@kingdee-kcode/coding-agent 16.1.9

package/src/task/index.ts

@@ -0,0 +1,1580 @@
+/**
+ * Task tool - Delegate tasks to specialized agents.
+ *
+ * Discovers agent definitions from:
+ *   - Bundled agents (shipped with omp-coding-agent)
+ *   - ~/.kcode/agent/agents/*.md (user-level)
+ *   - .kcode/agents/*.md (project-level)
+ *
+ * Supports:
+ *   - Single agent spawn per call (parallelism = parallel task calls)
+ *   - Batch spawning + shared context per call when `task.batch` is enabled
+ *   - Background execution through AsyncJobManager when `async.enabled` is enabled
+ *   - Progress tracking via JSON events
+ *   - Session artifacts for debugging
+ */
+import * as fs from "node:fs/promises";
+import * as os from "node:os";
+import path from "node:path";
+import type { AgentTool, AgentToolResult, AgentToolUpdateCallback } from "@oh-my-pi/pi-agent-core";
+import type { Usage } from "@oh-my-pi/pi-ai";
+import { $env, logger, prompt, Snowflake } from "@oh-my-pi/pi-utils";
+import type { ToolSession } from "..";
+import { resolveAgentModelPatterns } from "../config/model-resolver";
+import { MCPManager } from "../mcp/manager";
+import type { Theme } from "../modes/theme/theme";
+import planModeSubagentPrompt from "../prompts/system/plan-mode-subagent.md" with { type: "text" };
+import subagentUserPromptTemplate from "../prompts/system/subagent-user-prompt.md" with { type: "text" };
+import taskDescriptionTemplate from "../prompts/tools/task.md" with { type: "text" };
+import taskSummaryTemplate from "../prompts/tools/task-summary.md" with { type: "text" };
+import { truncateForPrompt } from "../tools/approval";
+import { isIrcEnabled } from "../tools/irc";
+import { formatBytes, formatDuration } from "../tools/render-utils";
+import {
+	type AgentDefinition,
+	type AgentProgress,
+	canSpawnAtDepth,
+	getTaskSchema,
+	type SingleResult,
+	type TaskItem,
+	type TaskParams,
+	type TaskToolDetails,
+	type TaskToolSchemaInstance,
+} from "./types";
+// Import review tools for side effects (registers subagent tool handlers)
+import "../tools/review";
+import type { AsyncJobManager } from "../async";
+import type { LocalProtocolOptions } from "../internal-urls";
+import { loadOverallPlanReference } from "../plan-mode/plan-handoff";
+import { AgentRegistry, MAIN_AGENT_ID } from "../registry/agent-registry";
+import { generateCommitMessage } from "../utils/commit-message-generator";
+import * as git from "../utils/git";
+import { type DiscoveryResult, discoverAgents, getAgent } from "./discovery";
+import { runSubprocess } from "./executor";
+import { generateTaskName } from "./name-generator";
+import { AgentOutputManager } from "./output-manager";
+import { mapWithConcurrencyLimit, Semaphore } from "./parallel";
+import { renderResult, renderCall as renderTaskCall } from "./render";
+import { repairTaskParams } from "./repair-args";
+import {
+	applyNestedPatches,
+	captureBaseline,
+	captureDeltaPatch,
+	cleanupIsolation,
+	cleanupTaskBranches,
+	commitToBranch,
+	ensureIsolation,
+	getRepoRoot,
+	type IsolationHandle,
+	mergeTaskBranches,
+	parseIsolationMode,
+	type WorktreeBaseline,
+} from "./worktree";
+
+function renderSubagentUserPrompt(assignment: string): string {
+	return prompt.render(subagentUserPromptTemplate, {
+		assignment: assignment.trim(),
+	});
+}
+
+function createUsageTotals(): Usage {
+	return {
+		input: 0,
+		output: 0,
+		cacheRead: 0,
+		cacheWrite: 0,
+		totalTokens: 0,
+		cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0, total: 0 },
+	};
+}
+
+function addUsageTotals(target: Usage, usage: Partial<Usage>): void {
+	const input = usage.input ?? 0;
+	const output = usage.output ?? 0;
+	const cacheRead = usage.cacheRead ?? 0;
+	const cacheWrite = usage.cacheWrite ?? 0;
+	const totalTokens = usage.totalTokens ?? input + output + cacheRead + cacheWrite;
+	const cost =
+		usage.cost ??
+		({
+			input: 0,
+			output: 0,
+			cacheRead: 0,
+			cacheWrite: 0,
+			total: 0,
+		} satisfies Usage["cost"]);
+
+	target.input += input;
+	target.output += output;
+	target.cacheRead += cacheRead;
+	target.cacheWrite += cacheWrite;
+	target.totalTokens += totalTokens;
+	target.cost.input += cost.input;
+	target.cost.output += cost.output;
+	target.cost.cacheRead += cost.cacheRead;
+	target.cost.cacheWrite += cost.cacheWrite;
+	target.cost.total += cost.total;
+}
+
+// Re-export types and utilities
+export { loadBundledAgents as BUNDLED_AGENTS } from "./agents";
+export { discoverCommands, expandCommand, getCommand } from "./commands";
+export { discoverAgents, getAgent } from "./discovery";
+export { AgentOutputManager } from "./output-manager";
+export type {
+	AgentDefinition,
+	AgentProgress,
+	SingleResult,
+	SubagentEventPayload,
+	SubagentLifecyclePayload,
+	SubagentProgressPayload,
+	TaskParams,
+	TaskToolDetails,
+} from "./types";
+export {
+	TASK_SUBAGENT_EVENT_CHANNEL,
+	TASK_SUBAGENT_LIFECYCLE_CHANNEL,
+	TASK_SUBAGENT_PROGRESS_CHANNEL,
+	taskSchema,
+} from "./types";
+
+// Built-in tools whose approval tier is "read" (see tool classes' `approval`).
+// An agent is read-only iff its declared tools are a non-empty subset of this set.
+// Fail-safe: any unknown tool makes the agent not read-only.
+export const READ_ONLY_TOOL_NAMES: ReadonlySet<string> = new Set([
+	"read",
+	"search",
+	"find",
+	"web_search",
+	"ast_grep",
+	"yield",
+	"irc",
+	"ask",
+	"job",
+	"todo",
+	"recall",
+	"reflect",
+	"retain",
+	"memory_edit",
+	"inspect_image",
+	"checkpoint",
+	"rewind",
+	"resolve",
+	"report_finding",
+	"search_tool_bm25",
+]);
+
+const PLAN_MODE_AGENT_TOOL_ALLOWLIST: ReadonlySet<string> = new Set(["ast_grep", "report_finding"]);
+
+export function isReadOnlyAgent(agent: AgentDefinition): boolean {
+	return !!agent.tools?.length && agent.tools.every(tool => READ_ONLY_TOOL_NAMES.has(tool));
+}
+
+/**
+ * Preview text for a child result. Falls back to "(no output)" — annotated
+ * with the request count when the child actually did work, so the parent can
+ * tell a no-op child from one that burned requests before being cancelled.
+ */
+export function formatResultOutputFallback(result: Pick<SingleResult, "output" | "stderr" | "requests">): string {
+	const base = result.output.trim() || result.stderr.trim();
+	if (base) return base;
+	return result.requests > 0 ? `(no output) after ${result.requests} req` : "(no output)";
+}
+
+/**
+ * Render the tool description from a cached agent list and current settings.
