Files
Moltbot/src/agents/model-selection.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

467 lines
13 KiB
TypeScript

import type { OpenClawConfig } from "../config/config.js";
import type { ModelCatalogEntry } from "./model-catalog.js";
import { resolveAgentModelPrimary } from "./agent-scope.js";
import { DEFAULT_MODEL, DEFAULT_PROVIDER } from "./defaults.js";
import { normalizeGoogleModelId } from "./models-config.providers.js";
export type ModelRef = {
provider: string;
model: string;
};
export type ThinkLevel = "off" | "minimal" | "low" | "medium" | "high" | "xhigh";
export type ModelAliasIndex = {
byAlias: Map<string, { alias: string; ref: ModelRef }>;
byKey: Map<string, string[]>;
};
const ANTHROPIC_MODEL_ALIASES: Record<string, string> = {
"opus-4.6": "claude-opus-4-6",
"opus-4.5": "claude-opus-4-5",
"sonnet-4.5": "claude-sonnet-4-5",
};
const OPENAI_CODEX_OAUTH_MODEL_PREFIXES = ["gpt-5.3-codex"] as const;
function normalizeAliasKey(value: string): string {
return value.trim().toLowerCase();
}
export function modelKey(provider: string, model: string) {
return `${provider}/${model}`;
}
export function normalizeProviderId(provider: string): string {
const normalized = provider.trim().toLowerCase();
if (normalized === "z.ai" || normalized === "z-ai") {
return "zai";
}
if (normalized === "opencode-zen") {
return "opencode";
}
if (normalized === "qwen") {
return "qwen-portal";
}
if (normalized === "kimi-code") {
return "kimi-coding";
}
return normalized;
}
export function isCliProvider(provider: string, cfg?: OpenClawConfig): boolean {
const normalized = normalizeProviderId(provider);
if (normalized === "claude-cli") {
return true;
}
if (normalized === "codex-cli") {
return true;
}
const backends = cfg?.agents?.defaults?.cliBackends ?? {};
return Object.keys(backends).some((key) => normalizeProviderId(key) === normalized);
}
function normalizeAnthropicModelId(model: string): string {
const trimmed = model.trim();
if (!trimmed) {
return trimmed;
}
const lower = trimmed.toLowerCase();
return ANTHROPIC_MODEL_ALIASES[lower] ?? trimmed;
}
function normalizeProviderModelId(provider: string, model: string): string {
if (provider === "anthropic") {
return normalizeAnthropicModelId(model);
}
if (provider === "google") {
return normalizeGoogleModelId(model);
}
return model;
}
function shouldUseOpenAICodexProvider(provider: string, model: string): boolean {
if (provider !== "openai") {
return false;
}
const normalized = model.trim().toLowerCase();
if (!normalized) {
return false;
}
return OPENAI_CODEX_OAUTH_MODEL_PREFIXES.some(
(prefix) => normalized === prefix || normalized.startsWith(`${prefix}-`),
);
}
export function normalizeModelRef(provider: string, model: string): ModelRef {
const normalizedProvider = normalizeProviderId(provider);
const normalizedModel = normalizeProviderModelId(normalizedProvider, model.trim());
if (shouldUseOpenAICodexProvider(normalizedProvider, normalizedModel)) {
return { provider: "openai-codex", model: normalizedModel };
}
return { provider: normalizedProvider, model: normalizedModel };
}
export function parseModelRef(raw: string, defaultProvider: string): ModelRef | null {
const trimmed = raw.trim();
if (!trimmed) {
return null;
}
const slash = trimmed.indexOf("/");
if (slash === -1) {
return normalizeModelRef(defaultProvider, trimmed);
}
const providerRaw = trimmed.slice(0, slash).trim();
const model = trimmed.slice(slash + 1).trim();
if (!providerRaw || !model) {
return null;
}
return normalizeModelRef(providerRaw, model);
}
export function resolveAllowlistModelKey(raw: string, defaultProvider: string): string | null {
const parsed = parseModelRef(raw, defaultProvider);
if (!parsed) {
return null;
}
return modelKey(parsed.provider, parsed.model);
}
export function buildConfiguredAllowlistKeys(params: {
cfg: OpenClawConfig | undefined;
defaultProvider: string;
}): Set<string> | null {
const rawAllowlist = Object.keys(params.cfg?.agents?.defaults?.models ?? {});
if (rawAllowlist.length === 0) {
return null;
}
const keys = new Set<string>();
for (const raw of rawAllowlist) {
const key = resolveAllowlistModelKey(String(raw ?? ""), params.defaultProvider);
if (key) {
keys.add(key);
}
}
return keys.size > 0 ? keys : null;
}
export function buildModelAliasIndex(params: {
cfg: OpenClawConfig;
defaultProvider: string;
}): ModelAliasIndex {
const byAlias = new Map<string, { alias: string; ref: ModelRef }>();
const byKey = new Map<string, string[]>();
const rawModels = params.cfg.agents?.defaults?.models ?? {};
