|
||
stream?: string;
|
||
data?: { phase?: string; aborted?: boolean };
|
||
};
|
||
return (
|
||
candidate.stream === "lifecycle" &&
|
||
candidate.data?.phase === "error" &&
|
||
candidate.data.aborted === true
|
||
);
|
||
}),
|
||
).toBe(true);
|
||
});
|
||
|
||
it("marks direct active-run cancellation aborted without a caller signal", async () => {
|
||
state.runWithModelFallbackMock.mockRejectedValueOnce(createAgentRunDirectAbortError());
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "hello",
|
||
to: "+1234567890",
|
||
}),
|
||
).rejects.toThrow("agent run aborted");
|
||
|
||
expect(
|
||
state.emitAgentEventMock.mock.calls.some(([event]) => {
|
||
const candidate = event as {
|
||
stream?: string;
|
||
data?: { phase?: string; aborted?: boolean; stopReason?: string };
|
||
};
|
||
return (
|
||
candidate.stream === "lifecycle" &&
|
||
candidate.data?.phase === "error" &&
|
||
candidate.data.aborted === true &&
|
||
candidate.data.stopReason === "aborted"
|
||
);
|
||
}),
|
||
).toBe(true);
|
||
});
|
||
|
||
it("propagates authProfileId from the switch error to the retried session entry", async () => {
|
||
let capturedAuthProfileProvider: string | undefined;
|
||
setupModelSwitchRetry({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
authProfileId: "profile-openai-prod",
|
||
authProfileIdSource: "user",
|
||
});
|
||
|
||
state.runAgentAttemptMock.mockImplementation(async (...args: unknown[]) => {
|
||
const attemptParams = args[0] as { authProfileProvider?: string } | undefined;
|
||
capturedAuthProfileProvider = attemptParams?.authProfileProvider;
|
||
return makeSuccessResult("openai", "gpt-5.4");
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(capturedAuthProfileProvider).toBe("openai");
|
||
expect(state.runWithModelFallbackMock).toHaveBeenCalledTimes(2);
|
||
});
|
||
|
||
it("does not persist a user live switch as an auto fallback probe result", async () => {
|
||
const sessionEntry: SessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
providerOverride: "openai",
|
||
modelOverride: "claude",
|
||
modelOverrideSource: "auto",
|
||
modelOverrideFallbackOriginProvider: "anthropic",
|
||
modelOverrideFallbackOriginModel: "claude",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
};
|
||
state.sessionEntryMock = sessionEntry;
|
||
const sessionStore: Record<string, SessionEntry> = { "agent:main:main": sessionEntry };
|
||
state.sessionStoreMock = sessionStore;
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
setupModelSwitchRetry({
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
authProfileId: "openai:primary",
|
||
authProfileIdSource: "user",
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const autoPinnedSwitchWrites = state.persistSessionEntryMock.mock.calls.filter((call) => {
|
||
const entry = (call[0] as { entry?: Record<string, unknown> } | undefined)?.entry;
|
||
return (
|
||
entry?.providerOverride === "openai" &&
|
||
entry?.modelOverride === "gpt-5.4" &&
|
||
entry?.modelOverrideSource === "auto" &&
|
||
entry?.modelOverrideFallbackOriginProvider === "anthropic"
|
||
);
|
||
});
|
||
expect(autoPinnedSwitchWrites).toHaveLength(0);
|
||
expectRecordFields(mockCallArg(state.updateSessionStoreAfterAgentRunMock), {
|
||
fallbackProvider: "openai",
|
||
fallbackModel: "gpt-5.4",
|
||
});
|
||
});
|
||
|
||
it("does not overwrite a concurrent user model switch after a primary probe", async () => {
|
||
const sessionEntry: SessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
providerOverride: "openai",
|
||
modelOverride: "claude",
|
||
modelOverrideSource: "auto",
