Files
openclaw/src/commands/agent.ts
Tyler Yust b8f66c260d Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447)
* Agents: add subagent orchestration controls

* Agents: add subagent orchestration controls (WIP uncommitted changes)

* feat(subagents): add depth-based spawn gating for sub-sub-agents

* feat(subagents): tool policy, registry, and announce chain for nested agents

* feat(subagents): system prompt, docs, changelog for nested sub-agents

* fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback

Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex),
the fallback candidate logic in resolveFallbackCandidates silently appended the
global primary model (opus) as a backstop. On reinjection/steer with a transient
error, the session could fall back to opus which has a smaller context window
and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? []
instead of undefined, preventing the implicit primary backstop.

Bug 2: Active subagents showed 'model n/a' in /subagents list because
resolveModelDisplay only read entry.model/modelProvider (populated after run
completes). Fix: fall back to modelOverride/providerOverride fields which are
populated at spawn time via sessions.patch.

Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could
theoretically escape runEmbeddedPiAgent and be treated as failover candidates
in runWithModelFallback, causing a switch to a model with a smaller context
window. Fix: in runWithModelFallback, detect context overflow errors via
isLikelyContextOverflowError and rethrow them immediately instead of trying the
next model candidate.

* fix(subagents): track spawn depth in session store and fix announce routing for nested agents

* Fix compaction status tracking and dedupe overflow compaction triggers

* fix(subagents): enforce depth block via session store and implement cascade kill

* fix: inject group chat context into system prompt

* fix(subagents): always write model to session store at spawn time

* Preserve spawnDepth when agent handler rewrites session entry

* fix(subagents): suppress announce on steer-restart

* fix(subagents): fallback spawned session model to runtime default

* fix(subagents): enforce spawn depth when caller key resolves by sessionId

* feat(subagents): implement active-first ordering for numeric targets and enhance task display

- Added a test to verify that subagents with numeric targets follow an active-first list ordering.
- Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity.
- Enhanced task display in command responses to prevent truncation of long task descriptions.
- Introduced new utility functions for compacting task text and managing subagent run states.

* fix(subagents): show model for active runs via run record fallback

When the spawned model matches the agent's default model, the session
store's override fields are intentionally cleared (isDefault: true).
The model/modelProvider fields are only populated after the run
completes. This left active subagents showing 'model n/a'.

Fix: store the resolved model on SubagentRunRecord at registration
time, and use it as a fallback in both display paths (subagents tool
and /subagents command) when the session store entry has no model info.

Changes:
- SubagentRunRecord: add optional model field
- registerSubagentRun: accept and persist model param
- sessions-spawn-tool: pass resolvedModel to registerSubagentRun
- subagents-tool: pass run record model as fallback to resolveModelDisplay
- commands-subagents: pass run record model as fallback to resolveModelDisplay

* feat(chat): implement session key resolution and reset on sidebar navigation

- Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar.
- Updated the `renderTab` function to handle session key changes when navigating to the chat tab.
- Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation.

* fix: subagent timeout=0 passthrough and fallback prompt duplication

Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default
- sessions-spawn-tool: default to undefined (not 0) when neither timeout param
  is provided; use != null check so explicit 0 passes through to gateway
- agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles
  0 → MAX_SAFE_TIMEOUT_MS)

Bug 2: model fallback no longer re-injects the original prompt as a duplicate
- agent.ts: track fallback attempt index; on retries use a short continuation
  message instead of the full original prompt since the session file already
  contains it from the first attempt
- Also skip re-sending images on fallback retries (already in session)

* feat(subagents): truncate long task descriptions in subagents command output

- Introduced a new utility function to format task previews, limiting their length to improve readability.
- Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately.
- Adjusted related tests to verify that long task descriptions are now truncated in the output.

* refactor(subagents): update subagent registry path resolution and improve command output formatting

- Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically.
- Enhanced the formatting of command output for active and recent subagents, adding separators for better readability.
- Updated related tests to reflect changes in command output structure.

* fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted

The previous fix (75a791106) correctly handled the case where
runTimeoutSeconds was explicitly set to 0 ("no timeout"). However,
when models omit the parameter entirely (which is common since the
schema marks it as optional), runTimeoutSeconds resolved to undefined.

undefined flowed through the chain as:
  sessions_spawn → timeout: undefined (since undefined != null is false)
  → gateway agent handler → agentCommand opts.timeout: undefined
  → resolveAgentTimeoutMs({ overrideSeconds: undefined })
  → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes)

This caused subagents to be killed at exactly 10 minutes even though
the user's intent (via TOOLS.md) was for subagents to run without a
timeout.