for (const [keyRaw, entryRaw] of Object.entries(rawModels)) {
const parsed = parseModelRef(String(keyRaw ?? ""), params.defaultProvider);
if (!parsed) {
continue;
}
const alias = String((entryRaw as { alias?: string } | undefined)?.alias ?? "").trim();
if (!alias) {
continue;
}
const aliasKey = normalizeAliasKey(alias);
byAlias.set(aliasKey, { alias, ref: parsed });
const key = modelKey(parsed.provider, parsed.model);
const existing = byKey.get(key) ?? [];
existing.push(alias);
byKey.set(key, existing);
}
return { byAlias, byKey };
}
export function resolveModelRefFromString(params: {
raw: string;
defaultProvider: string;
aliasIndex?: ModelAliasIndex;
}): { ref: ModelRef; alias?: string } | null {
const trimmed = params.raw.trim();
if (!trimmed) {
return null;
}
if (!trimmed.includes("/")) {
const aliasKey = normalizeAliasKey(trimmed);
const aliasMatch = params.aliasIndex?.byAlias.get(aliasKey);
if (aliasMatch) {
return { ref: aliasMatch.ref, alias: aliasMatch.alias };
}
}
const parsed = parseModelRef(trimmed, params.defaultProvider);
if (!parsed) {
return null;
}
return { ref: parsed };
}
export function resolveConfiguredModelRef(params: {
cfg: OpenClawConfig;
defaultProvider: string;
defaultModel: string;
}): ModelRef {
const rawModel = (() => {
const raw = params.cfg.agents?.defaults?.model as { primary?: string } | string | undefined;
if (typeof raw === "string") {
return raw.trim();
}
return raw?.primary?.trim() ?? "";
})();
if (rawModel) {
const trimmed = rawModel.trim();
const aliasIndex = buildModelAliasIndex({
cfg: params.cfg,
defaultProvider: params.defaultProvider,
});
if (!trimmed.includes("/")) {
const aliasKey = normalizeAliasKey(trimmed);
const aliasMatch = aliasIndex.byAlias.get(aliasKey);
if (aliasMatch) {
return aliasMatch.ref;
}
// Default to anthropic if no provider is specified, but warn as this is deprecated.
console.warn(
`[openclaw] Model "${trimmed}" specified without provider. Falling back to "anthropic/${trimmed}". Please use "anthropic/${trimmed}" in your config.`,
);
return { provider: "anthropic", model: trimmed };
}
const resolved = resolveModelRefFromString({
raw: trimmed,
defaultProvider: params.defaultProvider,
aliasIndex,
});
if (resolved) {
return resolved.ref;
}
}
return { provider: params.defaultProvider, model: params.defaultModel };
}
export function resolveDefaultModelForAgent(params: {
cfg: OpenClawConfig;
agentId?: string;
}): ModelRef {
const agentModelOverride = params.agentId
? resolveAgentModelPrimary(params.cfg, params.agentId)
: undefined;
const cfg =
agentModelOverride && agentModelOverride.length > 0
? {
...params.cfg,
agents: {
...params.cfg.agents,
defaults: {
...params.cfg.agents?.defaults,
model: {
...(typeof params.cfg.agents?.defaults?.model === "object"
? params.cfg.agents.defaults.model
: undefined),
primary: agentModelOverride,
},
},
},
}
: params.cfg;
return resolveConfiguredModelRef({
cfg,
defaultProvider: DEFAULT_PROVIDER,
defaultModel: DEFAULT_MODEL,
});
}
export function buildAllowedModelSet(params: {
cfg: OpenClawConfig;
catalog: ModelCatalogEntry[];
defaultProvider: string;
defaultModel?: string;
}): {
allowAny: boolean;
allowedCatalog: ModelCatalogEntry[];
allowedKeys: Set<string>;
} {
const rawAllowlist = (() => {
const modelMap = params.cfg.agents?.defaults?.models ?? {};
return Object.keys(modelMap);
})();
const allowAny = rawAllowlist.length === 0;
const defaultModel = params.defaultModel?.trim();
const defaultKey =
defaultModel && params.defaultProvider
? modelKey(params.defaultProvider, defaultModel)
: undefined;
const catalogKeys = new Set(params.catalog.map((entry) => modelKey(entry.provider, entry.id)));
if (allowAny) {
if (defaultKey) {
catalogKeys.add(defaultKey);
}
return {
allowAny: true,
allowedCatalog: params.catalog,
allowedKeys: catalogKeys,
};
}
const allowedKeys = new Set<string>();
const configuredProviders = (params.cfg.models?.providers ?? {}) as Record<string, unknown>;
for (const raw of rawAllowlist) {
const parsed = parseModelRef(String(raw), params.defaultProvider);
if (!parsed) {
continue;
}
const key = modelKey(parsed.provider, parsed.model);
const providerKey = normalizeProviderId(parsed.provider);
if (isCliProvider(parsed.provider, params.cfg)) {
allowedKeys.add(key);
} else if (catalogKeys.has(key)) {
allowedKeys.add(key);
} else if (configuredProviders[providerKey] != null) {
// Explicitly configured providers should be allowlist-able even when
// they don't exist in the curated model catalog.
