@deepseek-ai/dsh-tool-subagent 0.1.1-rc.2 → 0.1.2-alpha.2

package/lib/index.js

@@ -1,6 +1,236 @@
 import z from "@deepseek-ai/schemastery";
+import { scopeChainOf, scopeOf } from "@deepseek-ai/dsh-scope";
 import { defineTool } from "@deepseek-ai/dsh-tools";
-import { assertSubagentMaxDepth, settleRun } from "@deepseek-ai/dsh-subagent";
+import { assertSubagentMaxDepth, parentAgentOptionsForDelegation, settleRun } from "@deepseek-ai/dsh-subagent";
+import { ReasoningEffortId } from "@deepseek-ai/dsh-llm";
+import { z as z$1 } from "zod";
+z.object({
+	provider: z.string().min(1).required(),
+	model: z.string().min(1).required()
+});
+/**
+* Stable identity for one provider/model pair.
+* @param route - Exact provider/model route.
+* @returns Opaque key for equality checks.
+*/
+function modelRouteKey(route) {
+	return `${route.provider}\0${route.model}`;
+}
+/**
+* Reject malformed or duplicate route policy entries at a durable or configuration boundary.
+* @param routes - Candidate exact routes to validate.
+* @returns an assertion that the candidate is a validated exact-route array.
+*/
+function assertAllowedModelRoutes(routes) {
+	if (!Array.isArray(routes)) throw new Error("subagent model selection requires an array of routes");
+	const seen = /* @__PURE__ */ new Set();
+	const candidates = routes;
+	for (const candidate of candidates) {
+		if (typeof candidate !== "object" || candidate === null || Array.isArray(candidate) || !("provider" in candidate) || typeof candidate.provider !== "string" || !("model" in candidate) || typeof candidate.model !== "string" || candidate.provider.length === 0 || candidate.model.length === 0) throw new Error("subagent model selection requires non-empty provider and model ids");
+		const route = {
+			provider: candidate.provider,
+			model: candidate.model
+		};
+		const key = modelRouteKey(route);
+		if (seen.has(key)) throw new Error(`subagent model selection repeats route "${route.provider}/${route.model}"`);
+		seen.add(key);
+	}
+}
+/**
+* Whether a call explicitly selects any child LLM value.
+* @param request - Model-facing route fields from the tool call.
+* @returns Whether at least one route or effort field is present.
+*/
+function hasDelegationModelRequest(request) {
+	return request.provider !== void 0 || request.model !== void 0 || request.reasoning_effort !== void 0;
+}
+/** Reject an empty model-facing route value at the tool JSON boundary. */
+function assertNonEmpty(value, field) {
+	if (value !== void 0 && value.length === 0) throw new Error(`child LLM \`${field}\` must be non-empty`);
+}
+/**
+* Merge model-supplied selection fields over configured child defaults.
+* Provider and model form one route and must be supplied together. Changing
+* that route without an effort clears the configured route-owned effort.
+* @param parentOptions - Current parent values that supply missing child values.
+* @param configured - Tool-instance child defaults.
+* @param request - Model-facing route override.
+* @param enabled - Whether this tool instance permits model-facing selection.
+* @returns Child Agent options, preserving omission when no layer contributes one.
+*/
+function requestedAgentOptions(parentOptions, configured, request, enabled) {
+	if (!hasDelegationModelRequest(request)) return configured;
+	if (!enabled) throw new Error("child model selection is disabled for this tool instance");
+	assertNonEmpty(request.provider, "provider");
+	assertNonEmpty(request.model, "model");
+	assertNonEmpty(request.reasoning_effort, "reasoning_effort");
+	if (request.provider === void 0 !== (request.model === void 0)) throw new Error("child LLM `provider` and `model` must be supplied together");
+	const baselineProvider = configured?.provider ?? parentOptions.provider;
+	const baselineModel = configured?.model ?? parentOptions.model;
+	const routeChanged = request.provider !== void 0 && (request.provider !== baselineProvider || request.model !== baselineModel);
+	const { reasoningEffort: _configuredReasoningEffort, ...configuredWithoutReasoning } = configured ?? {};
+	return {
+		...routeChanged && request.reasoning_effort === void 0 ? configuredWithoutReasoning : configured,
+		...request.provider === void 0 ? {} : {
+			provider: request.provider,
+			model: request.model
+		},
+		...request.reasoning_effort === void 0 ? {} : { reasoningEffort: ReasoningEffortId(request.reasoning_effort) }
+	};
+}
+/**
+* Enforce a settings-owned route list at the operation that creates the child.