Fix: default runTimeoutSeconds to 0 (no timeout) when neither
runTimeoutSeconds nor timeoutSeconds is provided by the caller.
Subagent spawns are long-running by design and should not inherit the
600s agent-command default timeout.

* fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default)

* fix: thread timeout override through getReplyFromConfig dispatch path

getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override,
always falling back to the config default (600s). Add timeoutOverrideSeconds
to GetReplyOptions and pass it through as overrideSeconds so callers of the
dispatch chain can specify a custom timeout (0 = no timeout).

This complements the existing timeout threading in agentCommand and the
cron isolated-agent runner, which already pass overrideSeconds correctly.

* feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling

- Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution.
- Updated the `resolveFallbackCandidates` function to utilize the new normalization logic.
- Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms.
- Introduced a new test case to ensure that the normalization process works as expected for various input formats.

* feat(tests): add unit tests for steer failure behavior in openclaw-tools

- Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails.
- Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected.
- Enhanced the subagent registry with a new function to clear steer restart suppression.
- Updated related components to support the new test scenarios.

* fix(subagents): replace stop command with kill in slash commands and documentation

- Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs.
- Modified related documentation to reflect the change in command usage.
- Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling.
- Enhanced tests to ensure correct behavior of the updated commands and their interactions.

* feat(tests): add unit tests for readLatestAssistantReply function

- Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios.
- Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text.
- Mocked the gateway call to simulate different message histories for comprehensive testing.

* feat(tests): enhance subagent kill-all cascade tests and announce formatting

- Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents.
- Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content.
- Improved the handling of long findings and stats in the announce formatting logic to ensure concise output.
- Refactored related functions to enhance clarity and maintainability in the subagent registry and tools.

* refactor(subagent): update announce formatting and remove unused constants

- Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests.
- Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic.
- Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs.
- Cleaned up unused imports in the commands-subagents file to enhance code clarity.

* feat(tests): enhance billing error handling in user-facing text

- Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context.
- Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages.
- Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output.
- Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements.

* feat(subagent): enhance workflow guidance and auto-announcement clarity

- Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates.
- Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow.
- Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts.

* fix(cron): avoid announcing interim subagent spawn acks

* chore: clean post-rebase imports

* fix(cron): fall back to child replies when parent stays interim

* fix(subagents): make active-run guidance advisory

* fix(subagents): update announce flow to handle active descendants and enhance test coverage

- Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting.
- Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents.
- Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process.

* fix(subagents): enhance announce flow and formatting for user updates

- Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context.
- Refactored the announcement logic to improve clarity and ensure internal context remains private.
- Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates.
- Introduced a new function to build reply instructions based on session context, improving the overall announcement process.

* fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204)

* fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204)

* fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204)

* fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00

565 lines
20 KiB
TypeScript

import type { AgentCommandOpts } from "./agent/types.js";
import {
listAgentIds,
resolveAgentDir,
resolveAgentModelFallbacksOverride,
resolveAgentModelPrimary,
resolveAgentSkillsFilter,
resolveAgentWorkspaceDir,
} from "../agents/agent-scope.js";
import { ensureAuthProfileStore } from "../agents/auth-profiles.js";
import { clearSessionAuthProfileOverride } from "../agents/auth-profiles/session-override.js";
import { runCliAgent } from "../agents/cli-runner.js";
import { getCliSessionId } from "../agents/cli-session.js";
import { DEFAULT_MODEL, DEFAULT_PROVIDER } from "../agents/defaults.js";
import { AGENT_LANE_SUBAGENT } from "../agents/lanes.js";
import { loadModelCatalog } from "../agents/model-catalog.js";
import { runWithModelFallback } from "../agents/model-fallback.js";
import {
buildAllowedModelSet,
isCliProvider,
modelKey,
normalizeModelRef,
resolveConfiguredModelRef,
resolveThinkingDefault,
} from "../agents/model-selection.js";
import { runEmbeddedPiAgent } from "../agents/pi-embedded.js";
import { buildWorkspaceSkillSnapshot } from "../agents/skills.js";
import { getSkillsSnapshotVersion } from "../agents/skills/refresh.js";
import { resolveAgentTimeoutMs } from "../agents/timeout.js";
import { ensureAgentWorkspace } from "../agents/workspace.js";
import {
formatThinkingLevels,
formatXHighModelHint,
normalizeThinkLevel,
normalizeVerboseLevel,
supportsXHighThinking,
type ThinkLevel,
type VerboseLevel,
} from "../auto-reply/thinking.js";
import { formatCliCommand } from "../cli/command-format.js";
import { type CliDeps, createDefaultDeps } from "../cli/deps.js";
import { loadConfig } from "../config/config.js";
import {
resolveAgentIdFromSessionKey,
resolveSessionFilePath,
type SessionEntry,
updateSessionStore,
} from "../config/sessions.js";
import {
clearAgentRunContext,
emitAgentEvent,
registerAgentRunContext,
} from "../infra/agent-events.js";
import { getRemoteSkillEligibility } from "../infra/skills-remote.js";
import { normalizeAgentId } from "../routing/session-key.js";
import { defaultRuntime, type RuntimeEnv } from "../runtime.js";
import { applyVerboseOverride } from "../sessions/level-overrides.js";
import { applyModelOverrideToSessionEntry } from "../sessions/model-overrides.js";
import { resolveSendPolicy } from "../sessions/send-policy.js";
import { resolveMessageChannel } from "../utils/message-channel.js";
import { deliverAgentCommandResult } from "./agent/delivery.js";
import { resolveAgentRunContext } from "./agent/run-context.js";
import { updateSessionStoreAfterAgentRun } from "./agent/session-store.js";
import { resolveSession } from "./agent/session.js";
export async function agentCommand(
opts: AgentCommandOpts,
runtime: RuntimeEnv = defaultRuntime,
deps: CliDeps = createDefaultDeps(),
) {
const body = (opts.message ?? "").trim();
if (!body) {
throw new Error("Message (--message) is required");
}
if (!opts.to && !opts.sessionId && !opts.sessionKey && !opts.agentId) {
throw new Error("Pass --to <E.164>, --session-id, or --agent to choose a session");
}
const cfg = loadConfig();
const agentIdOverrideRaw = opts.agentId?.trim();
const agentIdOverride = agentIdOverrideRaw ? normalizeAgentId(agentIdOverrideRaw) : undefined;
if (agentIdOverride) {
const knownAgents = listAgentIds(cfg);
if (!knownAgents.includes(agentIdOverride)) {
throw new Error(
`Unknown agent id "${agentIdOverrideRaw}". Use "${formatCliCommand("openclaw agents list")}" to see configured agents.`,
);
}
}
if (agentIdOverride && opts.sessionKey) {
const sessionAgentId = resolveAgentIdFromSessionKey(opts.sessionKey);
if (sessionAgentId !== agentIdOverride) {
throw new Error(
`Agent id "${agentIdOverrideRaw}" does not match session key agent "${sessionAgentId}".`,
);
}
}
const agentCfg = cfg.agents?.defaults;
const sessionAgentId = agentIdOverride ?? resolveAgentIdFromSessionKey(opts.sessionKey?.trim());
const workspaceDirRaw = resolveAgentWorkspaceDir(cfg, sessionAgentId);
const agentDir = resolveAgentDir(cfg, sessionAgentId);
const workspace = await ensureAgentWorkspace({
dir: workspaceDirRaw,
ensureBootstrapFiles: !agentCfg?.skipBootstrap,