|
||
modelOverrideFallbackOriginProvider: "anthropic",
|
||
modelOverrideFallbackOriginModel: "claude",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
};
|
||
state.sessionEntryMock = sessionEntry;
|
||
const sessionStore: Record<string, SessionEntry> = { "agent:main:main": sessionEntry };
|
||
state.sessionStoreMock = sessionStore;
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
state.runWithModelFallbackMock.mockImplementationOnce(async (params: FallbackRunnerParams) => {
|
||
const result = await params.run(params.provider, params.model);
|
||
sessionStore["agent:main:main"] = {
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
providerOverride: "google",
|
||
modelOverride: "gemini-3-pro",
|
||
modelOverrideSource: "user",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
};
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(sessionStore["agent:main:main"], {
|
||
providerOverride: "google",
|
||
modelOverride: "gemini-3-pro",
|
||
modelOverrideSource: "user",
|
||
});
|
||
});
|
||
|
||
it("does not persist an automatic probe result after the session becomes locked", async () => {
|
||
const sessionEntry: SessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
providerOverride: "openai",
|
||
modelOverride: "claude",
|
||
modelOverrideSource: "auto",
|
||
modelOverrideFallbackOriginProvider: "anthropic",
|
||
modelOverrideFallbackOriginModel: "claude",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
};
|
||
state.sessionEntryMock = sessionEntry;
|
||
state.sessionStoreMock = { "agent:main:main": sessionEntry };
|
||
state.storePathMock = "/tmp/openclaw-session-store.json";
|
||
state.resolveAutoFallbackPrimaryProbeMock.mockReturnValue({
|
||
provider: "anthropic",
|
||
model: "claude",
|
||
fallbackProvider: "openai",
|
||
fallbackModel: "claude",
|
||
});
|
||
state.runWithModelFallbackMock.mockImplementationOnce(async (params: FallbackRunnerParams) => {
|
||
state.persistSessionEntryMock.mockClear();
|
||
const result = await params.run("openai", "claude");
|
||
const currentEntry = expectDefined(
|
||
(state.sessionStoreMock as Record<string, SessionEntry>)["agent:main:main"],
|
||
'(state.sessionStoreMock as Record<string, SessionEntry>)[ "agent:main... test invariant',
|
||
);
|
||
currentEntry.modelSelectionLocked = true;
|
||
state.isModelSelectionLockedMock.mockReturnValue(true);
|
||
return {
|
||
result,
|
||
provider: "openai",
|
||
model: "claude",
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("openai", "claude"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const autoProbeWrites = state.persistSessionEntryMock.mock.calls.filter((call) => {
|
||
const entry = (call[0] as { entry?: SessionEntry } | undefined)?.entry;
|
||
return entry?.modelOverrideSource === "auto" && entry?.modelOverride === "claude";
|
||
});
|
||
expect(autoProbeWrites).toHaveLength(0);
|
||
});
|
||
|
||
it("keeps aliased session auth profiles for codex-cli runs", async () => {
|
||
let capturedAuthProfileProvider: string | undefined;
|
||
const sessionEntry = {
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
providerOverride: "codex-cli",
|
||
modelOverride: "gpt-5.4",
|
||
authProfileOverride: "openai:work",
|
||
authProfileOverrideSource: "user",
|
||
skillsSnapshot: { prompt: "", skills: [], version: 0 },
|
||
} satisfies SessionEntry;
|
||
state.sessionEntryMock = sessionEntry;
|
||
state.runtimeConfigMock = {
|
||
agents: {
|
||
defaults: {
|
||
models: {
|
||
"codex-cli/gpt-5.4": {},
|
||
},
|
||
},
|
||
},
|
||
};
|
||
state.authProfileStoreMock = {
|
||
profiles: {
|
||
"openai:work": {
|
||