+* Pure inheritance remains outside this policy because no model-facing choice
+* occurred; any explicit route or effort field must resolve to an allowed route.
+* @param policy - Selection authority captured for this Session.
+* @param parentOptions - Current parent values that supply missing child values.
+* @param requested - Effective child options after request/config merging.
+* @param request - Model-facing selection fields from the tool call.
+*/
+function assertAllowedModelSelection(policy, parentOptions, requested, request) {
+	if (policy === void 0 || !hasDelegationModelRequest(request)) return;
+	const provider = requested?.provider ?? parentOptions.provider;
+	const model = requested?.model ?? parentOptions.model;
+	if (provider === void 0 || model === void 0) throw new Error("cannot select child LLM values without an effective provider and model");
+	if (policy.routes.some((route) => route.provider === provider && route.model === model)) return;
+	throw new Error(`child LLM route "${provider}/${model}" is not allowed for this Session`);
+}
+/**
+* Whether configured Agent options require route validation before delegation.
+* @param options - Tool-instance child defaults.
+* @returns Whether configured provider, model, or effort values must be resolved.
+*/
+function hasConfiguredLlmSelection(options) {
+	return options?.provider !== void 0 || options?.model !== void 0 || options?.reasoningEffort !== void 0;
+}
+/**
+* Resolve an effective child route through its live adapter before the child is
+* created. The LLM runtime owns provider lookup, exact-model metadata, effort
+* validation, and adapter defaults.
+* @param llm - Live LLM runtime.
+* @param parentOptions - Current parent values whose compatible fields the child inherits.
+* @param requested - Per-child options after request/config merging.
+* @param signal - Tool-call cancellation signal.
+* @param inheritParentReasoningEffort - Whether an omitted effort may inherit from the parent route.
+*/
+async function preflightChildLlmRoute(llm, parentOptions, requested, signal, inheritParentReasoningEffort = true) {
+	const provider = requested?.provider ?? parentOptions.provider;
+	const model = requested?.model ?? parentOptions.model;
+	if (provider === void 0 || model === void 0) throw new Error("cannot select child LLM values without an effective provider and model");
+	const routeChanged = provider !== parentOptions.provider || model !== parentOptions.model;
+	const reasoningEffort = requested?.reasoningEffort ?? (inheritParentReasoningEffort && !routeChanged ? parentOptions.reasoningEffort : void 0);
+	await llm.resolveCallConfig({
+		provider,
+		model,
+		...reasoningEffort === void 0 ? {} : { reasoningEffort }
+	}, signal);
+}
+//#endregion
+//#region lib/types/list-models.js
+/** Model-facing discovery of LLM routes available to child Agents. */
+/** Resolve one registered provider with a model-correctable diagnostic. */
+function registeredProvider(llm, policy, providerId) {
+	const providers = llm.listProviders();
+	const provider = providers.find((candidate) => candidate.id === providerId);
+	if (provider !== void 0) return provider;
+	const available = providers.filter((candidate) => policy.routes.some((route) => route.provider === candidate.id)).map((candidate) => candidate.id).join(", ") || "(none)";
+	throw new Error(`LLM provider "${providerId}" is not registered; available providers: ${available}`);
+}
+/** Render one advertised or resolved model. */
+function modelLine(provider, model) {
+	return `${provider}/${model.id} — ${model.name}${model.description === void 0 ? "" : `: ${model.description}`}`;
+}
+/** Read the requested provider, advertised models, or exact-model efforts. */
+async function listSubagentModels(ctx, policy, request, signal) {
+	const llm = ctx.get("llm");
+	if (llm === void 0) throw new Error("cannot discover child LLM routes because the `llm` service is unavailable");
+	if (request.model !== void 0 && request.provider === void 0) throw new Error("`model` requires `provider`");
+	if (request.provider === void 0) {
+		const providers = llm.listProviders().filter((provider) => policy.routes.some((route) => route.provider === provider.id));