+ */
+function renderDescription(
+	agents: AgentDefinition[],
+	maxConcurrency: number,
+	isolationEnabled: boolean,
+	disabledAgents: string[],
+	batchEnabled: boolean,
+	asyncEnabled: boolean,
+	ircEnabled: boolean,
+	parentSpawns: string,
+): string {
+	const spawningDisabled = parentSpawns === "";
+	let filteredAgents = disabledAgents.length > 0 ? agents.filter(a => !disabledAgents.includes(a.name)) : agents;
+	if (spawningDisabled) {
+		filteredAgents = [];
+	} else if (parentSpawns !== "*") {
+		const allowed = new Set(
+			parentSpawns
+				.split(",")
+				.map(s => s.trim())
+				.filter(Boolean),
+		);
+		filteredAgents = filteredAgents.filter(a => allowed.has(a.name));
+	}
+	const renderedAgents = filteredAgents.map(agent => ({
+		name: agent.name,
+		description: agent.description,
+		readOnly: isReadOnlyAgent(agent),
+	}));
+	return prompt.render(taskDescriptionTemplate, {
+		agents: renderedAgents,
+		spawningDisabled,
+		MAX_CONCURRENCY: maxConcurrency,
+		isolationEnabled,
+		batchEnabled,
+		asyncEnabled,
+		ircEnabled,
+	});
+}
+
+function createTaskModeError(text: string): AgentToolResult<TaskToolDetails> {
+	return {
+		content: [{ type: "text", text }],
+		details: { projectAgentsDir: null, results: [], totalDurationMs: 0 },
+	};
+}
+
+/**
+ * Reject fields the current configuration does not accept. `schema` is never
+ * accepted (structured output comes from the agent definition's `output`
+ * frontmatter, the inherited session schema, or an eval-workflow
+ * `agent(..., schema)` call); `tasks`/`context` require `task.batch`.
+ */
+function validateShapeParams(batchEnabled: boolean, params: TaskParams): string | undefined {
+	if ((params as Record<string, unknown>).schema !== undefined) {
+		return "The task tool does not accept `schema`. Rely on the selected agent definition's `output` schema or the inherited session schema; workflows needing ad-hoc structured output use eval `agent(prompt, schema)`.";
+	}
+	if (!batchEnabled) {
+		const disallowed = (["tasks", "context"] as const).filter(field => params[field] !== undefined);
+		if (disallowed.length > 0) {
+			return `task.batch is disabled, so the task tool does not accept ${disallowed.map(f => `\`${f}\``).join(" or ")}. Spawn one agent per call with \`assignment\`, or enable the task.batch setting.`;
+		}
+	}
+	return undefined;
+}
+
+/**
+ * Validate the spawn parameter contract against the wire shapes. `agent` is
+ * always required. With `task.batch` the model-facing shape is
+ * `{ agent, context, tasks[] }` — `tasks` non-empty with per-item assignments
+ * and unique ids, `context` non-empty, no top-level `assignment` alongside.
+ * The flat `{ agent, ...item }` form stays accepted at runtime under either
+ * setting (internal callers, stale transcripts). Returns a problem
+ * description, or undefined when valid.
+ */
+function validateSpawnParams(params: TaskParams, batchEnabled: boolean): string | undefined {
+	const agent = typeof params.agent === "string" ? params.agent.trim() : "";
+	if (!agent) {
+		return "Missing `agent`. Provide an agent type to spawn.";
+	}
+	const hasAssignment = typeof params.assignment === "string" && params.assignment.trim() !== "";
+	const tasks = params.tasks;
+	if (batchEnabled && tasks !== undefined) {
+		if (!Array.isArray(tasks) || tasks.length === 0) {
+			return "Missing `tasks`. Provide at least one task item ({ id?, description?, assignment }).";
+		}
+		if (hasAssignment) {
+			return "Top-level `assignment` is not part of the batch shape. Put the work in `tasks[]` items.";
+		}
+		for (let i = 0; i < tasks.length; i++) {
+			const item = tasks[i];
+			if (!item || typeof item.assignment !== "string" || item.assignment.trim() === "") {
+				return `Task ${i + 1}${item?.id ? ` (\`${item.id}\`)` : ""} is missing \`assignment\`. Every task needs complete, self-contained instructions.`;
+			}
+		}
+		const seen = new Map<string, string>();
+		for (const item of tasks) {
+			const id = item.id?.trim();
+			if (!id) continue;
+			const key = id.toLowerCase();
+			const existing = seen.get(key);
+			if (existing !== undefined) {
+				return `Duplicate task id ${existing === id ? `\`${id}\`` : `\`${existing}\` / \`${id}\``}. Provided ids must be unique within a call (case-insensitive).`;
+			}
+			seen.set(key, id);
+		}
+		if (typeof params.context !== "string" || params.context.trim() === "") {
+			return "Missing `context`. Provide the shared background for this batch — goal, constraints, and any contract the tasks share.";
+		}
+		return undefined;
+	}
+	if (!hasAssignment) {
+		return batchEnabled
+			? "Missing `tasks`. Provide a `tasks` array (one subagent per item) with a shared `context`."
+			: "Missing `assignment`. Provide complete, self-contained instructions for the agent.";
+	}
+	return undefined;
+}
+
+/**
+ * Normalize a validated call into its spawn list: the `tasks[]` batch when
+ * provided, otherwise the single top-level spawn.
+ */
+function resolveSpawnItems(params: TaskParams): TaskItem[] {
+	if (Array.isArray(params.tasks) && params.tasks.length > 0) {
+		return params.tasks;
+	}
+	return [{ id: params.id, description: params.description, role: params.role, assignment: params.assignment }];
+}
+
+/**
+ * Per-spawn params handed to the executor path: top-level call fields with the
+ * item's identity substituted in. `tasks` never leaks into a spawn; the shared
+ * `context` rides along unchanged. Keys are only materialized when present —
+ * `#runSpawn` distinguishes an absent `isolated` from an explicit one. The
+ * item's `isolated` (batch form) wins over the top-level flag (flat form).
+ */
+function spawnParamsFor(params: TaskParams, item: TaskItem): TaskParams {
+	const spawn: TaskParams = { agent: params.agent };
+	if (item.id !== undefined) spawn.id = item.id;
+	if (item.description !== undefined) spawn.description = item.description;
+	if (item.role !== undefined) spawn.role = item.role;
+	if (item.assignment !== undefined) spawn.assignment = item.assignment;
+	if (params.context !== undefined) spawn.context = params.context;
+	if (item.isolated !== undefined) {
+		spawn.isolated = item.isolated;
+	} else if ("isolated" in params) {
+		spawn.isolated = params.isolated;
+	}
+	return spawn;
+}
+
+/** Generic worker agents whose output sharpens with a tailored `role` rather than the bare type. */
+const GENERIC_SPAWN_AGENTS: ReadonlySet<string> = new Set(["task", "quick_task"]);
+
+/**
+ * Advisory — never a rejection — nudging the spawner toward tailored
+ * specialists when it spawns generic role-less workers and still holds spawn
+ * capacity (DepthCapacity: it currently has the `task` tool). Fires when a
+ * generic `task`/`quick_task` spawn carries no `role`, or when one call clones
+ * the same agent ≥2× all without roles. Returns undefined when no nudge applies.
+ */
+export function buildSpecializationAdvisory(
+	agentName: string | undefined,
+	items: TaskItem[],
+	depthCapacity: boolean,
+): string | undefined {
+	if (!depthCapacity) return undefined;
+	const rolelessCount = items.filter(item => !item.role?.trim()).length;
+	if (rolelessCount === 0) return undefined;
+	const generic = agentName !== undefined && GENERIC_SPAWN_AGENTS.has(agentName);
+	const cloned = items.length >= 2 && rolelessCount === items.length;
+	if (!generic && !cloned) return undefined;
+	const label = agentName ?? "task";
+	return (
+		`Tip: spawned ${rolelessCount} \`${label}\` worker${rolelessCount === 1 ? "" : "s"} without a \`role\`. ` +
+		`Tailored specialists outperform generic workers — give each spawn a \`role\` naming its expertise ` +
+		`(e.g. "Auth-flow security reviewer"). Depth budget remains, so decompose into named specialists ` +
+		`rather than cloning one generic worker.`
+	);
+}
+
+/**
+ * Suggestion — never a rejection — nudging the spawner to coordinate via `irc`
+ * when one call creates ≥2 live siblings and it still holds spawn capacity.
+ * Returns undefined when there is nothing to coordinate or IRC is unavailable.
