@oh-my-pi/pi-coding-agent 15.10.9 → 15.10.11

package/src/async/job-manager.ts

@@ -23,6 +23,12 @@ export interface AsyncJob {
 	 * supply an id (e.g. legacy tests, SDK consumers without an agent context).
 	 */
 	ownerId?: string;
+	/**
+	 * Job is registered but parked behind a caller-managed gate (e.g. a task
+	 * batch semaphore). Queued jobs do not count toward the running-job limit
+	 * until the caller invokes `markRunning()` from the run context.
+	 */
+	queued?: boolean;
 }
 
 export interface AsyncJobManagerOptions {
@@ -53,6 +59,8 @@ export interface AsyncJobRegisterOptions {
 	/** Registry id of the agent that owns this job; used to scope cancelAll. */
 	ownerId?: string;
 	onProgress?: (text: string, details?: Record<string, unknown>) => void | Promise<void>;
+	/** Register the job in queued state; see {@link AsyncJob.queued}. */
+	queued?: boolean;
 }
 
 /**
@@ -110,6 +118,17 @@ export class AsyncJobManager {
 		this.#retentionMs = Math.max(0, Math.floor(options.retentionMs ?? DEFAULT_RETENTION_MS));
 	}
 
+	/** True when the running-job count has reached the configured cap. */
+	get atCapacity(): boolean {
+		if (this.#disposed) return true;
+		// Mirror register(): queued jobs hold no execution slot.
+		let activeCount = 0;
+		for (const job of this.#jobs.values()) {
+			if (job.status === "running" && !job.queued) activeCount++;
+		}
+		return activeCount >= this.#maxRunningJobs;
+	}
+
 	register(
 		type: "bash" | "task",
 		label: string,
@@ -117,14 +136,21 @@ export class AsyncJobManager {
 			jobId: string;
 			signal: AbortSignal;
 			reportProgress: (text: string, details?: Record<string, unknown>) => Promise<void>;
+			/** Clear the queued flag once the job actually starts executing. */
+			markRunning: () => void;
 		}) => Promise<string>,
 		options?: AsyncJobRegisterOptions,
 	): string {
 		if (this.#disposed) {
 			throw new Error("Async job manager is disposed");
 		}
-		const runningCount = this.getRunningJobs().length;
-		if (runningCount >= this.#maxRunningJobs) {
+		// Queued jobs hold no execution slot yet — only count jobs that are
+		// actually running so a large parked batch cannot starve registration.
+		let activeCount = 0;
+		for (const existing of this.#jobs.values()) {
+			if (existing.status === "running" && !existing.queued) activeCount++;
+		}
+		if (activeCount >= this.#maxRunningJobs) {
 			throw new Error(
 				`Background job limit reached (${this.#maxRunningJobs}). Wait for running jobs to finish or cancel one.`,
 			);
@@ -144,6 +170,7 @@ export class AsyncJobManager {
 			abortController,
 			promise: Promise.resolve(),
 			ownerId: options?.ownerId,
+			queued: options?.queued === true,
 		};
 
 		const reportProgress = async (text: string, details?: Record<string, unknown>): Promise<void> => {
@@ -159,7 +186,14 @@ export class AsyncJobManager {
 		};
 		job.promise = (async () => {
 			try {
-				const text = await run({ jobId: id, signal: abortController.signal, reportProgress });
+				const text = await run({
+					jobId: id,
+					signal: abortController.signal,
+					reportProgress,
+					markRunning: () => {
+						job.queued = false;
+					},
+				});
 				if (job.status === "cancelled") {
 					job.resultText = text;
 					this.#scheduleEviction(id);
@@ -278,6 +312,26 @@ export class AsyncJobManager {
 		return before - this.#deliveries.length;
 	}
 