});
const workspaceDir = workspace.dir;
const configuredModel = resolveConfiguredModelRef({
cfg,
defaultProvider: DEFAULT_PROVIDER,
defaultModel: DEFAULT_MODEL,
});
const thinkingLevelsHint = formatThinkingLevels(configuredModel.provider, configuredModel.model);
const thinkOverride = normalizeThinkLevel(opts.thinking);
const thinkOnce = normalizeThinkLevel(opts.thinkingOnce);
if (opts.thinking && !thinkOverride) {
throw new Error(`Invalid thinking level. Use one of: ${thinkingLevelsHint}.`);
}
if (opts.thinkingOnce && !thinkOnce) {
throw new Error(`Invalid one-shot thinking level. Use one of: ${thinkingLevelsHint}.`);
}
const verboseOverride = normalizeVerboseLevel(opts.verbose);
if (opts.verbose && !verboseOverride) {
throw new Error('Invalid verbose level. Use "on", "full", or "off".');
}
const laneRaw = typeof opts.lane === "string" ? opts.lane.trim() : "";
const isSubagentLane = laneRaw === String(AGENT_LANE_SUBAGENT);
const timeoutSecondsRaw =
opts.timeout !== undefined
? Number.parseInt(String(opts.timeout), 10)
: isSubagentLane
? 0
: undefined;
if (
timeoutSecondsRaw !== undefined &&
(Number.isNaN(timeoutSecondsRaw) || timeoutSecondsRaw < 0)
) {
throw new Error("--timeout must be a non-negative integer (seconds; 0 means no timeout)");
}
const timeoutMs = resolveAgentTimeoutMs({
cfg,
overrideSeconds: timeoutSecondsRaw,
});
const sessionResolution = resolveSession({
cfg,
to: opts.to,
sessionId: opts.sessionId,
sessionKey: opts.sessionKey,
agentId: agentIdOverride,
});
const {
sessionId,
sessionKey,
sessionEntry: resolvedSessionEntry,
sessionStore,
storePath,
isNewSession,
persistedThinking,
persistedVerbose,
} = sessionResolution;
let sessionEntry = resolvedSessionEntry;
const runId = opts.runId?.trim() || sessionId;
try {
if (opts.deliver === true) {
const sendPolicy = resolveSendPolicy({
cfg,
entry: sessionEntry,
sessionKey,
channel: sessionEntry?.channel,
chatType: sessionEntry?.chatType,
});
if (sendPolicy === "deny") {
throw new Error("send blocked by session policy");
}
}
let resolvedThinkLevel =
thinkOnce ??
thinkOverride ??
persistedThinking ??
(agentCfg?.thinkingDefault as ThinkLevel | undefined);
const resolvedVerboseLevel =
verboseOverride ?? persistedVerbose ?? (agentCfg?.verboseDefault as VerboseLevel | undefined);
if (sessionKey) {
registerAgentRunContext(runId, {
sessionKey,
verboseLevel: resolvedVerboseLevel,
});
}
const needsSkillsSnapshot = isNewSession || !sessionEntry?.skillsSnapshot;
const skillsSnapshotVersion = getSkillsSnapshotVersion(workspaceDir);
const skillFilter = resolveAgentSkillsFilter(cfg, sessionAgentId);
const skillsSnapshot = needsSkillsSnapshot
? buildWorkspaceSkillSnapshot(workspaceDir, {
config: cfg,
eligibility: { remote: getRemoteSkillEligibility() },
snapshotVersion: skillsSnapshotVersion,
skillFilter,
})
: sessionEntry?.skillsSnapshot;
if (skillsSnapshot && sessionStore && sessionKey && needsSkillsSnapshot) {
const current = sessionEntry ?? {
sessionId,
updatedAt: Date.now(),
};
const next: SessionEntry = {
...current,
sessionId,
updatedAt: Date.now(),
skillsSnapshot,
};
sessionStore[sessionKey] = next;
await updateSessionStore(storePath, (store) => {
store[sessionKey] = next;
});
sessionEntry = next;
}
// Persist explicit /command overrides to the session store when we have a key.
if (sessionStore && sessionKey) {
const entry = sessionStore[sessionKey] ??
sessionEntry ?? { sessionId, updatedAt: Date.now() };
const next: SessionEntry = { ...entry, sessionId, updatedAt: Date.now() };
if (thinkOverride) {
next.thinkingLevel = thinkOverride;
}
applyVerboseOverride(next, verboseOverride);
sessionStore[sessionKey] = next;
await updateSessionStore(storePath, (store) => {
store[sessionKey] = next;
});
}
const agentModelPrimary = resolveAgentModelPrimary(cfg, sessionAgentId);
const cfgForModelSelection = agentModelPrimary
? {
...cfg,
agents: {
...cfg.agents,
defaults: {
...cfg.agents?.defaults,
model: {
...(typeof cfg.agents?.defaults?.model === "object"
? cfg.agents.defaults.model
: undefined),
primary: agentModelPrimary,
},
},
},
}
: cfg;
const configuredDefaultRef = resolveConfiguredModelRef({
cfg: cfgForModelSelection,
defaultProvider: DEFAULT_PROVIDER,