+ */
+export function buildCoordinationAdvisory(
+	items: TaskItem[],
+	depthCapacity: boolean,
+	ircEnabled: boolean,
+): string | undefined {
+	if (!depthCapacity || !ircEnabled || items.length < 2) return undefined;
+	return (
+		`Coordinate: ${items.length} siblings are running together. If their work overlaps, have them ` +
+		`message each other via \`irc\` (by id, or "all" to broadcast) before editing shared files — ` +
+		`live coordination beats a serial handoff. Check \`irc\` op:"list" to see who is doing what.`
+	);
+}
+
+/**
+ * Compose the non-blocking advisory appended to a `task` result: the
+ * specialization nudge, plus — only when the siblings keep running after this
+ * call (`willRunAsync`) — the coordination suggestion. Coordination is gated on
+ * async because a sync fanout's siblings have already finished, so a
+ * "coordinate while they run" hint would misfire. Returns undefined when
+ * neither applies.
+ */
+export function composeSpawnAdvisory(args: {
+	agentName: string | undefined;
+	items: TaskItem[];
+	depthCapacity: boolean;
+	ircEnabled: boolean;
+	willRunAsync: boolean;
+}): string | undefined {
+	return (
+		[
+			buildSpecializationAdvisory(args.agentName, args.items, args.depthCapacity),
+			args.willRunAsync ? buildCoordinationAdvisory(args.items, args.depthCapacity, args.ircEnabled) : undefined,
+		]
+			.filter(Boolean)
+			.join("\n\n") || undefined
+	);
+}
+
+/** Sentinel for async jobs whose subagent finished with a failing result; progress is already updated. */
+class TaskJobError extends Error {}
+
+/**
+ * Process-level memo for create-time agent discovery, keyed by resolved cwd.
+ *
+ * `TaskTool.create` runs for every (sub)agent session in this process and the
+ * walk-up + plugin-registry scan in `discoverAgents` is identical for a given
+ * cwd, so repeat creations reuse the first scan. Execution-time discovery
+ * (`#runSpawn`) intentionally stays fresh. The memo also tracks the live
+ * `discoverAgents` binding: test spies swap that binding, which invalidates
+ * the memo automatically.
+ */
+const discoveryMemo = new Map<string, Promise<DiscoveryResult>>();
+let discoveryMemoFn: typeof discoverAgents | undefined;
+
+function discoverAgentsForCreate(cwd: string): Promise<DiscoveryResult> {
+	const fn = discoverAgents;
+	if (discoveryMemoFn !== fn) {
+		discoveryMemoFn = fn;
+		discoveryMemo.clear();
+	}
+	const key = path.resolve(cwd);
+	let pending = discoveryMemo.get(key);
+	if (!pending) {
+		pending = fn(cwd);
+		discoveryMemo.set(key, pending);
+		pending.catch(() => {
+			if (discoveryMemo.get(key) === pending) discoveryMemo.delete(key);
+		});
+	}
+	return pending;
+}
+
+// ═══════════════════════════════════════════════════════════════════════════
+// Tool Class
+// ═══════════════════════════════════════════════════════════════════════════
+
+/**
+ * Task tool - Delegate tasks to specialized agents.
+ *
+ * Each call spawns one subagent — or, with `task.batch`, one per `tasks[]`
+ * item. When `async.enabled` is on, spawns run as AsyncJobManager jobs; when
+ * disabled, the tool blocks until every spawn finishes.
+ */
+export class TaskTool implements AgentTool<TaskToolSchemaInstance, TaskToolDetails, Theme> {
+	readonly name = "task";
+	readonly approval = "exec" as const;
+	readonly formatApprovalDetails = (args: unknown): string[] => {
+		const params = args as Partial<TaskParams>;
+		const lines: string[] = [];
+		if (typeof params.agent === "string") {
+			lines.push(`Agent: ${truncateForPrompt(params.agent)}`);
+		}
+		if (typeof params.role === "string" && params.role.trim()) {
+			lines.push(`Role: ${truncateForPrompt(params.role)}`);
+		}
+		if (typeof params.id === "string" && params.id.trim()) {
+			lines.push(`Task: ${truncateForPrompt(params.id)}`);
+		}
+		if (typeof params.assignment === "string") {
+			lines.push(`Assignment:\n${truncateForPrompt(params.assignment)}`);
+		}
+		if (typeof params.context === "string" && params.context.trim()) {
+			lines.push(`Context:\n${truncateForPrompt(params.context)}`);
+		}
+		const tasks = Array.isArray(params.tasks) ? params.tasks : [];
+		const firstTask = tasks[0];
+		if (firstTask) {
+			if (typeof firstTask.id === "string" && firstTask.id.trim()) {
+				lines.push(`Task: ${truncateForPrompt(firstTask.id)}`);
+			}
+			if (typeof firstTask.role === "string" && firstTask.role.trim()) {
+				lines.push(`Role: ${truncateForPrompt(firstTask.role)}`);
+			}
+			if (typeof firstTask.assignment === "string") {
+				lines.push(`Assignment:\n${truncateForPrompt(firstTask.assignment)}`);
+			}
+			if (tasks.length > 1) {
+				lines.push(`+${tasks.length - 1} more task${tasks.length === 2 ? "" : "s"}`);
+			}
+		}
+		return lines;
+	};
+	readonly label = "Task";
+	readonly summary = "Spawn subagents to complete delegated tasks";
+	readonly strict = true;
+	readonly loadMode = "discoverable";
+	readonly renderResult = renderResult;
+	// Suppress the streaming call preview once a (partial or final) result exists
+	// so the task renders as ONE block that transitions in place — not a pending
+	// call frame stacked above the result frame. Mirrors `taskToolRenderer`.
+	readonly mergeCallAndResult = true;
+	readonly #discoveredAgents: AgentDefinition[];
+	readonly #blockedAgent: string | undefined;
+	/**
+	 * One semaphore per TaskTool instance (i.e. per session): bounds concurrent
+	 * subagents across parallel `task` calls within the session. Sized from
+	 * `task.maxConcurrency` at first use; later setting changes do not resize it.
+	 */
+	#spawnSemaphore: Semaphore | undefined;
+
+	get parameters(): TaskToolSchemaInstance {
+		const isolationEnabled = this.session.settings.get("task.isolation.mode") !== "none";
+		return getTaskSchema({ isolationEnabled, batchEnabled: this.#isBatchEnabled() });
+	}
+
+	renderCall(args: unknown, options: Parameters<typeof renderTaskCall>[1], theme: Theme) {
+		return renderTaskCall(repairTaskParams(args as TaskParams), options, theme);
+	}
+
+	/** Dynamic description that reflects current disabled-agent settings */
+	get description(): string {
+		const disabledAgents = this.session.settings.get("task.disabledAgents") as string[];
+		const maxConcurrency = this.session.settings.get("task.maxConcurrency");
+		const isolationMode = this.session.settings.get("task.isolation.mode");
+		return renderDescription(
+			this.#discoveredAgents,
+			maxConcurrency,
+			isolationMode !== "none",
+			disabledAgents,
+			this.#isBatchEnabled(),
+			this.session.settings.get("async.enabled"),
+			isIrcEnabled(this.session.settings, this.session.taskDepth ?? 0),
+			this.session.getSessionSpawns() ?? "*",
+		);
+	}
+	private constructor(
+		private readonly session: ToolSession,
+		discoveredAgents: AgentDefinition[],
+	) {
+		this.#blockedAgent = $env.PI_BLOCKED_AGENT;
+		this.#discoveredAgents = discoveredAgents;
+	}
+
+	#isBatchEnabled(): boolean {
+		return this.session.settings.get("task.batch");
+	}
+
+	#getSpawnSemaphore(): Semaphore {
+		this.#spawnSemaphore ??= new Semaphore(this.session.settings.get("task.maxConcurrency"));
+		return this.#spawnSemaphore;
+	}
+
+	/**
+	 * Create a TaskTool instance with async agent discovery.