+	/**
+	 * Lift a foreground-wait suppression set via `acknowledgeDeliveries`. If the
+	 * job already finished while suppressed (its delivery enqueue was skipped),
+	 * re-enqueue the completion so the result is still delivered exactly once.
+	 */
+	resumeDeliveries(jobIds: string[]): void {
+		for (const rawId of jobIds) {
+			const jobId = rawId.trim();
+			if (!jobId) continue;
+			if (!this.#suppressedDeliveries.delete(jobId)) continue;
+			const job = this.#jobs.get(jobId);
+			if (!job || (job.status !== "completed" && job.status !== "failed")) continue;
+			const queued =
+				this.#deliveries.some(delivery => delivery.jobId === jobId) ||
+				this.#inFlightDeliveries.some(delivery => delivery.jobId === jobId);
+			if (queued) continue;
+			this.#enqueueDelivery(jobId, job.status === "completed" ? (job.resultText ?? "") : (job.errorText ?? ""));
+		}
+	}
+
 	/**
 	 * Cancel running jobs. With `filter.ownerId` set, cancels only jobs the
 	 * matching agent registered; with no filter, cancels every running job

package/src/autoresearch/dashboard.ts

@@ -66,7 +66,7 @@ export function createDashboardController(): DashboardController {
 
 					let scrollOffset = 0;
 					return {
-						render(width: number): string[] {
+						render(width: number): readonly string[] {
 							const terminalRows = process.stdout.rows ?? 40;
 							const header = renderExpandedHeader(runtime, width, theme);
 							const body = renderDashboardLines(runtime, width, theme, 0);

package/src/autoresearch/prompt-setup.md

@@ -18,16 +18,16 @@ Working directory: `{{working_dir}}`
 {{baseline_warning}}
 {{/if}}
 
-### What you must produce
+### What you MUST produce
 
-Write `./autoresearch.sh` at the working directory. It is the canonical benchmark entrypoint and must:
+Write `./autoresearch.sh` at the working directory. It is the canonical benchmark entrypoint and MUST:
 
 - exit 0 on success and non-zero on failure;
 - print the primary metric as a single line `METRIC <name>=<value>`;
 - print any secondary metrics as additional `METRIC <name>=<value>` lines;
 - run the same workload deterministically every time (no live network, no time-of-day dependencies, fixed seeds where applicable).
 
-You **may** edit anything else needed to make `autoresearch.sh` work — benchmark binaries, `Cargo.toml`, `package.json`, helper scripts, fixtures. All those edits are part of the harness baseline and will be committed for you when you call `init_experiment` on an autoresearch branch.
+You MAY edit anything else needed to make `autoresearch.sh` work — benchmark binaries, `Cargo.toml`, `package.json`, helper scripts, fixtures. All those edits are part of the harness baseline and will be committed for you when you call `init_experiment` on an autoresearch branch.
 
 ### Steps
 
@@ -38,6 +38,6 @@ You **may** edit anything else needed to make `autoresearch.sh` work — benchma
 
 ### Rules
 
-- Do **not** call `run_experiment`, `log_experiment`, or `update_notes` yet. They will error with "no active autoresearch session" until `init_experiment` runs.
-- Do **not** treat a compile-only check as a benchmark. The harness must actually execute the workload and emit `METRIC`.
-- Do **not** create `autoresearch.md`, `autoresearch.checks.sh`, `autoresearch.program.md`, `autoresearch.ideas.md`, `autoresearch.jsonl`, `.autoresearch/`, or `autoresearch.config.json`. Session state is tracked for you.
+- NEVER call `run_experiment`, `log_experiment`, or `update_notes` yet. They will error with "no active autoresearch session" until `init_experiment` runs.
+- NEVER treat a compile-only check as a benchmark. The harness MUST actually execute the workload and emit `METRIC`.
+- NEVER create `autoresearch.md`, `autoresearch.checks.sh`, `autoresearch.program.md`, `autoresearch.ideas.md`, `autoresearch.jsonl`, `.autoresearch/`, or `autoresearch.config.json`. Session state is tracked for you.

package/src/autoresearch/prompt.md

@@ -11,17 +11,17 @@ Primary goal:
 There is no goal recorded for this session yet. Infer what to optimize from the latest user message and the conversation; capture the goal in your notes (`update_notes`) once it is clear.
 {{/if}}
 
-Session state and run artifacts are managed for you. The benchmark entrypoint is `bash autoresearch.sh` (committed during Phase 1). Do not edit `autoresearch.sh` mid-segment unless you intentionally bump segment via `init_experiment new_segment: true`. Do not create `autoresearch.md` or `.autoresearch/` in this repo.
+Session state and run artifacts are managed for you. The benchmark entrypoint is `bash autoresearch.sh` (committed during Phase 1). NEVER edit `autoresearch.sh` mid-segment unless you intentionally bump segment via `init_experiment new_segment: true`. NEVER create `autoresearch.md` or `.autoresearch/` in this repo.
 