defaultModel: DEFAULT_MODEL,
});
const { provider: defaultProvider, model: defaultModel } = normalizeModelRef(
configuredDefaultRef.provider,
configuredDefaultRef.model,
);
let provider = defaultProvider;
let model = defaultModel;
const hasAllowlist = agentCfg?.models && Object.keys(agentCfg.models).length > 0;
const hasStoredOverride = Boolean(
sessionEntry?.modelOverride || sessionEntry?.providerOverride,
);
const needsModelCatalog = hasAllowlist || hasStoredOverride;
let allowedModelKeys = new Set<string>();
let allowedModelCatalog: Awaited<ReturnType<typeof loadModelCatalog>> = [];
let modelCatalog: Awaited<ReturnType<typeof loadModelCatalog>> | null = null;
if (needsModelCatalog) {
modelCatalog = await loadModelCatalog({ config: cfg });
const allowed = buildAllowedModelSet({
cfg,
catalog: modelCatalog,
defaultProvider,
defaultModel,
});
allowedModelKeys = allowed.allowedKeys;
allowedModelCatalog = allowed.allowedCatalog;
}
if (sessionEntry && sessionStore && sessionKey && hasStoredOverride) {
const entry = sessionEntry;
const overrideProvider = sessionEntry.providerOverride?.trim() || defaultProvider;
const overrideModel = sessionEntry.modelOverride?.trim();
if (overrideModel) {
const normalizedOverride = normalizeModelRef(overrideProvider, overrideModel);
const key = modelKey(normalizedOverride.provider, normalizedOverride.model);
if (
!isCliProvider(normalizedOverride.provider, cfg) &&
allowedModelKeys.size > 0 &&
!allowedModelKeys.has(key)
) {
const { updated } = applyModelOverrideToSessionEntry({
entry,
selection: { provider: defaultProvider, model: defaultModel, isDefault: true },
});
if (updated) {
sessionStore[sessionKey] = entry;
await updateSessionStore(storePath, (store) => {
store[sessionKey] = entry;
});
}
}
}
}
const storedProviderOverride = sessionEntry?.providerOverride?.trim();
const storedModelOverride = sessionEntry?.modelOverride?.trim();
if (storedModelOverride) {
const candidateProvider = storedProviderOverride || defaultProvider;
const normalizedStored = normalizeModelRef(candidateProvider, storedModelOverride);
const key = modelKey(normalizedStored.provider, normalizedStored.model);
if (
isCliProvider(normalizedStored.provider, cfg) ||
allowedModelKeys.size === 0 ||
allowedModelKeys.has(key)
) {
provider = normalizedStored.provider;
model = normalizedStored.model;
}
}
if (sessionEntry) {
const authProfileId = sessionEntry.authProfileOverride;
if (authProfileId) {
const entry = sessionEntry;
const store = ensureAuthProfileStore();
const profile = store.profiles[authProfileId];
if (!profile || profile.provider !== provider) {
if (sessionStore && sessionKey) {
await clearSessionAuthProfileOverride({
sessionEntry: entry,
sessionStore,
sessionKey,
storePath,
});
}
}
}
}
if (!resolvedThinkLevel) {
let catalogForThinking = modelCatalog ?? allowedModelCatalog;
if (!catalogForThinking || catalogForThinking.length === 0) {
modelCatalog = await loadModelCatalog({ config: cfg });
catalogForThinking = modelCatalog;
}
resolvedThinkLevel = resolveThinkingDefault({
cfg,
provider,
model,
catalog: catalogForThinking,
});
}
if (resolvedThinkLevel === "xhigh" && !supportsXHighThinking(provider, model)) {
const explicitThink = Boolean(thinkOnce || thinkOverride);
if (explicitThink) {
throw new Error(`Thinking level "xhigh" is only supported for ${formatXHighModelHint()}.`);
}
resolvedThinkLevel = "high";
if (sessionEntry && sessionStore && sessionKey && sessionEntry.thinkingLevel === "xhigh") {
const entry = sessionEntry;
entry.thinkingLevel = "high";
entry.updatedAt = Date.now();
sessionStore[sessionKey] = entry;
await updateSessionStore(storePath, (store) => {
store[sessionKey] = entry;
});
}
}
const sessionFile = resolveSessionFilePath(sessionId, sessionEntry, {
agentId: sessionAgentId,
});
const startedAt = Date.now();
let lifecycleEnded = false;
let result: Awaited<ReturnType<typeof runEmbeddedPiAgent>>;
let fallbackProvider = provider;
let fallbackModel = model;
try {
const runContext = resolveAgentRunContext(opts);
const messageChannel = resolveMessageChannel(
runContext.messageChannel,
opts.replyChannel ?? opts.channel,
);
const spawnedBy = opts.spawnedBy ?? sessionEntry?.spawnedBy;