+	 */
+	static async create(session: ToolSession): Promise<TaskTool> {
+		const { agents } = await discoverAgentsForCreate(session.cwd);
+		return new TaskTool(session, agents);
+	}
+
+	async execute(
+		toolCallId: string,
+		rawParams: unknown,
+		signal?: AbortSignal,
+		onUpdate?: AgentToolUpdateCallback<TaskToolDetails>,
+	): Promise<AgentToolResult<TaskToolDetails>> {
+		const params = repairTaskParams(rawParams as TaskParams);
+		const batchEnabled = this.#isBatchEnabled();
+		const validationError = validateShapeParams(batchEnabled, params) ?? validateSpawnParams(params, batchEnabled);
+		if (validationError) {
+			return createTaskModeError(validationError);
+		}
+
+		const spawnItems = resolveSpawnItems(params);
+		const selectedAgent = this.#discoveredAgents.find(agent => agent.name === params.agent);
+		const asyncEnabled = this.session.settings.get("async.enabled");
+		const manager = asyncEnabled ? this.session.asyncJobManager : undefined;
+		const depthCapacity = canSpawnAtDepth(
+			this.session.settings.get("task.maxRecursionDepth") ?? 2,
+			this.session.taskDepth ?? 0,
+		);
+		const ircEnabled = isIrcEnabled(this.session.settings, this.session.taskDepth ?? 0);
+		// Coordination only makes sense when the siblings keep running after this
+		// call returns (async). In the sync fallback they have already completed,
+		// so a "coordinate while they run" hint would misfire.
+		const willRunAsync = !!manager && selectedAgent?.blocking !== true;
+		const advisory = this.session.suppressSpawnAdvisory
+			? undefined
+			: composeSpawnAdvisory({
+					agentName: params.agent,
+					items: spawnItems,
+					depthCapacity,
+					ircEnabled,
+					willRunAsync,
+				});
+		// Returns a fresh result (copied content array, copied text part) rather
+		// than mutating the caller's — task results are short-lived here, but an
+		// in-place edit on a shared/cached AgentToolResult would be a hidden trap.
+		const withAdvisory = (result: AgentToolResult<TaskToolDetails>): AgentToolResult<TaskToolDetails> => {
+			if (!advisory) return result;
+			let appended = false;
+			const content = result.content.map(part => {
+				if (!appended && part.type === "text" && typeof part.text === "string") {
+					appended = true;
+					return { ...part, text: `${part.text}\n\n${advisory}` };
+				}
+				return part;
+			});
+			if (!appended) content.push({ type: "text", text: advisory });
+			return { ...result, content };
+		};
+		if (!asyncEnabled || !manager || selectedAgent?.blocking === true) {
+			// Sync fallback: async execution disabled, orphaned host that never
+			// wired a job manager, or an agent definition that declares
+			// `blocking: true`. The session-scoped semaphore still bounds fan-out
+			// across parallel task calls.
+			if (asyncEnabled && !manager) {
+				logger.warn("task: no AsyncJobManager registered; falling back to sync execution");
+			}
+			return withAdvisory(await this.#executeSyncFanout(toolCallId, params, spawnItems, signal, onUpdate));
+		}
+
+		// Resolve agent ids up front so the immediate result can name them.
+		const outputManager =
+			this.session.agentOutputManager ?? new AgentOutputManager(this.session.getArtifactsDir ?? (() => null));
+		const agentLabel = params.agent ?? "task";
+		const agentSource = selectedAgent?.source ?? "bundled";
+		const spawns: Array<{ agentId: string; item: TaskItem; progress: AgentProgress }> = [];
+		for (let index = 0; index < spawnItems.length; index++) {
+			const item = spawnItems[index];
+			const agentId = await outputManager.allocate(item.id?.trim() || generateTaskName());
+			const assignment = (item.assignment ?? "").trim();
+			spawns.push({
+				agentId,
+				item,
+				progress: {
+					index,
+					id: agentId,
+					agent: agentLabel,
+					agentSource,
+					status: "pending",
+					task: renderSubagentUserPrompt(assignment),
+					assignment,
+					description: item.description,
+					recentTools: [],
+					recentOutput: [],
+					toolCount: 0,
+					requests: 0,
+					tokens: 0,
+					cost: 0,
+					durationMs: 0,
+				},
+			});
+		}
+
+		// Aggregate async state for the one tool call: every spawn's job reports
+		// into the shared progress snapshot; the call stays "running" until all
+		// jobs settle, then turns "failed" if any spawn failed. The single-spawn
+		// case passes the job's own suggestion through (pre-batch behavior).
+		const single = spawns.length === 1;
+		let settledCount = 0;
+		let failedCount = 0;
+		let primaryJobId = spawns[0].agentId;
+		const buildAsyncDetails = (state: "running" | "completed" | "failed", jobId: string): TaskToolDetails => ({
+			projectAgentsDir: null,
+			results: [],
+			totalDurationMs: 0,
+			progress: spawns.map(spawn => ({ ...spawn.progress })),
+			async: {
+				state: single ? state : settledCount < spawns.length ? "running" : failedCount > 0 ? "failed" : "completed",
+				jobId: single ? jobId : primaryJobId,
+				type: "task",
+			},
+		});
+
+		const started: Array<{ agentId: string; jobId: string; description?: string }> = [];
+		const failedSchedules: string[] = [];
+		for (const spawn of spawns) {
+			try {
+				const jobId = this.#registerSpawnJob({
+					manager,
+					toolCallId,
+					spawnParams: spawnParamsFor(params, spawn.item),
+					agentId: spawn.agentId,
+					progress: spawn.progress,
+					ircEnabled,
+					buildDetails: buildAsyncDetails,
+					onUpdate,
+					onSettled: failed => {
+						settledCount += 1;
+						if (failed) failedCount += 1;
+					},
+				});
+				if (started.length === 0) primaryJobId = jobId;
+				started.push({ agentId: spawn.agentId, jobId, description: spawn.item.description });
+			} catch (error) {
+				const message = error instanceof Error ? error.message : String(error);
+				failedSchedules.push(`${spawn.agentId}: ${message}`);
+				spawn.progress.status = "failed";
+				settledCount += 1;
+				failedCount += 1;
+			}
+		}
+
+		if (started.length === 0) {
+			return {
+				content: [
+					{
+						type: "text",
+						text: `Failed to start background task job${single ? "" : "s"}: ${failedSchedules.join("; ")}`,
+					},
+				],
+				details: { projectAgentsDir: null, results: [], totalDurationMs: 0 },
+			};
+		}
+
+		if (single) {
+			const { agentId, jobId, description } = started[0];
+			const coordinationHint = ircEnabled
+				? `DM \`${agentId}\` via \`irc\` to coordinate while it runs; use \`job\` only to inspect (\`list\`), wait (\`poll\`), or cancel a stuck task.`
+				: `Use \`job\` to inspect (\`list\`), wait (\`poll\`), or cancel a stuck task.`;
+			const descriptionSuffix = description ? ` — ${description}` : "";
+			onUpdate?.({
+				content: [{ type: "text", text: `Spawned agent \`${agentId}\`...` }],
+				details: buildAsyncDetails("running", jobId),
+			});
+			return withAdvisory({
+				content: [
+					{
+						type: "text",
+						text: `Spawned agent \`${agentId}\` (job \`${jobId}\`)${descriptionSuffix}. The result will be delivered when it yields. ${coordinationHint}`,
+					},
+				],
+				details: buildAsyncDetails("running", jobId),
+			});
+		}
+
+		const coordinationHint = ircEnabled
+			? `DM these ids via \`irc\` to coordinate while they run; use \`job\` only to inspect (\`list\`), wait (\`poll\`), or cancel a stuck task.`
+			: `Use \`job\` to inspect (\`list\`), wait (\`poll\`), or cancel a stuck task by id.`;
+		const scheduleFailureSummary =
+			failedSchedules.length > 0
+				? ` Failed to schedule ${failedSchedules.length} spawn${failedSchedules.length === 1 ? "" : "s"}: ${failedSchedules.join("; ")}.`
+				: "";
+		const startedListing = started
+			.map(({ agentId, jobId, description }) => {
+				const prefix = `- \`${agentId}\` (job \`${jobId}\`)`;
+				return description ? `${prefix} — ${description}` : prefix;
+			})
+			.join("\n");
+		onUpdate?.({
+			content: [{ type: "text", text: `Spawned ${started.length} agents...` }],
+			details: buildAsyncDetails("running", primaryJobId),
+		});
+		return withAdvisory({
+			content: [
+				{
+					type: "text",
+					text: `Spawned ${started.length} background agents using ${agentLabel}.${scheduleFailureSummary} Each result will be delivered when that agent yields.\n${startedListing}\n${coordinationHint}`,
+				},
+			],
+			details: buildAsyncDetails("running", primaryJobId),
+		});
+	}
+
+	/**
+	 * Register one background job that runs a single spawn to completion and
+	 * delivers its yield text. The job body mirrors the sync path; `buildDetails`
+	 * supplies the (possibly batch-shared) progress snapshot and `onSettled`
+	 * feeds the caller's aggregate counters.
