@oh-my-pi/pi-coding-agent 15.10.11 → 15.11.0

package/src/cli/gallery-fixtures/agentic.ts

@@ -1,62 +1,76 @@
-// Gallery fixtures for the agentic orchestration tools (task, goal, job).
+// Gallery fixtures for the agentic orchestration tools (task, irc, goal, job).
+import type { Usage } from "@oh-my-pi/pi-ai";
+import type { TaskToolDetails } from "../../task/types";
+import type { IrcDetails } from "../../tools/irc";
 import type { GalleryFixture } from "./types";
 
+/** Message/activity timestamps are offsets from load time so gallery ages stay plausible. */
+const FIXTURE_NOW = Date.now();
+
+/** Plausible cumulative usage for a fixture subagent run. */
+const fixtureUsage = (tokens: { input: number; output: number }, costTotal: number): Usage => ({
+	input: tokens.input,
+	output: tokens.output,
+	cacheRead: 0,
+	cacheWrite: 0,
+	totalTokens: tokens.input + tokens.output,
+	cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0, total: costTotal },
+});
+
 export const agenticFixtures: Record<string, GalleryFixture> = {
 	task: {
 		label: "Task",
 		customRendered: true,
-		// Streaming: agent chosen, first task fully arrived, second still landing.
+		// Streaming: agent chosen, assignment still landing.
 		streamingArgs: {
 			agent: "task",
-			tasks: [
-				{
-					id: "AuthLoader",
-					description: "Load auth middleware",
-					assignment: "Read packages/server/src/auth/*.ts and summarize the session-cookie flow.",
-				},
-				{ id: "RateLimiter", description: "Audit rate limiter" },
-			],
+			id: "AuthLoader",
+			description: "Load auth middleware",
+			assignment: "Read packages/server/src/auth/*.ts and summarize the session-cookie",
 		},
 		args: {
 			agent: "task",
-			context: [
-				"# Goal",
-				"Harden the HTTP auth stack before the release cut.",
-				"# Constraints",
-				"Touch only files under packages/server/src/auth/. Do not run gates.",
-			].join("\n"),
-			tasks: [
-				{
-					id: "AuthLoader",
-					description: "Load auth middleware",
-					assignment:
-						"Read packages/server/src/auth/session.ts and middleware.ts, then document the session-cookie validation flow and any TODOs.",
-				},
-				{
-					id: "RateLimiter",
-					description: "Audit rate limiter",
-					assignment:
-						"Inspect packages/server/src/auth/rate-limit.ts. Confirm the 429 path sets Retry-After and report gaps.",
-				},
-				{
-					id: "TokenRotation",
-					description: "Check token rotation",
-					assignment:
-						"Trace refresh-token rotation in packages/server/src/auth/tokens.ts and flag any reuse window.",
-				},
-			],
+			id: "AuthLoader",
+			description: "Load auth middleware",
+			assignment:
+				"Read packages/server/src/auth/session.ts and middleware.ts, then document the session-cookie validation flow and any TODOs.",
 		},
 		result: {
 			content: [
 				{
 					type: "text",
-					text: "3 agents completed: AuthLoader, RateLimiter, TokenRotation.",
+					text: "Agent AuthLoader completed.",
 				},
 			],
 			details: {
 				projectAgentsDir: null,
 				totalDurationMs: 48_200,
-				usage: { cost: { total: 0.34 } },
+				usage: fixtureUsage({ input: 52_600, output: 8_800 }, 0.12),
+				progress: [
+					{
+						index: 0,
+						id: "AuthLoader",
+						agent: "task",
+						agentSource: "bundled",
+						status: "completed",
+						task: "Read packages/server/src/auth/session.ts and middleware.ts",
+						description: "Load auth middleware",
+						lastIntent: "Documenting session-cookie flow",
+						recentTools: [
+							{ tool: "read", args: "packages/server/src/auth/session.ts", endMs: 1_749_200_040_000 },
+							{ tool: "read", args: "packages/server/src/auth/middleware.ts", endMs: 1_749_200_052_000 },
+						],
+						recentOutput: ["Session validation runs in middleware.ts:42 via verifySessionCookie()."],
+						toolCount: 9,
+						requests: 6,
+						tokens: 61_400,
+						contextTokens: 23_100,
+						contextWindow: 200_000,
+						cost: 0.12,
+						durationMs: 41_900,
+						resolvedModel: "anthropic/claude-sonnet",
+					},
+				],
 				results: [
 					{
 						index: 0,
@@ -77,100 +91,31 @@ export const agenticFixtures: Record<string, GalleryFixture> = {