+		return providers.length === 0 ? "(no LLM providers)" : providers.map((provider) => `${provider.id} — ${provider.name}`).join("\n");
+	}
+	if (request.provider.length === 0) throw new Error("`provider` must be non-empty");
+	const allowedRoutes = policy.routes.filter((route) => route.provider === request.provider);
+	if (allowedRoutes.length === 0) throw new Error(`LLM provider "${request.provider}" is not allowed for this Session`);
+	const provider = registeredProvider(llm, policy, request.provider);
+	if (request.model === void 0) {
+		const models = (await llm.listModels(provider.id)).filter((model) => allowedRoutes.some((route) => route.model === model.id));
+		return models.length === 0 ? `(no advertised models for ${provider.id})` : models.map((model) => modelLine(provider.id, model)).join("\n");
+	}
+	if (request.model.length === 0) throw new Error("`model` must be non-empty");
+	if (!allowedRoutes.some((route) => route.model === request.model)) throw new Error(`child LLM route "${provider.id}/${request.model}" is not allowed for this Session`);
+	const model = await llm.resolveModelInfo(provider.id, request.model, signal);
+	const efforts = model.reasoning?.efforts.map((effort) => `${effort.id}${model.reasoning?.defaultEffort === effort.id ? " (default)" : ""} — ${effort.name}` + (effort.description === void 0 ? "" : `: ${effort.description}`)).join("\n") || "(no advertised reasoning efforts)";
+	return `${modelLine(provider.id, model)}\nReasoning efforts:\n${efforts}`;
+}
+/**
+* Register `list_subagent_models` for one owning delegation-tool instance.
+* @param ctx - Context whose tool registry owns the fixed discovery definition.
+* @param policy - Route policy captured for this Session.
+*/
+function registerListSubagentModels(ctx, policy) {
+	ctx.tools.register(defineTool({
+		name: "list_subagent_models",
+		description: "Discover LLM routes for subagents without changing the current Agent. Call with no arguments to list registered providers, with `provider` to list its advertised models, or with `provider` and `model` to inspect that exact model and its reasoning efforts. Catalog membership is advisory: an adapter may accept an unlisted model id. Use the returned ids with a delegation tool's `provider`, `model`, and `reasoning_effort` fields.",
+		parameters: {
+			provider: {
+				type: "string",
+				description: "Registered LLM provider id. Omit to list providers."
+			},
+			model: {
+				type: "string",
+				description: "Exact model id to inspect. Requires provider; omit to list that provider's advertised models."
+			}
+		},
+		output: {
+			schema: { type: "string" },
+			render: (_args, result) => [{
+				type: "text",
+				text: result
+			}]
+		},
+		execute(args, exec) {
+			return listSubagentModels(ctx, policy, args, exec.signal);
+		}
+	}));
+}
+/** Host-only projection of the durable model-selection policy. */
+const subagentModelSelectionProjectionDefinition = {
+	key: "subagentModelSelectionPolicy",
+	stateVersion: 1,
+	stateSchema: z$1.array(z$1.object({
+		provider: z$1.string().min(1),
+		model: z$1.string().min(1)
+	}).strict()).min(1).nullable(),
+	init: () => null,
+	apply: (policy, event) => {
+		if (policy !== null || event.type !== "subagent/model-selection-policy") return policy;
+		const { allowedModels } = event.data;
+		assertAllowedModelRoutes(allowedModels);
+		if (allowedModels.length === 0) throw new Error("subagent/model-selection-policy requires at least one route");
+		return allowedModels;
+	}
+};
+/**
+* Read the exact route list captured for a model-selectable definition.
+* @param projections - registry that owns the policy projection.
+* @param session - session whose durable decision is read.
+* @returns a detached route list, or undefined for the fixed-route definition.
+*/
+function subagentModelSelectionPolicy(projections, session) {
+	return projections.stateOf(session, "subagentModelSelectionPolicy")?.map((route) => ({ ...route }));
+}
+/**
+* Append the route policy once, before its definition can reach a model request.
+* @param projections - registry that owns the policy projection.
+* @param session - session receiving the model-selectable definition.
+* @param allowedModels - exact routes the definition may select explicitly.