+	 */
+	#registerSpawnJob(options: {
+		manager: AsyncJobManager;
+		toolCallId: string;
+		spawnParams: TaskParams;
+		agentId: string;
+		progress: AgentProgress;
+		ircEnabled: boolean;
+		buildDetails: (state: "running" | "completed" | "failed", jobId: string) => TaskToolDetails;
+		onUpdate?: AgentToolUpdateCallback<TaskToolDetails>;
+		onSettled?: (failed: boolean) => void;
+	}): string {
+		const { manager, toolCallId, spawnParams, agentId, progress, ircEnabled, buildDetails, onUpdate, onSettled } =
+			options;
+		const buildFollowUpHint = (aborted: boolean): string => {
+			if (aborted) {
+				return `\n\n${agentId} was aborted — transcript at history://${agentId}`;
+			}
+			const followUp = ircEnabled ? "message it via `irc` to follow up; " : "";
+			return `\n\n${agentId} is now idle — ${followUp}transcript at history://${agentId}`;
+		};
+		return manager.register(
+			"task",
+			agentId,
+			async ({ jobId: ownJobId, signal: runSignal, reportProgress, markRunning }) => {
+				const startedAt = Date.now();
+				const semaphore = this.#getSpawnSemaphore();
+				await semaphore.acquire();
+				const acquiredAt = Date.now();
+				if (runSignal.aborted) {
+					semaphore.release();
+					progress.status = "aborted";
+					onSettled?.(true);
+					throw new Error("Aborted before execution");
+				}
+				markRunning();
+				progress.status = "running";
+				await reportProgress(
+					`Running background task ${agentId}...`,
+					buildDetails("running", ownJobId) as unknown as Record<string, unknown>,
+				);
+				try {
+					const result = await this.#executeSync(
+						toolCallId,
+						spawnParams,
+						runSignal,
+						undefined,
+						agentId,
+						progress.index,
+						true,
+						{ invokedAt: startedAt, acquiredAt },
+					);
+					const finalText = result.content.find(part => part.type === "text")?.text ?? "(no output)";
+					const singleResult = result.details?.results[0];
+					// A missing result means the sync path failed at the tool level
+					// (results: []) — treat it as a failure, not success.
+					const resultFailed = !singleResult || (singleResult.aborted ?? false) || singleResult.exitCode !== 0;
+					progress.status = singleResult?.aborted ? "aborted" : resultFailed ? "failed" : "completed";
+					progress.durationMs = singleResult?.durationMs ?? Math.max(0, Date.now() - startedAt);
+					progress.tokens = singleResult?.tokens ?? 0;
+					progress.requests = singleResult?.requests ?? 0;
+					progress.contextTokens = singleResult?.contextTokens;
+					progress.contextWindow = singleResult?.contextWindow;
+					progress.cost = singleResult?.usage?.cost.total ?? 0;
+					progress.extractedToolData = singleResult?.extractedToolData;
+					progress.retryFailure = singleResult?.retryFailure;
+					progress.retryState = undefined;
+					onSettled?.(resultFailed);
+					const statusText = resultFailed
+						? `Background task ${agentId} failed.`
+						: `Background task ${agentId} complete.`;
+					await reportProgress(
+						statusText,
+						buildDetails(resultFailed ? "failed" : "completed", ownJobId) as unknown as Record<string, unknown>,
+					);
+					onUpdate?.({
+						content: [{ type: "text", text: statusText }],
+						details: buildDetails(resultFailed ? "failed" : "completed", ownJobId),
+					});
+					const deliveryText = `${finalText}${buildFollowUpHint(singleResult?.aborted === true)}`;
+					if (resultFailed) {
+						// Mark the job itself failed; the failed agent stays interrogable.
+						throw new TaskJobError(deliveryText);
+					}
+					return deliveryText;
+				} catch (error) {
+					if (error instanceof TaskJobError) {
+						throw error;
+					}
+					progress.status = "failed";
+					progress.durationMs = Math.max(0, Date.now() - startedAt);
+					onSettled?.(true);
+					const statusText = `Background task ${agentId} failed.`;
+					await reportProgress(statusText, buildDetails("failed", ownJobId) as unknown as Record<string, unknown>);
+					onUpdate?.({
+						content: [{ type: "text", text: statusText }],
+						details: buildDetails("failed", ownJobId),
+					});
+					const message = error instanceof Error ? error.message : String(error);
+					const hint = AgentRegistry.global().get(agentId) ? buildFollowUpHint(false) : "";
+					throw new TaskJobError(`${message}${hint}`);
+				} finally {
+					semaphore.release();
+				}
+			},
+			{
+				id: agentId,
+				queued: true,
+				ownerId: this.session.getAgentId?.() ?? undefined,
+				onProgress: (text, details) => {
+					const progressDetails = (details as TaskToolDetails | undefined) ?? buildDetails("running", agentId);
+					onUpdate?.({ content: [{ type: "text", text }], details: progressDetails });
+				},
+			},
+		);
+	}
+
+	/**
+	 * Sync fallback fan-out (no job manager, or a `blocking: true` agent): run
+	 * every spawn to completion inline and merge the per-spawn payloads into a
+	 * single tool result. The session-scoped semaphore still bounds concurrency
+	 * across parallel task calls.
+	 */
+	async #executeSyncFanout(
+		toolCallId: string,
+		params: TaskParams,
+		spawnItems: TaskItem[],
+		signal?: AbortSignal,
+		onUpdate?: AgentToolUpdateCallback<TaskToolDetails>,
+	): Promise<AgentToolResult<TaskToolDetails>> {
+		const semaphore = this.#getSpawnSemaphore();
+		if (spawnItems.length === 1) {
+			const invokedAt = Date.now();
+			await semaphore.acquire();
+			const acquiredAt = Date.now();
+			try {
+				return await this.#executeSync(
+					toolCallId,
+					spawnParamsFor(params, spawnItems[0]),
+					signal,
+					onUpdate,
+					undefined,
+					0,
+					false,
+					{ invokedAt, acquiredAt },
+				);
+			} finally {
+				semaphore.release();
+			}
+		}
+
+		const startTime = Date.now();
+		const latestProgress = new Map<number, AgentProgress>();
+		const emitCombined = () => {
+			onUpdate?.({
+				content: [{ type: "text", text: `Running ${spawnItems.length} agents...` }],
+				details: {
+					projectAgentsDir: null,
+					results: [],
+					totalDurationMs: Date.now() - startTime,
+					progress: Array.from(latestProgress.entries())
+						.sort((a, b) => a[0] - b[0])
+						.map(([, progress]) => progress),
+				},
+			});
+		};
+
+		const { results: payloads } = await mapWithConcurrencyLimit(
+			spawnItems,
+			spawnItems.length,
+			async (item, index, workerSignal) => {
+				const invokedAt = Date.now();
+				await semaphore.acquire();
+				const acquiredAt = Date.now();
+				try {
+					const itemOnUpdate: AgentToolUpdateCallback<TaskToolDetails> | undefined = onUpdate
+						? update => {
+								const progress = update.details?.progress?.[0];
+								if (progress) {
+									latestProgress.set(index, { ...progress, index });
+									emitCombined();
+								}
+							}
+						: undefined;
+					return await this.#executeSync(
+						toolCallId,
+						spawnParamsFor(params, item),
+						workerSignal,
+						itemOnUpdate,
+						undefined,
+						index,
+						false,
+						{ invokedAt, acquiredAt },
+					);
+				} finally {
+					semaphore.release();
+				}
+			},
+			signal,
+		);
+
+		const results: SingleResult[] = [];
+		const contentParts: string[] = [];
+		const outputPaths: string[] = [];
+		const usageTotals = createUsageTotals();
+		let hasUsage = false;
+		let projectAgentsDir: string | null = null;
+		for (let index = 0; index < spawnItems.length; index++) {
+			const payload = payloads[index];
+			if (!payload) {
+				contentParts.push(`Task ${spawnItems[index].id?.trim() || `#${index + 1}`}: cancelled before start.`);
+				continue;
+			}
+			projectAgentsDir ??= payload.details?.projectAgentsDir ?? null;
+			const text = payload.content.find(part => part.type === "text")?.text;
+			if (text) contentParts.push(text);
+			for (const result of payload.details?.results ?? []) {
+				results.push({ ...result, index });
+				if (result.usage) {
+					addUsageTotals(usageTotals, result.usage);
+					hasUsage = true;
+				}
+				if (result.outputPath) outputPaths.push(result.outputPath);
+			}
+		}
+
+		return {
+			content: [{ type: "text", text: contentParts.join("\n\n") }],
+			details: {
+				projectAgentsDir,
+				results,
+				totalDurationMs: Date.now() - startTime,
+				usage: hasUsage ? usageTotals : undefined,
+				outputPaths: outputPaths.length > 0 ? outputPaths : undefined,
+			},
+		};
+	}
+
+	/**
+	 * Synchronous execution of one spawn. Used as the body of every
+	 * async job and directly by the sync fallback (no job manager / blocking
+	 * agent) and by in-process callers that need the result inline (e.g. the
+	 * commit flow's analyze_files tool).