 						truncated: false,
 						durationMs: 41_900,
 						tokens: 61_400,
+						requests: 6,
 						contextTokens: 23_100,
 						contextWindow: 200_000,
 						resolvedModel: "anthropic/claude-sonnet",
-						usage: { cost: { total: 0.12 } },
+						usage: fixtureUsage({ input: 52_600, output: 8_800 }, 0.12),
 						outputMeta: { lineCount: 3, charCount: 214 },
 					},
-					{
-						index: 1,
-						id: "RateLimiter",
-						agent: "task",
-						agentSource: "bundled",
-						description: "Audit rate limiter",
-						task: "Inspect packages/server/src/auth/rate-limit.ts",
-						assignment:
-							"Inspect packages/server/src/auth/rate-limit.ts. Confirm the 429 path sets Retry-After and report gaps.",
-						exitCode: 0,
-						output: [
-							"rate-limit.ts uses a fixed-window counter keyed by client IP.",
-							"429 responses set Retry-After (rate-limit.ts:57).",
-							"Gap: no per-account limit, so a botnet across IPs bypasses the cap.",
-						].join("\n"),
-						stderr: "",
-						truncated: false,
-						durationMs: 38_500,
-						tokens: 54_800,
-						contextTokens: 19_700,
-						contextWindow: 200_000,
-						resolvedModel: "anthropic/claude-sonnet",
-						usage: { cost: { total: 0.1 } },
-						outputMeta: { lineCount: 3, charCount: 198 },
-					},
-					{
-						index: 2,
-						id: "TokenRotation",
-						agent: "task",
-						agentSource: "bundled",
-						description: "Check token rotation",
-						task: "Trace refresh-token rotation in packages/server/src/auth/tokens.ts",
-						assignment:
-							"Trace refresh-token rotation in packages/server/src/auth/tokens.ts and flag any reuse window.",
-						exitCode: 0,
-						output: [
-							"Refresh tokens rotate on every use (tokens.ts:120) and the old jti is revoked.",
-							"Reuse of a rotated token triggers full-family revocation — no reuse window found.",
-						].join("\n"),
-						stderr: "",
-						truncated: false,
-						durationMs: 48_200,
-						tokens: 49_200,
-						contextTokens: 17_500,
-						contextWindow: 200_000,
-						resolvedModel: "anthropic/claude-sonnet",
-						usage: { cost: { total: 0.12 } },
-						outputMeta: { lineCount: 2, charCount: 160 },
-					},
 				],
-			},
+			} satisfies TaskToolDetails,
 		},
 		errorResult: {
 			isError: true,
 			content: [
 				{
 					type: "text",
-					text: "1 of 3 agents failed: RateLimiter.",
+					text: "Agent RateLimiter failed.",
 				},
 			],
 			details: {
 				projectAgentsDir: null,
-				totalDurationMs: 39_400,
-				usage: { cost: { total: 0.21 } },
+				totalDurationMs: 9_800,
+				usage: fixtureUsage({ input: 10_900, output: 1_400 }, 0.1),
 				results: [
 					{
 						index: 0,
-						id: "AuthLoader",
-						agent: "task",
-						agentSource: "bundled",
-						description: "Load auth middleware",
-						task: "Read packages/server/src/auth/session.ts and middleware.ts",
-						assignment:
-							"Read packages/server/src/auth/session.ts and middleware.ts, then document the session-cookie validation flow and any TODOs.",
-						exitCode: 0,
-						output: "Session validation runs in middleware.ts:42 via verifySessionCookie().",
-						stderr: "",
-						truncated: false,
-						durationMs: 31_200,
-						tokens: 58_100,
-						contextTokens: 21_900,
-						contextWindow: 200_000,
-						resolvedModel: "anthropic/claude-sonnet",
-						usage: { cost: { total: 0.11 } },
-						outputMeta: { lineCount: 1, charCount: 70 },
-					},
-					{
-						index: 1,
 						id: "RateLimiter",
 						agent: "task",
 						agentSource: "bundled",
@@ -184,15 +129,185 @@ export const agenticFixtures: Record<string, GalleryFixture> = {
 						truncated: false,
 						durationMs: 9_800,
 						tokens: 12_300,
+						requests: 3,
 						contextTokens: 6_400,
 						contextWindow: 200_000,
 						resolvedModel: "anthropic/claude-sonnet",
-						usage: { cost: { total: 0.1 } },
+						usage: fixtureUsage({ input: 10_900, output: 1_400 }, 0.1),
 						error: "Subagent exited 1: target file packages/server/src/auth/rate-limit.ts does not exist.",
 						outputMeta: { lineCount: 0, charCount: 0 },
 					},
 				],
-			},
+			} satisfies TaskToolDetails,
+		},
+	},
+
+	irc: {
+		label: "IRC",