+*/
+function recordSubagentModelSelection(projections, session, allowedModels) {
+	if (subagentModelSelectionPolicy(projections, session) !== void 0) return;
+	session.append("subagent/model-selection-policy", { allowedModels: allowedModels.map((route) => ({ ...route })) });
+}
+//#endregion
 //#region lib/types/index.js
 /**
 * Model-facing delegation through one configured `ctx.subagents` provider.
@@ -15,18 +245,19 @@ const name = "tool-subagent";
 const inject = [
 	"tools",
 	"subagents",
-	"systemPrompt"
+	"systemPrompt",
+	"sessionProjections"
 ];
-/** Prompt order after bounded delegation policy and before child reporting. */
-const SUBAGENT_SECTION_ORDER = 116.5;
 const Config = z.object({
 	provider: z.string().required(),
 	toolName: z.string().default("subagent"),
+	modelSelectionSettings: z.boolean().default(false),
 	enableRunInBackground: z.boolean().default(true),
 	backgroundMode: z.union(["one-shot", "continuable"]).default("one-shot"),
 	agentOptions: z.object({
 		provider: z.string(),
 		model: z.string(),
+		reasoningEffort: z.string().min(1),
 		maxTokens: z.number().step(1).min(1).max(Number.MAX_SAFE_INTEGER)
 	}).default(void 0),
 	persona: z.string(),
@@ -133,164 +364,286 @@ function apply(ctx, config) {
 	const backgroundEnabled = config.enableRunInBackground !== false;
 	const continuable = (config.backgroundMode ?? "one-shot") === "continuable";
 	const toolName = config.toolName ?? "subagent";
-	let disposeTool;
-	const mount = (provider) => {
-		if (typeof config.maxDepth === "number" && !provider.capabilities.depthLimit) throw new Error(`tool-subagent: provider "${provider.name}" cannot enforce maxDepth (no depthLimit capability) — set maxDepth: 'provider-managed' to leave the recursion budget to the provider`);
-		const wording = providerWording(provider.inheritsParentContext);
-		if (continuable && provider.prepareContinuable === void 0) throw new Error(`tool-subagent: provider "${provider.name}" does not support \`backgroundMode: continuable\``);
-		disposeTool = ctx.tools.register(defineTool({
-			name: toolName,
-			description: wording.description + (backgroundEnabled ? continuable ? " This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result." : " This call waits for the result by default. Set `run_in_background: true` to return a job id; collect with `job_output` and stop with `job_kill`." : " This call waits for the subagent and returns its result."),
-			parameters: {
-				description: {
-					type: "string",
-					required: true,
-					description: "A short (3-5 word) description of the delegated task, for display."
-				},
-				prompt: {
-					type: "string",
-					required: true,
-					description: wording.promptDescription
-				},
-				...backgroundEnabled ? { run_in_background: {
-					type: "boolean",
-					description: continuable ? "Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it." : "Whether to run as a background job and return its id. Defaults to false; collect with job_output or stop with job_kill."
-				} } : {}
-			},
-			output: {
-				schema: { oneOf: [
-					{
-						type: "object",
-						additionalProperties: false,
-						properties: {
-							kind: {
+	const modelSelectionCapable = config.modelSelectionSettings === true;
+	ctx.sessionProjections.register(subagentModelSelectionProjectionDefinition);
+	const assertSubagentProviderConfiguration = (subagentProvider) => {
+		if (typeof config.maxDepth === "number" && !subagentProvider.capabilities.depthLimit) throw new Error(`tool-subagent: provider "${subagentProvider.name}" cannot enforce maxDepth (no depthLimit capability) — set maxDepth: 'provider-managed' to leave the recursion budget to the provider`);
+		if (config.agentOptions !== void 0 && !subagentProvider.capabilities.agentOptions) throw new Error(`tool-subagent: provider "${subagentProvider.name}" does not support child agentOptions`);
+		if (modelSelectionCapable && !subagentProvider.capabilities.agentOptions) throw new Error(`tool-subagent: provider "${subagentProvider.name}" does not support child model selection`);
+		if (continuable && subagentProvider.prepareContinuable === void 0) throw new Error(`tool-subagent: provider "${subagentProvider.name}" does not support \`backgroundMode: continuable\``);
+	};
+	ctx.on("subagent/provider-added", (subagentProvider) => {
+		if (subagentProvider.name === config.provider) assertSubagentProviderConfiguration(subagentProvider);
+	});
+	const initialProvider = ctx.subagents.getProvider(config.provider);
+	if (initialProvider !== void 0) assertSubagentProviderConfiguration(initialProvider);
+	const install = (runtimeCtx, modelSelectionPolicy) => {