+	 */
+	async #executeSync(
+		toolCallId: string,
+		params: TaskParams,
+		signal?: AbortSignal,
+		onUpdate?: AgentToolUpdateCallback<TaskToolDetails>,
+		preAllocatedId?: string,
+		spawnIndex = 0,
+		detached = false,
+		launchTiming?: { invokedAt: number; acquiredAt: number },
+	): Promise<AgentToolResult<TaskToolDetails>> {
+		return this.#runSpawn(toolCallId, params, signal, onUpdate, preAllocatedId, spawnIndex, detached, launchTiming);
+	}
+
+	/** Spawn a fresh subagent and run it to completion. */
+	async #runSpawn(
+		toolCallId: string,
+		params: TaskParams,
+		signal?: AbortSignal,
+		onUpdate?: AgentToolUpdateCallback<TaskToolDetails>,
+		preAllocatedId?: string,
+		spawnIndex = 0,
+		detached = false,
+		launchTiming?: { invokedAt: number; acquiredAt: number },
+	): Promise<AgentToolResult<TaskToolDetails>> {
+		const startTime = Date.now();
+		const { agents, projectAgentsDir } = await discoverAgents(this.session.cwd);
+		const agentName = params.agent ?? "";
+		const sharedContext = this.#isBatchEnabled() ? params.context?.trim() || undefined : undefined;
+		const assignment = (params.assignment ?? "").trim();
+		const isolationMode = this.session.settings.get("task.isolation.mode");
+		const isolationRequested = "isolated" in params ? params.isolated === true : false;
+		const isIsolated = isolationMode !== "none" && isolationRequested;
+		const mergeMode = this.session.settings.get("task.isolation.merge");
+		const commitStyle = this.session.settings.get("task.isolation.commits");
+		const taskDepth = this.session.taskDepth ?? 0;
+		const subagentLspEnabled = (this.session.enableLsp ?? true) && this.session.settings.get("task.enableLsp");
+
+		if (isolationMode === "none" && "isolated" in params) {
+			return {
+				content: [{ type: "text", text: "Task isolation is disabled." }],
+				details: { projectAgentsDir, results: [], totalDurationMs: 0 },
+			};
+		}
+
+		// Validate agent exists
+		const agent = getAgent(agents, agentName);
+		if (!agent) {
+			const available = agents.map(a => a.name).join(", ") || "none";
+			return {
+				content: [{ type: "text", text: `Unknown agent "${agentName}". Available: ${available}` }],
+				details: { projectAgentsDir, results: [], totalDurationMs: 0 },
+			};
+		}
+
+		// Check if agent is disabled in settings
+		const disabledAgents = this.session.settings.get("task.disabledAgents") as string[];
+		if (disabledAgents.length > 0 && disabledAgents.includes(agentName)) {
+			const enabled = agents.filter(a => !disabledAgents.includes(a.name)).map(a => a.name);
+			return {
+				content: [
+					{
+						type: "text",
+						text: `Agent "${agentName}" is disabled in settings. Enable it via /agents, or use a different agent type.${enabled.length > 0 ? ` Available: ${enabled.join(", ")}` : ""}`,
+					},
+				],
+				details: { projectAgentsDir, results: [], totalDurationMs: 0 },
+			};
+		}
+
+		const planModeState = this.session.getPlanModeState?.();
+		const planModeBaseTools = ["read", "search", "find", "lsp", "web_search"];
+		const planModeTools = [
+			...planModeBaseTools,
+			...(agent.tools ?? []).filter(
+				tool => PLAN_MODE_AGENT_TOOL_ALLOWLIST.has(tool) && !planModeBaseTools.includes(tool),
+			),
+		];
+		const effectiveAgent: typeof agent = planModeState?.enabled
+			? {
+					...agent,
+					systemPrompt: `${planModeSubagentPrompt}\n\n${agent.systemPrompt}`,
+					tools: planModeTools,
+					spawns: undefined,
+				}
+			: agent;
+
+		// Apply per-agent model override from settings (highest priority)
+		const agentModelOverrides = this.session.settings.get("task.agentModelOverrides");
+		const settingsModelOverride = agentModelOverrides[agentName];
+		const parentActiveModelPattern = this.session.getActiveModelString?.();
+		const modelOverride = resolveAgentModelPatterns({
+			settingsOverride: settingsModelOverride,
+			agentModel: effectiveAgent.model,
+			settings: this.session.settings,
+			activeModelPattern: parentActiveModelPattern,
+			fallbackModelPattern: this.session.getModelString?.(),
+		});
+		const thinkingLevelOverride = effectiveAgent.thinkingLevel;
+
+		// Output schema priority: agent frontmatter > inherited parent session.
+		// The task call itself never carries a schema; workflows needing ad-hoc
+		// structured output go through eval agent(prompt, schema).
+		const effectiveOutputSchema = effectiveAgent.output ?? this.session.outputSchema;
+
+		let repoRoot: string | null = null;
+		let baseline: WorktreeBaseline | null = null;
+		if (isIsolated) {
+			try {
+				repoRoot = await getRepoRoot(this.session.cwd);
+				baseline = await captureBaseline(repoRoot);
+			} catch (err) {
+				const message = err instanceof Error ? err.message : String(err);
+				return {
+					content: [{ type: "text", text: `Isolated task execution requires a git repository. ${message}` }],
+					details: { projectAgentsDir, results: [], totalDurationMs: Date.now() - startTime },
+				};
+			}
+		}
+
+		const preferredIsolationBackend = parseIsolationMode(isolationMode);
+
+		// Derive artifacts directory
+		const sessionFile = this.session.getSessionFile();
+		const artifactsDir = sessionFile ? sessionFile.slice(0, -6) : null;
+		const tempArtifactsDir = artifactsDir ? null : path.join(os.tmpdir(), `omp-task-${Snowflake.next()}`);
+		const effectiveArtifactsDir = artifactsDir || tempArtifactsDir!;
+
+		const localProtocolOptions: LocalProtocolOptions = this.session.localProtocolOptions ?? {
+			getArtifactsDir: this.session.getArtifactsDir ?? (() => null),
+			getSessionId: this.session.getSessionId ?? (() => null),
+		};
+
+		// Subagents adopt the parent's ArtifactManager so artifact IDs are unique
+		// across the whole tree and outputs land flat in the parent's dir.