type: "api_key",
|
||
provider: "openai",
|
||
key: "sk-test",
|
||
},
|
||
},
|
||
};
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockImplementation(async (...args: unknown[]) => {
|
||
const attemptParams = args[0] as { authProfileProvider?: string } | undefined;
|
||
capturedAuthProfileProvider = attemptParams?.authProfileProvider;
|
||
return makeSuccessResult("codex-cli", "gpt-5.4");
|
||
});
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(capturedAuthProfileProvider).toBe("codex-cli");
|
||
expect(state.clearSessionAuthProfileOverrideMock).not.toHaveBeenCalled();
|
||
});
|
||
|
||
it("hydrates stripped persisted skill snapshots before running the CLI path", async () => {
|
||
const persistedSnapshot = {
|
||
prompt: "persisted prompt",
|
||
skills: [{ name: "cli-skill" }],
|
||
skillFilter: ["cli-skill"],
|
||
version: 0,
|
||
};
|
||
const rebuiltSkills = [
|
||
{
|
||
name: "cli-skill",
|
||
description: "CLI skill",
|
||
filePath: "/tmp/workspace/skills/cli-skill/SKILL.md",
|
||
baseDir: "/tmp/workspace/skills/cli-skill",
|
||
source: "# CLI skill",
|
||
},
|
||
];
|
||
state.sessionEntryMock = {
|
||
sessionId: "session-1",
|
||
updatedAt: Date.now(),
|
||
skillsSnapshot: persistedSnapshot,
|
||
};
|
||
state.buildWorkspaceSkillSnapshotMock.mockReturnValue({
|
||
prompt: "rebuilt prompt",
|
||
skills: [{ name: "different-skill" }],
|
||
resolvedSkills: rebuiltSkills,
|
||
version: 99,
|
||
});
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const result = await params.run(params.provider, params.model);
|
||
return {
|
||
result,
|
||
provider: params.provider,
|
||
model: params.model,
|
||
attempts: [],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock.mockResolvedValue(makeSuccessResult("anthropic", "claude"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
const attemptParams = mockCallArg(state.runAgentAttemptMock) as {
|
||
skillsSnapshot?: Record<string, unknown>;
|
||
};
|
||
expectRecordFields(attemptParams?.skillsSnapshot, {
|
||
prompt: "persisted prompt",
|
||
skills: [{ name: "cli-skill" }],
|
||
skillFilter: ["cli-skill"],
|
||
version: 0,
|
||
resolvedSkills: rebuiltSkills,
|
||
});
|
||
expect(state.buildWorkspaceSkillSnapshotMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it("classifies empty embedded run results before model fallback accepts them", async () => {
|
||
let observedClassification: unknown;
|
||
state.runWithModelFallbackMock.mockImplementation(async (params: FallbackRunnerParams) => {
|
||
const primaryResult = await params.run(params.provider, params.model);
|
||
observedClassification = await params.classifyResult?.({
|
||
provider: params.provider,
|
||
model: params.model,
|
||
result: primaryResult,
|
||
attempt: 1,
|
||
total: 2,
|
||
});
|
||
const fallbackResult = await params.run("openai", "gpt-5.4");
|
||
return {
|
||
result: fallbackResult,
|
||
provider: "openai",
|
||
model: "gpt-5.4",
|
||
attempts: [
|
||
{
|
||
provider: params.provider,
|
||
model: params.model,
|
||
error: "empty result",
|
||
reason: "format",
|
||
code: "empty_result",
|
||
},
|
||
],
|
||
};
|
||
});
|
||
state.runAgentAttemptMock
|
||
.mockResolvedValueOnce(makeEmptyResult("anthropic", "claude"))
|
||
.mockResolvedValueOnce(makeSuccessResult("openai", "gpt-5.4"));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectRecordFields(observedClassification, {
|
||
reason: "format",
|
||
code: "empty_result",
|
||
});
|
||
expect(state.runAgentAttemptMock).toHaveBeenCalledTimes(2);
|
||
expectRecordFields(mockCallArg(state.runAgentAttemptMock, 1), {
|
||
providerOverride: "openai",
