Files
openclaw/src/agents/model-selection.ts

627 lines
18 KiB
TypeScript
Raw Normal View History

2026-01-30 03:15:10 +01:00
import type { OpenClawConfig } from "../config/config.js";
import { resolveAgentModelPrimaryValue, toAgentModelListLike } from "../config/model-input.js";
import { createSubsystemLogger } from "../logging/subsystem.js";
import { resolveAgentConfig, resolveAgentEffectiveModelPrimary } from "./agent-scope.js";
import { DEFAULT_MODEL, DEFAULT_PROVIDER } from "./defaults.js";
import type { ModelCatalogEntry } from "./model-catalog.js";
import { splitTrailingAuthProfile } from "./model-ref-profile.js";
import { normalizeGoogleModelId } from "./models-config.providers.js";
2025-12-23 23:45:20 +00:00
const log = createSubsystemLogger("model-selection");
2025-12-23 23:45:20 +00:00
export type ModelRef = {
provider: string;
model: string;
};
export type ThinkLevel = "off" | "minimal" | "low" | "medium" | "high" | "xhigh" | "adaptive";
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-4.6": "claude-sonnet-4-6",
"sonnet-4.5": "claude-sonnet-4-5",
};
const CLAUDE_46_MODEL_RE = /claude-(?:opus|sonnet)-4(?:\.|-)6(?:$|[-.])/i;
function normalizeAliasKey(value: string): string {
return value.trim().toLowerCase();
}
2025-12-23 23:45:20 +00:00
export function modelKey(provider: string, model: string) {
return `${provider}/${model}`;
}
export function normalizeProviderId(provider: string): string {
2026-01-06 11:35:23 -03:00
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";
}
if (normalized === "bedrock" || normalized === "aws-bedrock") {
return "amazon-bedrock";
}
// Backward compatibility for older provider naming.
if (normalized === "bytedance" || normalized === "doubao") {
return "volcengine";
}
2026-01-06 11:35:23 -03:00
return normalized;
}
export function findNormalizedProviderValue<T>(
entries: Record<string, T> | undefined,
provider: string,
): T | undefined {
if (!entries) {
return undefined;
}
const providerKey = normalizeProviderId(provider);
for (const [key, value] of Object.entries(entries)) {
if (normalizeProviderId(key) === providerKey) {
return value;
}
}
return undefined;
}
export function findNormalizedProviderKey(
entries: Record<string, unknown> | undefined,
provider: string,
): string | undefined {
if (!entries) {
return undefined;
}
const providerKey = normalizeProviderId(provider);
return Object.keys(entries).find((key) => normalizeProviderId(key) === providerKey);
}
2026-01-30 03:15:10 +01:00
export function isCliProvider(provider: string, cfg?: OpenClawConfig): boolean {
2026-01-10 23:31:25 +00:00
const normalized = normalizeProviderId(provider);
if (normalized === "claude-cli") {
return true;
}
if (normalized === "codex-cli") {
return true;
}
2026-01-10 23:31:25 +00:00
const backends = cfg?.agents?.defaults?.cliBackends ?? {};
return Object.keys(backends).some((key) => normalizeProviderId(key) === normalized);
2026-01-10 23:31:25 +00:00
}
2026-01-09 05:07:16 +00:00
function normalizeAnthropicModelId(model: string): string {
const trimmed = model.trim();
if (!trimmed) {
return trimmed;
}
2026-01-09 05:07:16 +00:00
const lower = trimmed.toLowerCase();
return ANTHROPIC_MODEL_ALIASES[lower] ?? trimmed;
2026-01-09 05:07:16 +00:00
}
function normalizeProviderModelId(provider: string, model: string): string {
if (provider === "anthropic") {
return normalizeAnthropicModelId(model);
}
if (provider === "vercel-ai-gateway" && !model.includes("/")) {
// Allow Vercel-specific Claude refs without an upstream prefix.
const normalizedAnthropicModel = normalizeAnthropicModelId(model);
if (normalizedAnthropicModel.startsWith("claude-")) {
return `anthropic/${normalizedAnthropicModel}`;
}
}
if (provider === "google") {
return normalizeGoogleModelId(model);
}
// OpenRouter-native models (e.g. "openrouter/aurora-alpha") need the full
// "openrouter/<name>" as the model ID sent to the API. Models from external
// providers already contain a slash (e.g. "anthropic/claude-sonnet-4-5") and
// are passed through as-is (#12924).