// When a session has an explicit model override, prevent the fallback logic
// from silently appending the global primary model as a backstop. Passing an
// empty array (instead of undefined) tells resolveFallbackCandidates to skip
// the implicit primary append, so the session stays on its overridden model.
const agentFallbacksOverride = resolveAgentModelFallbacksOverride(cfg, sessionAgentId);
const effectiveFallbacksOverride = storedModelOverride
? (agentFallbacksOverride ?? [])
: agentFallbacksOverride;
// Track model fallback attempts so retries on an existing session don't
// re-inject the original prompt as a duplicate user message.
let fallbackAttemptIndex = 0;
const fallbackResult = await runWithModelFallback({
cfg,
provider,
model,
agentDir,
fallbacksOverride: effectiveFallbacksOverride,
run: (providerOverride, modelOverride) => {
const isFallbackRetry = fallbackAttemptIndex > 0;
fallbackAttemptIndex += 1;
// On fallback retries the session file already contains the original
// prompt from the first attempt. Re-injecting the full prompt would
// create a duplicate user message. Use a short continuation hint
// instead so the model picks up where it left off.
const effectivePrompt = isFallbackRetry
? "Continue where you left off. The previous model attempt failed or timed out."
: body;
if (isCliProvider(providerOverride, cfg)) {
const cliSessionId = getCliSessionId(sessionEntry, providerOverride);
return runCliAgent({
sessionId,
sessionKey,
agentId: sessionAgentId,
sessionFile,
workspaceDir,
config: cfg,
prompt: effectivePrompt,
provider: providerOverride,
model: modelOverride,
thinkLevel: resolvedThinkLevel,
timeoutMs,
runId,
extraSystemPrompt: opts.extraSystemPrompt,
cliSessionId,
images: isFallbackRetry ? undefined : opts.images,
streamParams: opts.streamParams,
});
}
const authProfileId =
providerOverride === provider ? sessionEntry?.authProfileOverride : undefined;
return runEmbeddedPiAgent({
sessionId,
sessionKey,
agentId: sessionAgentId,
messageChannel,
agentAccountId: runContext.accountId,
messageTo: opts.replyTo ?? opts.to,
messageThreadId: opts.threadId,
groupId: runContext.groupId,
groupChannel: runContext.groupChannel,
groupSpace: runContext.groupSpace,
spawnedBy,
currentChannelId: runContext.currentChannelId,
currentThreadTs: runContext.currentThreadTs,
replyToMode: runContext.replyToMode,
hasRepliedRef: runContext.hasRepliedRef,
senderIsOwner: true,
sessionFile,
workspaceDir,
config: cfg,
skillsSnapshot,
prompt: effectivePrompt,
images: isFallbackRetry ? undefined : opts.images,
clientTools: opts.clientTools,
provider: providerOverride,
model: modelOverride,
authProfileId,
authProfileIdSource: authProfileId
? sessionEntry?.authProfileOverrideSource
: undefined,
thinkLevel: resolvedThinkLevel,
verboseLevel: resolvedVerboseLevel,
timeoutMs,
runId,
lane: opts.lane,
abortSignal: opts.abortSignal,
extraSystemPrompt: opts.extraSystemPrompt,
inputProvenance: opts.inputProvenance,
streamParams: opts.streamParams,
agentDir,
onAgentEvent: (evt) => {
// Track lifecycle end for fallback emission below.
if (
evt.stream === "lifecycle" &&
typeof evt.data?.phase === "string" &&
(evt.data.phase === "end" || evt.data.phase === "error")
) {
lifecycleEnded = true;
}
},
});
},
});
result = fallbackResult.result;
fallbackProvider = fallbackResult.provider;
fallbackModel = fallbackResult.model;
if (!lifecycleEnded) {
emitAgentEvent({
runId,
stream: "lifecycle",
data: {
phase: "end",
startedAt,
endedAt: Date.now(),
aborted: result.meta.aborted ?? false,
},
});
}
} catch (err) {
if (!lifecycleEnded) {
emitAgentEvent({
runId,
stream: "lifecycle",
data: {
phase: "error",
startedAt,
endedAt: Date.now(),
error: String(err),
},
});
}
throw err;
}
// Update token+model fields in the session store.
if (sessionStore && sessionKey) {
await updateSessionStoreAfterAgentRun({
cfg,
contextTokensOverride: agentCfg?.contextTokens,
sessionId,
sessionKey,
storePath,
sessionStore,
defaultProvider: provider,
defaultModel: model,
fallbackProvider,
fallbackModel,
result,
});
}
const payloads = result.payloads ?? [];
return await deliverAgentCommandResult({
cfg,
deps,
runtime,
opts,
sessionEntry,
result,
payloads,
});
} finally {
clearAgentRunContext(runId);
}
}