|
||
modelOverride: "gpt-5.4",
|
||
isFallbackRetry: true,
|
||
});
|
||
const deliveryParams = requireRecord(
|
||
mockCallArg(state.deliverAgentCommandResultMock),
|
||
"delivery params",
|
||
);
|
||
const result = requireRecord(deliveryParams.result, "delivery result");
|
||
const meta = requireRecord(result.meta, "delivery result meta");
|
||
const agentMeta = requireRecord(meta.agentMeta, "delivery agent meta");
|
||
const fallbackAttempts = requireArray(agentMeta.fallbackAttempts, "fallback attempts");
|
||
expectRecordFields(fallbackAttempts[0], {
|
||
provider: "anthropic",
|
||
model: "claude",
|
||
reason: "format",
|
||
});
|
||
});
|
||
|
||
it("emits a failure lifecycle after delivering a preserved exhausted result", async () => {
|
||
const exhaustedResult = {
|
||
payloads: [{ text: "Terminal tool summary", isError: true }],
|
||
meta: {
|
||
durationMs: 100,
|
||
aborted: false,
|
||
stopReason: "end_turn",
|
||
error: {
|
||
kind: "incomplete_turn",
|
||
message: "All fallback candidates ended incomplete",
|
||
fallbackSafe: true,
|
||
terminalPresentation: true,
|
||
},
|
||
agentMeta: { provider: "anthropic", model: "claude" },
|
||
},
|
||
};
|
||
state.runAgentAttemptMock.mockImplementationOnce(async (attemptParams: unknown) => {
|
||
const params = attemptParams as {
|
||
deferTerminalLifecycle?: boolean;
|
||
onAgentEvent?: (event: { stream: string; data: Record<string, unknown> }) => void;
|
||
};
|
||
expect(params.deferTerminalLifecycle).toBe(true);
|
||
params.onAgentEvent?.({
|
||
stream: "lifecycle",
|
||
data: {
|
||
phase: "finishing",
|
||
error: "All fallback candidates ended incomplete",
|
||
},
|
||
});
|
||
return exhaustedResult;
|
||
});
|
||
state.runWithModelFallbackMock.mockImplementationOnce(async (params: FallbackRunnerParams) => ({
|
||
outcome: "exhausted",
|
||
result: await params.run("anthropic", "claude"),
|
||
provider: "anthropic",
|
||
model: "claude",
|
||
attempts: [
|
||
{
|
||
provider: "anthropic",
|
||
model: "claude",
|
||
error: "All fallback candidates ended incomplete",
|
||
reason: "format",
|
||
},
|
||
],
|
||
}));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(state.deliverAgentCommandResultMock).toHaveBeenCalledTimes(1);
|
||
const lifecycleEvents = state.emitAgentEventMock.mock.calls
|
||
.map((call) => call[0] as { stream?: string; data?: Record<string, unknown> })
|
||
.filter((event) => event.stream === "lifecycle");
|
||
expect(lifecycleEvents.some((event) => event.data?.phase === "finishing")).toBe(false);
|
||
expect(lifecycleEvents.some((event) => event.data?.phase === "end")).toBe(false);
|
||
expect(lifecycleEvents).toEqual(
|
||
expect.arrayContaining([
|
||
expect.objectContaining({
|
||
data: expect.objectContaining({
|
||
phase: "error",
|
||
error: "All fallback candidates ended incomplete",
|
||
fallbackExhaustedFailure: true,
|
||
}),
|
||
}),
|
||
]),
|
||
);
|
||
});
|
||
|
||
it("emits a failure lifecycle for completed non-fallbackable error results", async () => {
|
||
const terminalErrorResult = {
|
||
payloads: [{ text: "Command may have changed state", isError: true }],
|
||
meta: {
|
||
durationMs: 100,
|
||
aborted: false,
|
||
stopReason: "end_turn",
|
||
replayInvalid: true,
|
||
error: {
|
||
kind: "incomplete_turn",
|
||
message: "raw provider detail should stay private",
|
||
fallbackSafe: false,
|
||
},
|
||
agentMeta: { provider: "anthropic", model: "claude" },
|
||
},
|
||
};
|
||
state.runAgentAttemptMock.mockImplementationOnce(async (attemptParams: unknown) => {
|
||
const params = attemptParams as {
|
||