+		// Streaming: recipient known; the message body still arriving.
+		streamingArgs: { op: "send", to: "AuthLoader", message: "Are you still touching" },
+		args: {
+			op: "send",
+			to: "AuthLoader",
+			message: "Are you still touching src/server/auth.ts? I need to add a 401 path.",
+			await: true,
+		},
+		result: {
+			content: [
+				{
+					type: "text",
+					text: [
+						"Delivered to 1 peer(s):",
+						"- AuthLoader: revived",
+						"",
+						"Reply from AuthLoader:",
+						"Done with auth.ts — go ahead, just rebase past my session-store rename.",
+					].join("\n"),
+				},
+			],
+			details: {
+				op: "send",
+				from: "Main",
+				to: "AuthLoader",
+				receipts: [{ to: "AuthLoader", outcome: "revived" }],
+				waited: {
+					id: "7181122334455667789",
+					from: "AuthLoader",
+					to: "Main",
+					body: "Done with auth.ts — go ahead, just rebase past my session-store rename.",
+					ts: FIXTURE_NOW - 5_000,
+					replyTo: "7181122334455667788",
+				},
+			} satisfies IrcDetails,
+		},
+		errorResult: {
+			isError: true,
+			content: [
+				{
+					type: "text",
+					text: 'No recipients received the message.\n- RateLimiter: failed — unknown agent "RateLimiter"',
+				},
+			],
+			details: {
+				op: "send",
+				from: "Main",
+				to: "RateLimiter",
+				receipts: [{ to: "RateLimiter", outcome: "failed", error: 'unknown agent "RateLimiter"' }],
+			} satisfies IrcDetails,
+		},
+	},
+
+	irc_wait: {
+		label: "IRC (wait)",
+		customRendered: true,
+		renderer: "irc",
+		streamingArgs: { op: "wait", from: "AuthLoader" },
+		args: { op: "wait", from: "AuthLoader", timeoutMs: 60_000 },
+		result: {
+			content: [
+				{
+					type: "text",
+					text: "[7181122334455667790] AuthLoader: session-store rename is merged; auth.ts is yours.",
+				},
+			],
+			details: {
+				op: "wait",
+				from: "Main",
+				waited: {
+					id: "7181122334455667790",
+					from: "AuthLoader",
+					to: "Main",
+					body: "session-store rename is merged; auth.ts is yours.",
+					ts: FIXTURE_NOW - 30_000,
+				},
+			} satisfies IrcDetails,
+		},
+	},
+
+	irc_inbox: {
+		label: "IRC (inbox)",
+		customRendered: true,
+		renderer: "irc",
+		streamingArgs: { op: "inbox" },
+		args: { op: "inbox", peek: true },
+		result: {
+			content: [
+				{
+					type: "text",
+					text: [
+						"2 unread message(s):",
+						"- [7181122334455667791] AuthLoader: hub table reads unreadCount — ping me when the bus lands.",
+						"- [7181122334455667792] RateLimiter (reply to 7181122334455667791): bus is in; receipts carry outcome.",
+					].join("\n"),
+				},
+			],
+			details: {
+				op: "inbox",
+				from: "Main",
+				inbox: [
+					{
+						id: "7181122334455667791",
+						from: "AuthLoader",
+						to: "Main",
+						body: "hub table reads unreadCount — ping me when the bus lands.",
+						ts: FIXTURE_NOW - 4 * 60_000,
+					},
+					{
+						id: "7181122334455667792",
+						from: "RateLimiter",
+						to: "Main",
+						body: "bus is in; receipts carry outcome.",
+						ts: FIXTURE_NOW - 60_000,
+						replyTo: "7181122334455667791",
+					},
+				],
+			} satisfies IrcDetails,
+		},
+	},
+
+	irc_list: {
+		label: "IRC (list)",
+		customRendered: true,
+		renderer: "irc",
+		streamingArgs: { op: "list" },
+		args: { op: "list" },
+		result: {
+			content: [
+				{
+					type: "text",
+					text: [
+						"2 peer(s):",
+						"- AuthLoader [task · sub · idle] — parent Main, active 2m ago",
+						"- RateLimiter [task · sub · parked] — unread 2, parent Main, active 12m ago",
+						"",
+						"Parked agents are revived automatically when you message them.",
+					].join("\n"),
+				},
+			],
+			details: {
+				op: "list",
+				from: "Main",
+				peers: [
+					{
+						id: "AuthLoader",
+						displayName: "task",
+						kind: "sub",
+						status: "idle",
+						parentId: "Main",
+						unread: 0,
+						lastActivity: FIXTURE_NOW - 2 * 60_000,
+					},
+					{
+						id: "RateLimiter",
+						displayName: "task",
+						kind: "sub",
+						status: "parked",
+						parentId: "Main",
+						unread: 2,
+						lastActivity: FIXTURE_NOW - 12 * 60_000,
+					},
+				],
+			} satisfies IrcDetails,
 		},
 	},
 