+		const parentArtifactManager = this.session.getArtifactManager?.() ?? undefined;
+
+		// When the session is executing an approved plan, hand the overall plan to
+		// every subagent so they share the main agent's plan context. Skipped in
+		// plan mode (read-only exploration uses planModeSubagentPrompt instead) and
+		// when no plan file exists at the session's reference path.
+		const planReference = planModeState?.enabled
+			? undefined
+			: await loadOverallPlanReference(
+					this.session.getPlanReferencePath?.() ?? "local://PLAN.md",
+					localProtocolOptions,
+				);
+
+		try {
+			// Check self-recursion prevention
+			if (this.#blockedAgent && agentName === this.#blockedAgent) {
+				return {
+					content: [
+						{
+							type: "text",
+							text: `Cannot spawn ${this.#blockedAgent} agent from within itself (recursion prevention). Use a different agent type.`,
+						},
+					],
+					details: { projectAgentsDir, results: [], totalDurationMs: Date.now() - startTime },
+				};
+			}
+
+			// Check spawn restrictions from parent
+			const parentSpawns = this.session.getSessionSpawns() ?? "*";
+			const allowedSpawns = parentSpawns.split(",").map(s => s.trim());
+			const isSpawnAllowed = (): boolean => {
+				if (parentSpawns === "") return false; // Empty = deny all
+				if (parentSpawns === "*") return true; // Wildcard = allow all
+				return allowedSpawns.includes(agentName);
+			};
+
+			if (!isSpawnAllowed()) {
+				const allowed = parentSpawns === "" ? "none (spawns disabled for this agent)" : parentSpawns;
+				return {
+					content: [{ type: "text", text: `Cannot spawn '${agentName}'. Allowed: ${allowed}` }],
+					details: { projectAgentsDir, results: [], totalDurationMs: Date.now() - startTime },
+				};
+			}
+
+			await fs.mkdir(effectiveArtifactsDir, { recursive: true });
+
+			// Allocate a unique ID across the session to prevent artifact collisions
+			let agentId: string;
+			if (preAllocatedId) {
+				agentId = preAllocatedId;
+			} else {
+				const outputManager =
+					this.session.agentOutputManager ?? new AgentOutputManager(this.session.getArtifactsDir ?? (() => null));
+				agentId = await outputManager.allocate(params.id?.trim() || generateTaskName());
+			}
+
+			const availableSkills = [...(this.session.skills ?? [])];
+			// Resolve autoload skills from agent definition against available skills
+			const resolvedAutoloadSkills =
+				agent.autoloadSkills?.length && availableSkills.length > 0
+					? agent.autoloadSkills
+							.map(name => availableSkills.find(s => s.name === name))
+							.filter((s): s is NonNullable<typeof s> => s !== undefined)
+					: [];
+			const contextFiles = this.session.contextFiles?.filter(
+				file => path.basename(file.path).toLowerCase() !== "agents.md",
+			);
+			const promptTemplates = this.session.promptTemplates;
+			const parentEvalSessionId = this.session.getEvalSessionId?.() ?? undefined;
+			const mcpManager = this.session.mcpManager ?? MCPManager.instance();
+
+			// Progress tracking for the single agent
+			let latestProgress: AgentProgress = {
+				index: spawnIndex,
+				id: agentId,
+				agent: agentName,
+				agentSource: agent.source,
+				status: "pending",
+				task: renderSubagentUserPrompt(assignment),
+				assignment,
+				recentTools: [],
+				recentOutput: [],
+				toolCount: 0,
+				requests: 0,
+				tokens: 0,
+				cost: 0,
+				durationMs: 0,
+				modelOverride,
+				description: params.description,
+			};
+			const emitProgress = () => {
+				onUpdate?.({
+					content: [{ type: "text", text: `Running agent ${agentId}...` }],
+					details: {
+						projectAgentsDir,
+						results: [],
+						totalDurationMs: Date.now() - startTime,
+						progress: [latestProgress],
+					},
+				});
+			};
+			emitProgress();
+
+			const buildCommitMessageFn = () =>
+				commitStyle === "ai" && this.session.modelRegistry
+					? async (diff: string) => {
+							return generateCommitMessage(
+								diff,
+								this.session.modelRegistry!,
+								this.session.settings,
+								this.session.getSessionId?.() ?? undefined,
+							);
+						}
+					: undefined;
+
+			const sharedRunOptions = {
+				cwd: this.session.cwd,
+				agent: effectiveAgent,
+				task: renderSubagentUserPrompt(assignment),
+				assignment,
+				context: sharedContext,
+				planReference,
+				description: params.description,
+				role: params.role,
+				index: spawnIndex,
+				parentToolCallId: toolCallId,
+				detached,
+				id: agentId,
+				taskDepth,
+				invokedAt: launchTiming?.invokedAt,
+				acquiredAt: launchTiming?.acquiredAt,
+				modelOverride,
+				parentActiveModelPattern,
+				thinkingLevel: thinkingLevelOverride,
+				outputSchema: effectiveOutputSchema,
+				sessionFile,
+				persistArtifacts: !!artifactsDir,
+				artifactsDir: effectiveArtifactsDir,
+				enableLsp: subagentLspEnabled,
+				signal,
+				eventBus: this.session.eventBus,
+				onProgress: (progress: AgentProgress) => {
+					// Shallow snapshot; recentTools is mutated in place by the
+					// executor, the rest is reassigned or immutable. A deep clone
+					// here cost O(extractedToolData) per progress event.
+					latestProgress = { ...progress, recentTools: progress.recentTools.slice() };
+					emitProgress();
+				},
+				authStorage: this.session.authStorage,
+				modelRegistry: this.session.modelRegistry,
+				settings: this.session.settings,
+				mcpManager,
+				contextFiles,
+				skills: availableSkills,
+				autoloadSkills: resolvedAutoloadSkills,
+				workspaceTree: this.session.workspaceTree,
+				promptTemplates,
+				rules: this.session.rules,
+				preloadedExtensionPaths: this.session.extensionPaths,
+				preloadedCustomToolPaths: this.session.customToolPaths,
+				localProtocolOptions,
+				parentArtifactManager,
+				parentHindsightSessionState: this.session.getHindsightSessionState?.(),
+				parentMnemopiSessionState: this.session.getMnemopiSessionState?.(),
+				parentTelemetry: this.session.getTelemetry?.(),
+				parentEvalSessionId,
+				parentAgentId: this.session.getAgentId?.() ?? MAIN_AGENT_ID,
+			};
+
+			const runTask = async (): Promise<SingleResult> => {
+				if (!isIsolated) {
+					return runSubprocess(sharedRunOptions);
+				}
+
+				const taskStart = Date.now();
+				let isolationHandle: IsolationHandle | undefined;
+				try {
+					if (!repoRoot || !baseline) {
+						throw new Error("Isolated task execution not initialized.");
+					}
+					const taskBaseline = structuredClone(baseline);
+
+					isolationHandle = await ensureIsolation(repoRoot, agentId, preferredIsolationBackend);
+					const isolationDir = isolationHandle.mergedDir;
+
+					// Isolated runs re-discover extensions/custom tools inside the
+					// worktree instead of reusing the parent's source paths.