onAgentEvent?: (event: { stream: string; data: Record<string, unknown> }) => void;
|
||
};
|
||
params.onAgentEvent?.({
|
||
stream: "lifecycle",
|
||
data: {
|
||
phase: "finishing",
|
||
error: "Command may have changed state",
|
||
replayInvalid: true,
|
||
},
|
||
});
|
||
return terminalErrorResult;
|
||
});
|
||
state.runWithModelFallbackMock.mockImplementationOnce(async (params: FallbackRunnerParams) => ({
|
||
outcome: "completed",
|
||
result: await params.run("anthropic", "claude"),
|
||
provider: "anthropic",
|
||
model: "claude",
|
||
attempts: [],
|
||
}));
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expect(state.deliverAgentCommandResultMock).toHaveBeenCalledTimes(1);
|
||
const lifecycleEvents = state.emitAgentEventMock.mock.calls
|
||
.map((call) => call[0] as { stream?: string; data?: Record<string, unknown> })
|
||
.filter((event) => event.stream === "lifecycle");
|
||
expect(lifecycleEvents).toEqual(
|
||
expect.arrayContaining([
|
||
expect.objectContaining({
|
||
data: expect.objectContaining({
|
||
phase: "error",
|
||
error: "Command may have changed state",
|
||
replayInvalid: true,
|
||
}),
|
||
}),
|
||
]),
|
||
);
|
||
expect(
|
||
lifecycleEvents.some(
|
||
(event) => event.data?.phase === "end" || event.data?.fallbackExhaustedFailure === true,
|
||
),
|
||
).toBe(false);
|
||
expect(JSON.stringify(lifecycleEvents)).not.toContain("raw provider detail");
|
||
});
|
||
|
||
it.each([
|
||
{
|
||
name: "updates hasSessionModelOverride for fallback resolution after switch",
|
||
switchOptions: { provider: "openai", model: "gpt-5.4" },
|
||
expectedOverride: true,
|
||
},
|
||
{
|
||
name: "does not flip hasSessionModelOverride on auth-only switch with same model",
|
||
switchOptions: {
|
||
provider: "anthropic",
|
||
model: "claude",
|
||
authProfileId: "profile-99",
|
||
authProfileIdSource: "user" as const,
|
||
},
|
||
expectedOverride: false,
|
||
},
|
||
{
|
||
name: "flips hasSessionModelOverride on provider-only switch with same model",
|
||
switchOptions: { provider: "openai", model: "claude" },
|
||
expectedOverride: true,
|
||
},
|
||
])("$name", async ({ switchOptions, expectedOverride }) => {
|
||
setupModelSwitchRetry(switchOptions);
|
||
state.runAgentAttemptMock.mockResolvedValue(
|
||
makeSuccessResult(switchOptions.provider, switchOptions.model),
|
||
);
|
||
state.resolveEffectiveModelFallbacksMock.mockClear();
|
||
|
||
await runBasicAgentCommand();
|
||
|
||
expectFallbackOverrideCalls(false, expectedOverride);
|
||
});
|
||
|
||
it("sends internal completion wakes to ACP sessions as plain prompt text", async () => {
|
||
setupAcpSession();
|
||
|
||
await agentCommand({
|
||
message: [
|
||
INTERNAL_RUNTIME_CONTEXT_BEGIN,
|
||
"OpenClaw runtime context (internal):",
|
||
"hidden task completion event",
|
||
INTERNAL_RUNTIME_CONTEXT_END,
|
||
].join("\n"),
|
||
sessionKey: "agent:main:main",
|
||
internalEvents: [
|
||
{
|
||
type: "task_completion",
|
||
source: "subagent",
|
||
childSessionKey: "agent:main:subagent:child",
|
||
childSessionId: "child-session-id",
|
||
announceType: "subagent task",
|
||
taskLabel: "inspect ACP delivery",
|
||
status: "ok",
|
||
statusLabel: "completed successfully",
|
||
result: "child output",
|
||
replyInstruction: "Summarize the result for the user.",
|
||
},
|
||
],
|
||
});
|
||
|
||
expect(state.acpRunTurnMock).toHaveBeenCalledTimes(1);
|
||
const runTurnParams = mockCallArg(state.acpRunTurnMock) as { text?: string };
|
||
expect(runTurnParams.text).toContain("A background task completed.");
|
||
expect(runTurnParams.text).toContain("inspect ACP delivery");