package/src/cli/gallery-fixtures/types.ts

@@ -36,6 +36,11 @@ export interface GalleryFixture {
 	 * real one keeps the gallery honest for these tools.
 	 */
 	customRendered?: boolean;
+	/**
+	 * Renderer-registry key to use when the fixture key is a variant of a tool
+	 * (e.g. `irc_wait` → `irc`). Defaults to the fixture key.
+	 */
+	renderer?: string;
 	/**
 	 * Arguments shown during the streaming state — a partial view of {@link args}
 	 * as if the tool-call JSON were still arriving. May include `__partialJson`

package/src/cli-commands.ts

@@ -38,6 +38,18 @@ export const commands: CommandEntry[] = [
 	{ name: "search", load: () => import("./commands/web-search").then(m => m.default), aliases: ["q"] },
 ];
 
+const RESERVED_TOP_LEVEL_WORDS = new Map<string, string>([
+	[
+		"extensions",
+		'`omp extensions` is not a management command. Use `omp plugin list` / `omp plugin install`, or run `omp launch extensions` if you meant to send "extensions" as a prompt.',
+	],
+]);
+
+export function reservedTopLevelWordMessage(first: string | undefined, argc = 1): string | undefined {
+	if (argc !== 1 || !first || first.startsWith("-") || first.startsWith("@")) return undefined;
+	return RESERVED_TOP_LEVEL_WORDS.get(first);
+}
+
 /**
  * Return true when `first` matches a registered subcommand name or alias.
  *
@@ -48,3 +60,20 @@ export function isSubcommand(first: string | undefined): boolean {
 	if (!first || first.startsWith("-") || first.startsWith("@")) return false;
 	return commands.some(entry => entry.name === first || entry.aliases?.includes(first));
 }
+
+export type ResolvedCliArgv = { argv: string[] } | { error: string };
+
+/**
+ * Decide what the CLI runner should do with raw argv: reject bare reserved
+ * management words, pass help/version through untouched, and route everything
+ * that is not a known subcommand to `launch`.
+ */
+export function resolveCliArgv(argv: string[]): ResolvedCliArgv {
+	const first = argv[0];
+	const reservedMessage = reservedTopLevelWordMessage(first, argv.length);
+	if (reservedMessage) return { error: reservedMessage };
+	if (first === "--help" || first === "-h" || first === "--version" || first === "-v" || first === "help") {
+		return { argv };
+	}
+	return { argv: isSubcommand(first) ? argv : ["launch", ...argv] };
+}