if (provider === "openrouter" && !model.includes("/")) {
return `openrouter/${model}`;
}
return model;
}
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
export function normalizeModelRef(provider: string, model: string): ModelRef {
const normalizedProvider = normalizeProviderId(provider);
const normalizedModel = normalizeProviderModelId(normalizedProvider, model.trim());
return { provider: normalizedProvider, model: normalizedModel };
}
export function parseModelRef(raw: string, defaultProvider: string): ModelRef | null {
2025-12-23 23:45:20 +00:00
const trimmed = raw.trim();
if (!trimmed) {
return null;
}
2025-12-23 23:45:20 +00:00
const slash = trimmed.indexOf("/");
if (slash === -1) {
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
return normalizeModelRef(defaultProvider, trimmed);
2025-12-23 23:45:20 +00:00
}
2026-01-06 11:35:23 -03:00
const providerRaw = trimmed.slice(0, slash).trim();
2025-12-23 23:45:20 +00:00
const model = trimmed.slice(slash + 1).trim();
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
if (!providerRaw || !model) {
return null;
}
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
return normalizeModelRef(providerRaw, model);
2025-12-23 23:45:20 +00:00
}
export function inferUniqueProviderFromConfiguredModels(params: {
cfg: OpenClawConfig;
model: string;
}): string | undefined {
const model = params.model.trim();
if (!model) {
return undefined;
}
const configuredModels = params.cfg.agents?.defaults?.models;
if (!configuredModels) {
return undefined;
}
const normalized = model.toLowerCase();
const providers = new Set<string>();
for (const key of Object.keys(configuredModels)) {
const ref = key.trim();
if (!ref || !ref.includes("/")) {
continue;
}
const parsed = parseModelRef(ref, DEFAULT_PROVIDER);
if (!parsed) {
continue;
}
if (parsed.model === model || parsed.model.toLowerCase() === normalized) {
providers.add(parsed.provider);
if (providers.size > 1) {
return undefined;
}
}
}
if (providers.size !== 1) {
return undefined;
}
return providers.values().next().value;
}
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: {
2026-01-30 03:15:10 +01:00
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 { model } = splitTrailingAuthProfile(params.raw);
if (!model) {
return null;
}
if (!model.includes("/")) {
const aliasKey = normalizeAliasKey(model);
const aliasMatch = params.aliasIndex?.byAlias.get(aliasKey);
if (aliasMatch) {
return { ref: aliasMatch.ref, alias: aliasMatch.alias };
}
}
const parsed = parseModelRef(model, params.defaultProvider);
if (!parsed) {
return null;
}
return { ref: parsed };
}
export function resolveConfiguredModelRef(params: {
2026-01-30 03:15:10 +01:00
cfg: OpenClawConfig;
defaultProvider: string;
defaultModel: string;
}): ModelRef {
const rawModel = resolveAgentModelPrimaryValue(params.cfg.agents?.defaults?.model) ?? "";
if (rawModel) {
2025-12-26 00:43:44 +01:00
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.
log.warn(
`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;
}
}
2025-12-26 00:43:44 +01:00
return { provider: params.defaultProvider, model: params.defaultModel };
}
export function resolveDefaultModelForAgent(params: {
2026-01-30 03:15:10 +01:00
cfg: OpenClawConfig;
agentId?: string;
}): ModelRef {
const agentModelOverride = params.agentId
? resolveAgentEffectiveModelPrimary(params.cfg, params.agentId)
: undefined;
const cfg =
agentModelOverride && agentModelOverride.length > 0
? {
...params.cfg,
agents: {
...params.cfg.agents,
defaults: {
...params.cfg.agents?.defaults,
model: {
...toAgentModelListLike(params.cfg.agents?.defaults?.model),
primary: agentModelOverride,
},
},
},
}
: params.cfg;
return resolveConfiguredModelRef({
cfg,
defaultProvider: DEFAULT_PROVIDER,
defaultModel: DEFAULT_MODEL,
});
}
export function resolveSubagentConfiguredModelSelection(params: {
cfg: OpenClawConfig;
agentId: string;
}): string | undefined {
const agentConfig = resolveAgentConfig(params.cfg, params.agentId);
return (
normalizeModelSelection(agentConfig?.subagents?.model) ??
normalizeModelSelection(params.cfg.agents?.defaults?.subagents?.model) ??
normalizeModelSelection(agentConfig?.model)
);
}
export function resolveSubagentSpawnModelSelection(params: {
cfg: OpenClawConfig;
agentId: string;
modelOverride?: unknown;
}): string {
const runtimeDefault = resolveDefaultModelForAgent({
cfg: params.cfg,
agentId: params.agentId,
});
return (
normalizeModelSelection(params.modelOverride) ??
resolveSubagentConfiguredModelSelection({
cfg: params.cfg,
agentId: params.agentId,
}) ??
normalizeModelSelection(resolveAgentModelPrimaryValue(params.cfg.agents?.defaults?.model)) ??