+					const result = await runSubprocess({
+						...sharedRunOptions,
+						worktree: isolationDir,
+						preloadedExtensionPaths: undefined,
+						preloadedCustomToolPaths: undefined,
+					});
+					if (mergeMode === "branch" && result.exitCode === 0) {
+						try {
+							const commitResult = await commitToBranch(
+								isolationDir,
+								taskBaseline,
+								agentId,
+								params.description,
+								buildCommitMessageFn(),
+							);
+							return {
+								...result,
+								branchName: commitResult?.branchName,
+								nestedPatches: commitResult?.nestedPatches,
+							};
+						} catch (mergeErr) {
+							// Agent succeeded but branch commit failed — clean up stale branch
+							const branchName = `omp/task/${agentId}`;
+							await git.branch.tryDelete(repoRoot, branchName);
+							const msg = mergeErr instanceof Error ? mergeErr.message : String(mergeErr);
+							return { ...result, error: `Merge failed: ${msg}` };
+						}
+					}
+					if (result.exitCode === 0) {
+						try {
+							const delta = await captureDeltaPatch(isolationDir, taskBaseline);
+							const patchPath = path.join(effectiveArtifactsDir, `${agentId}.patch`);
+							await Bun.write(patchPath, delta.rootPatch);
+							return {
+								...result,
+								patchPath,
+								nestedPatches: delta.nestedPatches,
+							};
+						} catch (patchErr) {
+							const msg = patchErr instanceof Error ? patchErr.message : String(patchErr);
+							return { ...result, error: `Patch capture failed: ${msg}` };
+						}
+					}
+					return result;
+				} catch (err) {
+					const message = err instanceof Error ? err.message : String(err);
+					return {
+						index: spawnIndex,
+						id: agentId,
+						agent: agent.name,
+						agentSource: agent.source,
+						task: renderSubagentUserPrompt(assignment),
+						assignment,
+						description: params.description,
+						exitCode: 1,
+						output: "",
+						stderr: message,
+						truncated: false,
+						durationMs: Date.now() - taskStart,
+						tokens: 0,
+						requests: 0,
+						modelOverride,
+						error: message,
+					};
+				} finally {
+					if (isolationHandle) {
+						await cleanupIsolation(isolationHandle);
+					}
+				}
+			};
+
+			const result = await runTask();
+
+			let mergeSummary = "";
+			let changesApplied: boolean | null = null;
+			let hadAnyChanges = false;
+			let mergedBranchForNestedPatches = false;
+			if (isIsolated && repoRoot) {
+				try {
+					if (mergeMode === "branch") {
+						if (!result.branchName || result.exitCode !== 0 || result.aborted) {
+							changesApplied = true;
+							mergeSummary = "\n\nNo changes to apply.";
+						} else {
+							const mergeResult = await mergeTaskBranches(repoRoot, [
+								{ branchName: result.branchName, taskId: result.id, description: result.description },
+							]);
+							mergedBranchForNestedPatches = mergeResult.merged.includes(result.branchName);
+							changesApplied = mergeResult.failed.length === 0;
+							hadAnyChanges = changesApplied && mergeResult.merged.length > 0;
+
+							if (changesApplied) {
+								mergeSummary = hadAnyChanges
+									? `\n\nMerged branch: ${result.branchName}`
+									: "\n\nNo changes to apply.";
+							} else {
+								const conflictPart = mergeResult.conflict ? `\nConflict: ${mergeResult.conflict}` : "";
+								mergeSummary = `\n\n<system-notification>Branch merge failed: ${result.branchName}.${conflictPart}\nThe unmerged branch remains for manual resolution.</system-notification>`;
+							}
+							if (mergeResult.stashConflict) {
+								mergeSummary += `\n\n<system-notification>${mergeResult.stashConflict}</system-notification>`;
+							}
+
+							// Clean up the merged branch (keep failed ones for manual resolution)
+							if (changesApplied) {
+								await cleanupTaskBranches(repoRoot, [result.branchName]);
+							}
+						}
+					} else {
+						// Patch mode: apply the patch from a successful run. A failed or
+						// aborted run has nothing to apply and must not block the result.
+						const succeeded = result.exitCode === 0 && !result.error && !result.aborted;
+						if (!succeeded) {
+							changesApplied = true;
+							hadAnyChanges = false;
+						} else if (!result.patchPath) {
+							changesApplied = false;
+							hadAnyChanges = false;
+						} else {
+							const patchText = await Bun.file(result.patchPath).text();
+							if (!patchText.trim()) {
+								changesApplied = true;
+								hadAnyChanges = false;
+							} else {
+								const normalized = patchText.endsWith("\n") ? patchText : `${patchText}\n`;
+								changesApplied = await git.patch.canApplyText(repoRoot, normalized);
+								if (changesApplied) {
+									try {
+										await git.patch.applyText(repoRoot, normalized);
+										hadAnyChanges = true;
+									} catch {
+										changesApplied = false;
+										hadAnyChanges = false;
+									}
+								}
+							}
+						}
+
+						if (changesApplied) {
+							mergeSummary = hadAnyChanges ? "\n\nApplied patches: yes" : "\n\nNo changes to apply.";
+						} else {
+							const notification =
+								"<system-notification>Patches were not applied and must be handled manually.</system-notification>";
+							const patchList = result.patchPath ? `\n\nPatch artifact:\n- ${result.patchPath}` : "";
+							mergeSummary = `\n\n${notification}${patchList}`;
+						}
+					}
+				} catch (mergeErr) {
+					const msg = mergeErr instanceof Error ? mergeErr.message : String(mergeErr);
+					changesApplied = false;
+					hadAnyChanges = false;
+					mergeSummary = `\n\n<system-notification>Merge phase failed: ${msg}\nTask outputs are preserved but changes were not applied.</system-notification>`;
+				}
+			}
+
+			// Apply nested repo patches (separate from parent git)
+			if (isIsolated && repoRoot && (mergeMode === "branch" || changesApplied !== false)) {
+				const nestedPatches = result.nestedPatches ?? [];
+				const eligible =
+					nestedPatches.length > 0 &&
+					result.exitCode === 0 &&
+					!result.aborted &&
+					(mergeMode !== "branch" || mergedBranchForNestedPatches);
+				if (eligible) {
+					try {
+						await applyNestedPatches(repoRoot, nestedPatches, buildCommitMessageFn());
+					} catch {
+						// Nested patch failures are non-fatal to the parent merge
+						mergeSummary +=
+							"\n\n<system-notification>Some nested repository patches failed to apply.</system-notification>";
+					}
+				}
+			}
+
+			// Cleanup temp directory if used
+			const shouldCleanupTempArtifacts =
+				tempArtifactsDir && (!isIsolated || changesApplied === true || changesApplied === null);
+			if (shouldCleanupTempArtifacts) {
+				await fs.rm(tempArtifactsDir, { recursive: true, force: true });
+			}
+
+			return this.#buildResultPayload(result, projectAgentsDir, Date.now() - startTime, mergeSummary);
+		} catch (err) {
+			return {
+				content: [{ type: "text", text: `Task execution failed: ${err}` }],
+				details: { projectAgentsDir, results: [], totalDurationMs: Date.now() - startTime },
+			};
+		}
+	}
+
+	/** Build the tool result (summary text + details) for a settled run. */
+	#buildResultPayload(
+		result: SingleResult,
+		projectAgentsDir: string | null,
+		totalDurationMs: number,
+		mergeSummary: string,
+	): AgentToolResult<TaskToolDetails> {
+		const status = result.aborted
+			? "cancelled"
+			: result.exitCode === 0 && result.error
+				? "merge failed"
+				: result.exitCode === 0
+					? "completed"
+					: `failed (exit ${result.exitCode})`;
+		const output = formatResultOutputFallback(result);
+		const outputCharCount = result.outputMeta?.charCount ?? output.length;
+		const fullOutputThreshold = 5000;
+		let preview = output;
+		let truncated = false;
+		if (outputCharCount > fullOutputThreshold) {
+			const slice = output.slice(0, fullOutputThreshold);
+			const lastNewline = slice.lastIndexOf("\n");
+			preview = lastNewline >= 0 ? slice.slice(0, lastNewline) : slice;
+			truncated = true;
+		}
+		const summary = prompt.render(taskSummaryTemplate, {
+			agentName: result.agent,
+			id: result.id,
+			status,
+			duration: formatDuration(totalDurationMs),
+			preview,
+			truncated,
+			meta: result.outputMeta
+				? {
+						lineCount: result.outputMeta.lineCount,
+						charSize: formatBytes(result.outputMeta.charCount),
+					}
+				: undefined,
+			mergeSummary,
+		});
+
+		return {
+			content: [{ type: "text", text: summary }],
+			details: {
+				projectAgentsDir,
+				results: [result],
+				totalDurationMs,
+				usage: result.usage,
+				outputPaths: result.outputPath ? [result.outputPath] : undefined,
+			},
+		};
+	}
+}