package/src/cli.ts

@@ -43,19 +43,33 @@ async function showHelp(config: CliConfig): Promise<void> {
 	}
 }
 /**
- * Smoke-test entry. Spawns bundled workers, pings them, exits.
+ * Smoke-test entry. Spawns bundled workers, serves the stats dashboard once,
+ * pings everything, then exits.
  *
- * Purpose: catch the silent worker-load regressions that hit compiled
- * binaries (issues #1011 and #1027). Version/help paths do not spawn worker
- * modules on a fresh install, so this probe is the minimal end-to-end test
- * that proves `new Worker(...)` resolves and bundled worker modules evaluate.
+ * Purpose: catch the silent worker-load and bundled-asset regressions that hit
+ * compiled binaries and the npm CLI bundle. Version/help paths do not spawn
+ * worker modules or serve dashboard assets on a fresh install, so this probe is
+ * the minimal end-to-end test that proves those distribution-only paths work.
  * Wired into `scripts/install-tests/run-ci.sh` so binary / source-link /
  * tarball installs all exercise it on every CI run.
  */
 async function runSmokeTest(): Promise<void> {
-	const { smokeTestSyncWorker } = await import("@oh-my-pi/omp-stats");
+	const { smokeTestSyncWorker, startServer } = await import("@oh-my-pi/omp-stats");
 	const { smokeTestTinyTitleWorker } = await import("./tiny/title-client");
 	await smokeTestSyncWorker();
+
+	const statsServer = await startServer(0);
+	try {
+		const response = await fetch(`http://127.0.0.1:${statsServer.port}/`);
+		if (!response.ok) throw new Error(`stats dashboard smoke failed: HTTP ${response.status}`);
+		const html = await response.text();
+		if (!html.includes('<div id="root"></div>') || !html.includes("index.js")) {
+			throw new Error("stats dashboard smoke failed: dashboard HTML was not served");
+		}
+	} finally {
+		statsServer.stop();
+	}
+
 	await smokeTestTinyTitleWorker();
 	process.stdout.write("smoke-test: ok\n");
 }
@@ -161,20 +175,19 @@ export async function runCli(argv: string[]): Promise<void> {
 	if (await runWorkerEntrypoint(argv[0])) {
 		return;
 	}
-	const [{ run }, { commands, isSubcommand }] = await Promise.all([
+	const [{ run }, { commands, resolveCliArgv }] = await Promise.all([
 		import("@oh-my-pi/pi-utils/cli"),
 		import("./cli-commands"),
 	]);
 	// --help and --version are handled by run() directly, don't rewrite those.
 	// Everything else that isn't a known subcommand routes to "launch".
-	const first = argv[0];
-	const runArgv =
-		first === "--help" || first === "-h" || first === "--version" || first === "-v" || first === "help"
-			? argv
-			: isSubcommand(first)
-				? argv
-				: ["launch", ...argv];
-	return run({ bin: APP_NAME, version: VERSION, argv: runArgv, commands, help: showHelp });
+	const resolved = resolveCliArgv(argv);
+	if ("error" in resolved) {
+		process.stderr.write(`error: ${resolved.error}\n`);
+		process.exitCode = 1;
+		return;
+	}
+	return run({ bin: APP_NAME, version: VERSION, argv: resolved.argv, commands, help: showHelp });
 }
 
 // Floating call instead of top-level await: TLA forces `--bytecode` (CJS

package/src/commands/launch.ts

@@ -56,6 +56,10 @@ export default class Index extends Command {
 			description: "Output mode: text (default), json, rpc, or rpc-ui",
 			options: ["text", "json", "rpc", "acp", "rpc-ui"],
 		}),
+		config: Flags.string({
+			description: "Load an extra config.yml-style overlay for this run (repeatable)",
+			multiple: true,
+		}),
 		print: Flags.boolean({
 			char: "p",
 			description: "Non-interactive mode: process prompt and exit",

package/src/commit/agentic/tools/analyze-file.ts

@@ -43,6 +43,9 @@ function buildToolSession(
 		settings: options.settings,
 		authStorage: options.authStorage,
 		modelRegistry: options.modelRegistry,
+		// The task tool no longer takes a per-call schema; the inherited session
+		// schema drives structured output for every spawn from this session.
+		outputSchema: analyzeFileOutputSchema,
 	};
 }
 