+		const modelSelectionEnabled = modelSelectionPolicy !== void 0;
+		if (modelSelectionPolicy !== void 0) registerListSubagentModels(runtimeCtx, modelSelectionPolicy);
+		let mounted;
+		const mount = (subagentProvider) => {
+			assertSubagentProviderConfiguration(subagentProvider);
+			const wording = providerWording(subagentProvider.inheritsParentContext);
+			const providerRouteDefaults = subagentProvider.agentRouteDefaults;
+			const choiceDescription = !modelSelectionEnabled ? "" : (providerRouteDefaults !== void 0 ? " Child LLM selection is optional. Omit `provider`, `model`, and `reasoning_effort` to use configured child defaults and this provider's route defaults. Supply `provider` and `model` together after using `list_subagent_models` to inspect advertised routes and efforts. Changing the effective route without naming an effort uses the selected model's default effort." : " Child LLM selection is optional. Omit `provider`, `model`, and `reasoning_effort` to use configured child defaults and inherit compatible missing values from the parent Agent. Supply `provider` and `model` together after using `list_subagent_models` to inspect advertised routes and efforts. Changing the effective route without naming an effort uses the selected model's default effort.") + (subagentProvider.inheritsParentContext ? " Changing the route can prevent provider-side reuse of the inherited conversation prefix." : "");
+			mounted = {
+				subagentProvider,
+				disposeTool: runtimeCtx.tools.register(defineTool({
+					name: toolName,
+					description: wording.description + (backgroundEnabled ? continuable ? " This tool runs in the background by default, immediately returns a durable subagent id, and keeps the child conversation available for later turns. When that run settles, the runtime sends the parent a notice containing its outcome and any final assistant message; `send_message` starts a later turn in the same child conversation. Set `run_in_background: false` only when your next action depends on receiving the result." : " This call waits for the result by default. Set `run_in_background: true` to return a job id; collect with `job_output` and stop with `job_kill`." : " This call waits for the subagent and returns its result.") + choiceDescription,
+					parameters: {
+						description: {
+							type: "string",
+							required: true,
+							description: "A short (3-5 word) description of the delegated task, for display."
+						},
+						prompt: {
+							type: "string",
+							required: true,
+							description: wording.promptDescription
+						},
+						...modelSelectionEnabled ? {
+							provider: {
 								type: "string",
-								required: true,
-								const: "background"
+								description: providerRouteDefaults !== void 0 ? "LLM provider route for the child. Supply together with model; omit both to use configured child defaults or this provider's route defaults." : "LLM provider route for the child. Supply together with model; omit both to use configured child defaults or inherit the parent route."
 							},
-							jobId: {
-								type: "string",
-								required: true
-							}
-						}
-					},
-					{
-						type: "object",
-						additionalProperties: false,
-						properties: {
-							kind: {
+							model: {
 								type: "string",
-								required: true,
-								const: "continuable"
+								description: providerRouteDefaults !== void 0 ? "Model id interpreted by provider. Supply together with provider; omit both to use configured child defaults or this provider's route defaults." : "Model id interpreted by provider. Supply together with provider; omit both to use configured child defaults or inherit the parent route."
 							},
-							subagentId: {
+							reasoning_effort: {
 								type: "string",
-								required: true
+								description: providerRouteDefaults !== void 0 ? "Adapter-owned reasoning effort for the effective child route. Omit to use a compatible configured effort or the selected model's default." : "Adapter-owned reasoning effort for the effective child route. Omit to inherit a compatible configured/parent effort or use a newly selected model's default."
 							}
-						}
+						} : {},
+						...backgroundEnabled ? { run_in_background: {
+							type: "boolean",
+							description: continuable ? "Whether to run in the background and return a durable subagent id immediately. Defaults to true. Set false to wait for the result when your next action depends on it." : "Whether to run as a background job and return its id. Defaults to false; collect with job_output or stop with job_kill."