`${runtimeDefault.provider}/${runtimeDefault.model}`
);
}
2025-12-23 23:45:20 +00:00
export function buildAllowedModelSet(params: {
2026-01-30 03:15:10 +01:00
cfg: OpenClawConfig;
2025-12-23 23:45:20 +00:00
catalog: ModelCatalogEntry[];
defaultProvider: string;
defaultModel?: string;
2025-12-23 23:45:20 +00:00
}): {
allowAny: boolean;
allowedCatalog: ModelCatalogEntry[];
allowedKeys: Set<string>;
} {
const rawAllowlist = (() => {
const modelMap = params.cfg.agents?.defaults?.models ?? {};
return Object.keys(modelMap);
})();
2025-12-23 23:45:20 +00:00
const allowAny = rawAllowlist.length === 0;
const defaultModel = params.defaultModel?.trim();
const defaultRef =
defaultModel && params.defaultProvider
? parseModelRef(defaultModel, params.defaultProvider)
: null;
const defaultKey = defaultRef ? modelKey(defaultRef.provider, defaultRef.model) : undefined;
const catalogKeys = new Set(params.catalog.map((entry) => modelKey(entry.provider, entry.id)));
2025-12-23 23:45:20 +00:00
if (allowAny) {
if (defaultKey) {
catalogKeys.add(defaultKey);
}
2025-12-23 23:45:20 +00:00
return {
allowAny: true,
allowedCatalog: params.catalog,
allowedKeys: catalogKeys,
};
}
const allowedKeys = new Set<string>();
const syntheticCatalogEntries = new Map<string, ModelCatalogEntry>();
2025-12-23 23:45:20 +00:00
for (const raw of rawAllowlist) {
const parsed = parseModelRef(String(raw), params.defaultProvider);
if (!parsed) {
continue;
}
2025-12-23 23:45:20 +00:00
const key = modelKey(parsed.provider, parsed.model);
// Explicit allowlist entries are always trusted, even when bundled catalog
// data is stale and does not include the configured model yet.
allowedKeys.add(key);
if (!catalogKeys.has(key) && !syntheticCatalogEntries.has(key)) {
syntheticCatalogEntries.set(key, {
id: parsed.model,
name: parsed.model,
provider: parsed.provider,
});
2025-12-23 23:45:20 +00:00
}
}
if (defaultKey) {
allowedKeys.add(defaultKey);
}
const allowedCatalog = [
...params.catalog.filter((entry) => allowedKeys.has(modelKey(entry.provider, entry.id))),
...syntheticCatalogEntries.values(),
];
2025-12-23 23:45:20 +00:00
2026-01-10 23:31:25 +00:00
if (allowedCatalog.length === 0 && allowedKeys.size === 0) {
if (defaultKey) {
catalogKeys.add(defaultKey);
}
2025-12-23 23:45:20 +00:00
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: {
2026-01-30 03:15:10 +01:00
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: {
2026-01-30 03:15:10 +01:00
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: {
2026-01-30 03:15:10 +01:00
cfg: OpenClawConfig;
provider: string;
model: string;
catalog?: ModelCatalogEntry[];
}): ThinkLevel {
const normalizedProvider = normalizeProviderId(params.provider);
const modelLower = params.model.toLowerCase();
const perModelThinking =
params.cfg.agents?.defaults?.models?.[modelKey(params.provider, params.model)]?.params
?.thinking;
if (
perModelThinking === "off" ||
perModelThinking === "minimal" ||
perModelThinking === "low" ||
perModelThinking === "medium" ||
perModelThinking === "high" ||
perModelThinking === "xhigh" ||
perModelThinking === "adaptive"
) {
return perModelThinking;
}
const configured = params.cfg.agents?.defaults?.thinkingDefault;
if (configured) {
return configured;
}
const isAnthropicFamilyModel =
normalizedProvider === "anthropic" ||
normalizedProvider === "amazon-bedrock" ||
modelLower.includes("anthropic/") ||
modelLower.includes(".anthropic.");
if (isAnthropicFamilyModel && CLAUDE_46_MODEL_RE.test(modelLower)) {
return "adaptive";
}
const candidate = params.catalog?.find(
2026-01-03 12:35:09 +00:00
(entry) => entry.provider === params.provider && entry.id === params.model,
);
if (candidate?.reasoning) {
return "low";
}
return "off";
}
/** Default reasoning level when session/directive do not set it: "on" if model supports reasoning, else "off". */
export function resolveReasoningDefault(params: {
provider: string;
model: string;
catalog?: ModelCatalogEntry[];
}): "on" | "off" {
const key = modelKey(params.provider, params.model);
const candidate = params.catalog?.find(
(entry) =>
(entry.provider === params.provider && entry.id === params.model) ||
(entry.provider === key && entry.id === params.model),
);
return candidate?.reasoning === true ? "on" : "off";
}
/**
* Resolve the model configured for Gmail hook processing.
* Returns null if hooks.gmail.model is not set.
*/
export function resolveHooksGmailModel(params: {
2026-01-30 03:15:10 +01:00
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;
}
/**
* Normalize a model selection value (string or `{primary?: string}`) to a
* plain trimmed string. Returns `undefined` when the input is empty/missing.
* Shared by sessions-spawn and cron isolated-agent model resolution.
*/
export function normalizeModelSelection(value: unknown): string | undefined {
if (typeof value === "string") {
const trimmed = value.trim();
return trimmed || undefined;
}
if (!value || typeof value !== "object") {
return undefined;
}
const primary = (value as { primary?: unknown }).primary;
if (typeof primary === "string" && primary.trim()) {
return primary.trim();
}
return undefined;
}