@@ -59,29 +62,45 @@ export function createAnalyzeFileTool(options: {
 		label: "Analyze Files",
 		description: "Spawn quick_task agents to analyze files.",
 		parameters: analyzeFileSchema,
-		async execute(toolCallId, params, onUpdate, ctx, signal) {
+		async execute(toolCallId, params, _onUpdate, ctx, signal) {
 			const toolSession = buildToolSession(ctx, options);
+			// The hand-built ToolSession carries no asyncJobManager, so every
+			// execute() below takes the task tool's sync fallback and resolves
+			// with the subagent's result inline — exactly what this flow needs.
+			// The tool's session semaphore bounds the parallel fan-out.
 			const taskTool = await TaskTool.create(toolSession);
 			const numstat = options.state.overview?.numstat ?? [];
-			const tasks = params.files.map((file, index) => {
-				const relatedFiles = formatRelatedFiles(params.files, file, numstat);
-				const assignment = prompt.render(analyzeFilePrompt, {
-					file,
-					goal: params.goal,
-					related_files: relatedFiles,
-				});
-				return {
-					id: `AnalyzeFile${index + 1}`,
-					description: `Analyze ${file}`,
-					assignment,
-				};
-			});
-			const taskParams: TaskParams = {
-				agent: "quick_task",
-				schema: JSON.stringify(analyzeFileOutputSchema),
-				tasks,
+
+			const analyses = await Promise.all(
+				params.files.map((file, index) => {
+					const relatedFiles = formatRelatedFiles(params.files, file, numstat);
+					const assignment = prompt.render(analyzeFilePrompt, {
+						file,
+						goal: params.goal,
+						related_files: relatedFiles,
+					});
+					const taskParams: TaskParams = {
+						agent: "quick_task",
+						id: `AnalyzeFile${index + 1}`,
+						description: `Analyze ${file}`,
+						assignment,
+					};
+					return taskTool.execute(`${toolCallId}-${index + 1}`, taskParams, signal);
+				}),
+			);
+			const results = analyses.flatMap(analysis => analysis.details?.results ?? []);
+			const text = analyses
+				.map(analysis => analysis.content.find(part => part.type === "text")?.text ?? "")
+				.filter(Boolean)
+				.join("\n\n");
+			return {
+				content: [{ type: "text", text: text || "(no output)" }],
+				details: {
+					projectAgentsDir: null,
+					results,
+					totalDurationMs: analyses.reduce((sum, analysis) => sum + (analysis.details?.totalDurationMs ?? 0), 0),
+				},
 			};
-			return taskTool.execute(toolCallId, taskParams, signal, onUpdate);
 		},
 	};
 }

package/src/commit/model-selection.ts

@@ -48,7 +48,7 @@ export async function resolvePrimaryModel(
 	}
 	return {
 		model,
-		apiKey: modelRegistry.resolver(model.provider, { baseUrl: model.baseUrl }),
+		apiKey: modelRegistry.resolver(model.provider, { baseUrl: model.baseUrl, modelId: model.id }),
 		thinkingLevel: resolved?.thinkingLevel,
 	};
 }
@@ -68,6 +68,7 @@ export async function resolveSmolModel(
 				model: resolvedSmol.model,
 				apiKey: modelRegistry.resolver(resolvedSmol.model.provider, {
 					baseUrl: resolvedSmol.model.baseUrl,
+					modelId: resolvedSmol.model.id,
 				}),
 				thinkingLevel: resolvedSmol.thinkingLevel,
 			};
@@ -82,7 +83,7 @@ export async function resolveSmolModel(
 		if (apiKey) {
 			return {
 				model: candidate,
-				apiKey: modelRegistry.resolver(candidate.provider, { baseUrl: candidate.baseUrl }),
+				apiKey: modelRegistry.resolver(candidate.provider, { baseUrl: candidate.baseUrl, modelId: candidate.id }),
 			};
 		}
 	}