+						} } : {}
 					},
-					{
-						type: "object",
-						additionalProperties: false,
-						properties: {
-							kind: {
-								type: "string",
-								required: true,
-								const: "foreground"
+					output: {
+						schema: { oneOf: [
+							{
+								type: "object",
+								additionalProperties: false,
+								properties: {
+									kind: {
+										type: "string",
+										required: true,
+										const: "background"
+									},
+									jobId: {
+										type: "string",
+										required: true
+									}
+								}
 							},
-							runId: {
-								type: "string",
-								required: true
+							{
+								type: "object",
+								additionalProperties: false,
+								properties: {
+									kind: {
+										type: "string",
+										required: true,
+										const: "continuable"
+									},
+									subagentId: {
+										type: "string",
+										required: true
+									}
+								}
 							},
-							output: {
-								type: "array",
-								required: true,
-								items: { type: "json" }
+							{
+								type: "object",
+								additionalProperties: false,
+								properties: {
+									kind: {
+										type: "string",
+										required: true,
+										const: "foreground"
+									},
+									runId: {
+										type: "string",
+										required: true
+									},
+									output: {
+										type: "array",
+										required: true,
+										items: { type: "json" }
+									}
+								}
 							}
+						] },
+						render: (_args, value) => [{
+							type: "text",
+							text: value.kind === "background" ? `started background subagent job ${value.jobId}` : value.kind === "continuable" ? `started subagent ${value.subagentId}` : outputValueText(value.output)
+						}]
+					},
+					isConcurrencySafe: () => true,
+					async execute(args, exec) {
+						const parent = exec.agent;
+						if (!parent) throw new Error("subagent tool requires a calling agent (exec.agent was undefined)");
+						const modelRequest = args;
+						const parentOptions = parentAgentOptionsForDelegation(parent);
+						const requiresRoutePreflight = hasDelegationModelRequest(modelRequest) || hasConfiguredLlmSelection(config.agentOptions);
+						const requestedChildAgentOptions = requestedAgentOptions(parentOptions, requiresRoutePreflight && providerRouteDefaults !== void 0 ? {
+							...providerRouteDefaults,
+							...config.agentOptions
+						} : config.agentOptions, modelRequest, modelSelectionEnabled);
+						assertAllowedModelSelection(modelSelectionPolicy, parentOptions, requestedChildAgentOptions, modelRequest);
+						if (requiresRoutePreflight) {
+							const llm = runtimeCtx.get("llm");
+							if (llm === void 0) throw new Error("cannot resolve the selected child LLM route because the `llm` service is unavailable");
+							await preflightChildLlmRoute(llm, parentOptions, requestedChildAgentOptions, exec.signal, providerRouteDefaults === void 0);
+							if (runtimeCtx.subagents.getProvider(config.provider) !== subagentProvider) throw new Error(`subagent provider "${config.provider}" changed while resolving the child LLM route; retry the delegation`);
 						}
-					}
-				] },
-				render: (_args, value) => [{
-					type: "text",
-					text: value.kind === "background" ? `started background subagent job ${value.jobId}` : value.kind === "continuable" ? `started subagent ${value.subagentId}` : outputValueText(value.output)
-				}]
-			},
-			isConcurrencySafe: () => true,
-			async execute(args, exec) {
-				const parent = exec.agent;
-				if (!parent) throw new Error("subagent tool requires a calling agent (exec.agent was undefined)");
-				const maxDepth = typeof config.maxDepth === "number" ? config.maxDepth : void 0;
-				const request = {
-					label: args.description,
-					prompt: [{
-						type: "text",
-						text: args.prompt
-					}],
-					parent,
-					...config.agentOptions !== void 0 ? { agentOptions: config.agentOptions } : {},
-					...config.persona !== void 0 ? { persona: config.persona } : {},
-					...config.toolFilter !== void 0 ? { toolFilter: config.toolFilter } : {},
-					...maxDepth !== void 0 ? { maxDepth } : {}
-				};
-				if (resolveDelegationRun(args, {
-					backgroundEnabled,
-					continuable
-				}).runInBackground) {