allowedKeys.add(key);
}
}
if (defaultKey) {
allowedKeys.add(defaultKey);
}
const allowedCatalog = params.catalog.filter((entry) =>
allowedKeys.has(modelKey(entry.provider, entry.id)),
);
if (allowedCatalog.length === 0 && allowedKeys.size === 0) {
if (defaultKey) {
catalogKeys.add(defaultKey);
}
return {
allowAny: true,
allowedCatalog: params.catalog,
allowedKeys: catalogKeys,
};
}
return { allowAny: false, allowedCatalog, allowedKeys };
}
export type ModelRefStatus = {
key: string;
inCatalog: boolean;
allowAny: boolean;
allowed: boolean;
};
export function getModelRefStatus(params: {
cfg: OpenClawConfig;
catalog: ModelCatalogEntry[];
ref: ModelRef;
defaultProvider: string;
defaultModel?: string;
}): ModelRefStatus {
const allowed = buildAllowedModelSet({
cfg: params.cfg,
catalog: params.catalog,
defaultProvider: params.defaultProvider,
defaultModel: params.defaultModel,
});
const key = modelKey(params.ref.provider, params.ref.model);
return {
key,
inCatalog: params.catalog.some((entry) => modelKey(entry.provider, entry.id) === key),
allowAny: allowed.allowAny,
allowed: allowed.allowAny || allowed.allowedKeys.has(key),
};
}
export function resolveAllowedModelRef(params: {
cfg: OpenClawConfig;
catalog: ModelCatalogEntry[];
raw: string;
defaultProvider: string;
defaultModel?: string;
}):
| { ref: ModelRef; key: string }
| {
error: string;
} {
const trimmed = params.raw.trim();
if (!trimmed) {
return { error: "invalid model: empty" };
}
const aliasIndex = buildModelAliasIndex({
cfg: params.cfg,
defaultProvider: params.defaultProvider,
});
const resolved = resolveModelRefFromString({
raw: trimmed,
defaultProvider: params.defaultProvider,
aliasIndex,
});
if (!resolved) {
return { error: `invalid model: ${trimmed}` };
}
const status = getModelRefStatus({
cfg: params.cfg,
catalog: params.catalog,
ref: resolved.ref,
defaultProvider: params.defaultProvider,
defaultModel: params.defaultModel,
});
if (!status.allowed) {
return { error: `model not allowed: ${status.key}` };
}
return { ref: resolved.ref, key: status.key };
}
export function resolveThinkingDefault(params: {
cfg: OpenClawConfig;
provider: string;
model: string;
catalog?: ModelCatalogEntry[];
}): ThinkLevel {
const configured = params.cfg.agents?.defaults?.thinkingDefault;
if (configured) {
return configured;
}
const candidate = params.catalog?.find(
(entry) => entry.provider === params.provider && entry.id === params.model,
);
if (candidate?.reasoning) {
return "low";
}
return "off";
}
/**
* Resolve the model configured for Gmail hook processing.
* Returns null if hooks.gmail.model is not set.
*/
export function resolveHooksGmailModel(params: {
cfg: OpenClawConfig;
defaultProvider: string;
}): ModelRef | null {
const hooksModel = params.cfg.hooks?.gmail?.model;
if (!hooksModel?.trim()) {
return null;
}
const aliasIndex = buildModelAliasIndex({
cfg: params.cfg,
defaultProvider: params.defaultProvider,
});
const resolved = resolveModelRefFromString({
raw: hooksModel,
defaultProvider: params.defaultProvider,
aliasIndex,
});
return resolved?.ref ?? null;
}