|
||
expect(runTurnParams.text).toContain("child output");
|
||
expect(runTurnParams.text).not.toContain(INTERNAL_RUNTIME_CONTEXT_BEGIN);
|
||
expect(runTurnParams.text).not.toContain(INTERNAL_RUNTIME_CONTEXT_END);
|
||
|
||
expect(state.persistAcpTurnTranscriptMock).toHaveBeenCalledTimes(1);
|
||
const transcriptParams = mockCallArg(state.persistAcpTurnTranscriptMock) as {
|
||
body?: string;
|
||
transcriptBody?: string;
|
||
};
|
||
expect(transcriptParams.body).toBe(runTurnParams.text);
|
||
expect(transcriptParams.transcriptBody).toContain("A background task completed.");
|
||
expect(transcriptParams.transcriptBody).not.toContain(INTERNAL_RUNTIME_CONTEXT_BEGIN);
|
||
expect(transcriptParams.transcriptBody).not.toContain(INTERNAL_RUNTIME_CONTEXT_END);
|
||
});
|
||
|
||
it("keeps session provenance for internal ACP turns", async () => {
|
||
setupAcpSession();
|
||
|
||
await agentCommand({
|
||
message: "internal ACP turn",
|
||
sessionKey: "agent:main:main",
|
||
sessionEffects: "internal",
|
||
});
|
||
|
||
expect(state.registerAgentRunContextMock).toHaveBeenCalledWith(
|
||
"session-1",
|
||
expect.objectContaining({
|
||
sessionKey: "agent:main:main",
|
||
sessionId: "session-1",
|
||
isControlUiVisible: false,
|
||
}),
|
||
);
|
||
});
|
||
|
||
it("allows manual ACP spawn turns when ACP dispatch is disabled", async () => {
|
||
setupAcpSession();
|
||
state.resolveAcpDispatchPolicyErrorMock.mockReturnValue(
|
||
new Error("ACP dispatch is disabled by policy (`acp.dispatch.enabled=false`)."),
|
||
);
|
||
|
||
await agentCommand({
|
||
message: "bootstrap ACP child",
|
||
sessionKey: "agent:main:main",
|
||
acpTurnSource: "manual_spawn",
|
||
});
|
||
|
||
expect(state.resolveAcpExplicitTurnPolicyErrorMock).toHaveBeenCalledTimes(1);
|
||
expect(state.resolveAcpDispatchPolicyErrorMock).not.toHaveBeenCalled();
|
||
expect(state.acpRunTurnMock).toHaveBeenCalledTimes(1);
|
||
});
|
||
|
||
it("keeps ordinary ACP turns blocked when ACP dispatch is disabled", async () => {
|
||
setupAcpSession();
|
||
state.resolveAcpDispatchPolicyErrorMock.mockReturnValue(
|
||
new Error("ACP dispatch is disabled by policy (`acp.dispatch.enabled=false`)."),
|
||
);
|
||
|
||
await expect(
|
||
agentCommand({
|
||
message: "automatic ACP turn",
|
||
sessionKey: "agent:main:main",
|
||
}),
|
||
).rejects.toThrow("ACP dispatch is disabled");
|
||
|
||
expect(state.resolveAcpExplicitTurnPolicyErrorMock).not.toHaveBeenCalled();
|
||
expect(state.resolveAcpDispatchPolicyErrorMock).toHaveBeenCalledTimes(1);
|
||
expect(state.acpRunTurnMock).not.toHaveBeenCalled();
|
||
expect(state.emitAcpLifecycleErrorMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({ terminalOutcome: "blocked" }),
|
||
);
|
||
});
|
||
|
||
it("preserves ACP cancelled results without a stop reason", async () => {
|
||
setupAcpSession();
|
||
state.acpRunTurnMock.mockImplementationOnce(async (params: unknown) => {
|
||
const onEvent = (params as { onEvent?: (event: unknown) => void }).onEvent;
|
||
onEvent?.({ type: "done", status: "cancelled" });
|
||
});
|
||
|
||
await agentCommand({
|
||
message: "cancelled ACP turn",
|
||
sessionKey: "agent:main:main",
|
||
});
|
||
|
||
expect(state.emitAcpLifecycleEndMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({ resultStatus: "cancelled", stopReason: undefined }),
|
||
);
|
||
expect(state.buildAcpResultMock).toHaveBeenCalledWith(
|
||
expect.objectContaining({ resultStatus: "cancelled", stopReason: undefined }),
|
||
);
|
||
});
|
||
});
|
||
/* oxlint-disable max-lines -- TODO: split this grandfathered oversized file. */
|