-					if (continuable) return {
-						kind: "continuable",
-						subagentId: (await ctx.subagents.startContinuable({
-							provider: config.provider,
+						exec.signal.throwIfAborted();
+						const maxDepth = typeof config.maxDepth === "number" ? config.maxDepth : void 0;
+						const request = {
 							label: args.description,
-							request,
+							prompt: [{
+								type: "text",
+								text: args.prompt
+							}],
+							parent,
+							...requestedChildAgentOptions !== void 0 ? { agentOptions: requestedChildAgentOptions } : {},
+							...config.persona !== void 0 ? { persona: config.persona } : {},
+							...config.toolFilter !== void 0 ? { toolFilter: config.toolFilter } : {},
+							...maxDepth !== void 0 ? { maxDepth } : {}
+						};
+						if (resolveDelegationRun(args, {
+							backgroundEnabled,
+							continuable
+						}).runInBackground) {
+							if (continuable) return {
+								kind: "continuable",
+								subagentId: (await runtimeCtx.subagents.startContinuable({
+									provider: config.provider,
+									label: args.description,
+									request,
+									signal: exec.signal
+								})).childId
+							};
+							const jobs = runtimeCtx.get("jobs");
+							if (jobs === void 0) throw new Error("background jobs unavailable: load @deepseek-ai/dsh-jobs and @deepseek-ai/dsh-tool-jobs");
+							return {
+								kind: "background",
+								jobId: jobs.start({
+									kind: "subagent",
+									label: args.description,
+									owner: parent,
+									run: () => {
+										const controller = new AbortController();
+										return {
+											cancel: (reason) => {
+												controller.abort(reason ?? "background subagent task killed");
+											},
+											done: settleStart(runtimeCtx.subagents.start(config.provider, {
+												...request,
+												signal: controller.signal
+											}), controller.signal)
+										};
+									}
+								})
+							};
+						}
+						return settleForegroundRun(await runtimeCtx.subagents.start(config.provider, {
+							...request,
 							signal: exec.signal
-						})).childId
-					};
-					const jobs = ctx.get("jobs");
-					if (jobs === void 0) throw new Error("background jobs unavailable: load @deepseek-ai/dsh-jobs and @deepseek-ai/dsh-tool-jobs");
-					return {
-						kind: "background",
-						jobId: jobs.start({
-							kind: "subagent",
-							label: args.description,
-							owner: parent,
-							run: () => {
-								const controller = new AbortController();
-								return {
-									cancel: (reason) => {
-										controller.abort(reason ?? "background subagent task killed");
-									},
-									done: settleStart(ctx.subagents.start(config.provider, {
-										...request,
-										signal: controller.signal
-									}), controller.signal)
-								};
-							}
-						})
-					};
-				}
-				return settleForegroundRun(await ctx.subagents.start(config.provider, {
-					...request,
-					signal: exec.signal
-				}));
+						}));
+					}
+				}))
+			};
+		};
+		runtimeCtx.on("subagent/provider-added", (subagentProvider) => {
+			if (subagentProvider.name === config.provider && mounted === void 0) mount(subagentProvider);
+		});
+		runtimeCtx.on("subagent/provider-removed", (name) => {
+			if (name !== config.provider || mounted === void 0) return;
+			mounted.disposeTool();
+			mounted = void 0;
+		});
+		const present = runtimeCtx.subagents.getProvider(config.provider);
+		if (present !== void 0) mount(present);
+		else runtimeCtx.logger.info(`subagent provider "${config.provider}" not registered yet; the "${config.toolName ?? "subagent"}" tool will register when it appears`);
+		if (backgroundEnabled && continuable) runtimeCtx.systemPrompt.section({
+			name: `tool:${toolName}`,
+			order: runtimeCtx.systemPrompt.getSectionOrder("TOOL_SUBAGENT"),
+			text: (context) => mounted === void 0 || runtimeCtx.tools.get(toolName, context.scope) === void 0 ? "" : `Use ${toolName} in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set \`run_in_background: false\` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.`
+		});
+	};
+	if (config.modelSelectionSettings !== true) {
+		install(ctx, void 0);
+		return;
+	}
+	const settings = ctx.get("subagentModelSelection");
+	if (settings === void 0) throw new Error("tool-subagent: `modelSelectionSettings` requires @deepseek-ai/dsh-tool-subagent/model-selection-settings in the Host scope");
+	const compositionScope = scopeOf(ctx);
+	if (compositionScope === void 0) throw new Error("tool-subagent: `modelSelectionSettings` requires an Agent or preset scope");
+	const selectForAgent = (agent) => {
+		let allowedModels = subagentModelSelectionPolicy(ctx.sessionProjections, agent.session);
+		if (allowedModels === void 0) {
+			const parentId = agent.session.header.origin === "subagent" ? agent.session.header.parentSession : void 0;
+			if (parentId !== void 0) {
+				const parent = ctx.get("agents")?.get(parentId);
+				allowedModels = parent === void 0 ? void 0 : subagentModelSelectionPolicy(ctx.sessionProjections, parent.session);
+			} else if (agent.session.firstLiveSeq === 0) {
+				const current = settings.current();
+				allowedModels = current.enabled ? current.allowedModels : void 0;
 			}
-		}));
+		}
+		if (allowedModels !== void 0) recordSubagentModelSelection(ctx.sessionProjections, agent.session, allowedModels);
+		return allowedModels === void 0 ? void 0 : { routes: allowedModels };
 	};
-	ctx.on("subagent/provider-added", (provider) => {
-		if (provider.name === config.provider && disposeTool === void 0) mount(provider);
-	});
-	ctx.on("subagent/provider-removed", (name) => {
-		if (name !== config.provider || disposeTool === void 0) return;
-		disposeTool();
-		disposeTool = void 0;
+	const agent = ctx.agent;
+	if (agent !== void 0) {
+		install(ctx, selectForAgent(agent));
+		return;
+	}
+	const agents = ctx.get("agents");
+	/* v8 ignore next -- Agent and preset scopes are minted only by the Agent registry. */
+	if (agents === void 0) throw new Error("tool-subagent: scoped model-selection settings require the Agent registry");
+	const scopedInstalls = /* @__PURE__ */ new WeakMap();
+	const installing = /* @__PURE__ */ new WeakSet();
+	const belongsToComposition = (candidate) => scopeChainOf(scopeOf(candidate.ctx)).includes(compositionScope);
+	const installScoped = (candidate) => {
+		if (scopedInstalls.has(candidate) || installing.has(candidate)) return;
+		installing.add(candidate);
+		let fiber;
+		try {
+			const policy = selectForAgent(candidate);
+			fiber = candidate.ctx.inject([
+				"tools",
+				"subagents",
+				"systemPrompt"
+			], (runtimeCtx) => {
+				install(runtimeCtx, policy);
+			});
+		} finally {
+			installing.delete(candidate);
+		}
+		scopedInstalls.set(candidate, fiber);
+	};
+	const removeScoped = (candidate) => {
+		const fiber = scopedInstalls.get(candidate);
+		if (fiber === void 0) return;
+		scopedInstalls.delete(candidate);
+		/* v8 ignore next 3 -- Cordis Fiber disposal contains registration cleanup failures; this is the final diagnostic sink. */
+		fiber.dispose().catch((error) => {
+			ctx.logger.warn(`tool-subagent: failed to remove recomposed Agent "${candidate.id}" definitions: ${String(error)}`);
+		});
+	};
+	const reconcileComposedAgents = () => {
+		for (const candidate of agents.list()) if (belongsToComposition(candidate)) installScoped(candidate);
+		else removeScoped(candidate);
+	};
+	ctx.on("agent/created", ({ agent: created }) => {
+		installScoped(created);
 	});
-	const present = ctx.subagents.getProvider(config.provider);
-	if (present !== void 0) mount(present);
-	else ctx.logger.info(`subagent provider "${config.provider}" not registered yet; the "${config.toolName ?? "subagent"}" tool will register when it appears`);
-	if (backgroundEnabled && continuable) ctx.systemPrompt.section({
-		name: `tool:${toolName}`,
-		order: SUBAGENT_SECTION_ORDER,
-		text: (context) => disposeTool === void 0 || ctx.tools.get(toolName, context.scope) === void 0 ? "" : `Use ${toolName} in the background by default. Start independent delegations together in one assistant message and continue useful work while they run. Set \`run_in_background: false\` only when your next action depends on that subagent's result. When a background run settles, the runtime sends you a notice containing its outcome and any final assistant message.`
+	ctx.on("agent/disposed", ({ agent: disposed }) => {
+		removeScoped(disposed);
 	});
+	ctx.on("tools/change", reconcileComposedAgents);
 }
 //#endregion
 export { Config, apply, inject, name };