 Working directory: `{{working_dir}}`
 {{#if has_branch}}Active branch: `{{branch}}`{{/if}}
 {{#if has_baseline_commit}}Baseline commit: `{{baseline_commit}}`{{/if}}
 
-You are running an autonomous experiment loop. Keep iterating until the user interrupts you or the configured maximum iteration count is reached.
+You are running an autonomous experiment loop. You MUST keep iterating until the user interrupts you or the configured maximum iteration count is reached.
 
 ### Available tools
 - `init_experiment` — open or reconfigure the session. Pass `new_segment: true` to start a fresh baseline within the current session.
-- `run_experiment` — run the benchmark (`bash autoresearch.sh`). Output is captured automatically and `METRIC name=value` / `ASI key=value` lines printed by the harness are parsed back to you. The command is fixed; if you need a different workload, edit `autoresearch.sh` and bump segment via `init_experiment new_segment: true`.
+- `run_experiment` — run the benchmark (`bash autoresearch.sh`). Output is captured automatically and `METRIC name=value` / `ASI key=value` lines printed by the harness are parsed back to you. The command is fixed.
 - `log_experiment` — record the result. On `keep`, modified files are committed for you; on `discard`/`crash`/`checks_failed`, the worktree is reverted. Pass `flag_runs` to mark earlier runs as suspect; flagged runs are excluded from baseline and best-metric math.
 - `update_notes` — replace the durable session playbook (`body`) or append to the ideas backlog (`append_idea`). The notes are injected into your system prompt every iteration.
 
@@ -97,7 +97,7 @@ Finish the `log_experiment` step before starting another benchmark.
 {{/if}}
 
 ### Guardrails
-- Do not game the benchmark.
-- Do not overfit to synthetic inputs if the real workload is broader.
-- Preserve correctness.
+- NEVER game the benchmark.
+- NEVER overfit to synthetic inputs if the real workload is broader.
+- MUST preserve correctness.
 - If the user sends another message while a run is in progress, finish the current run and logging cycle first, then address the new input in the next iteration.

package/src/capability/fs.ts

@@ -15,6 +15,16 @@ export async function readFile(filePath: string): Promise<string | null> {
 	}
 
 	try {
+		// Gate on the file type first: discovery scans foreign config dirs
+		// (~/.claude, ~/.cursor, project trees), and reading a FIFO/socket/char
+		// device with `.text()` blocks until EOF — i.e. forever — hanging
+		// startup with zero output. `stat` follows symlinks, so symlinked
+		// context files (CLAUDE.md -> AGENTS.md) still resolve.
+		const stats = await fs.promises.stat(abs);
+		if (!stats.isFile()) {
+			contentCache.set(abs, null);
+			return null;
+		}
 		const content = await Bun.file(abs).text();
 		contentCache.set(abs, content);
 		return content;

package/src/cli/args.ts

@@ -1,7 +1,7 @@
 /**
  * CLI argument parsing and help display
  */
-import { type Effort, THINKING_EFFORTS } from "@oh-my-pi/pi-ai/effort";
+import { type Effort, THINKING_EFFORTS } from "@oh-my-pi/pi-catalog/effort";
 import { APP_NAME, CONFIG_DIR_NAME, logger } from "@oh-my-pi/pi-utils";
 import chalk from "chalk";
 import { parseEffort } from "../thinking";

package/src/cli/auth-gateway-cli.ts

@@ -24,14 +24,12 @@ import {
 	type CredentialCompletionResult,
 	completeSimple,
 	DEFAULT_AUTH_GATEWAY_BIND,
-	type GeneratedProvider,
-	getBundledModels,
-	getBundledProviders,
 	type Model,
 	RemoteAuthCredentialStore,
 	type SnapshotResponse,
 	startAuthGateway,
 } from "@oh-my-pi/pi-ai";
+import { type GeneratedProvider, getBundledModels, getBundledProviders } from "@oh-my-pi/pi-catalog/models";
 import { getConfigRootDir, isEnoent, VERSION } from "@oh-my-pi/pi-utils";
 import chalk from "chalk";
 import { type AuthBrokerClientConfig, resolveAuthBrokerConfig } from "../session/auth-broker-config";

package/src/cli/dry-balance-cli.ts

@@ -12,10 +12,10 @@ import type {
 	SimpleStreamOptions,
 } from "@oh-my-pi/pi-ai";
 import { streamSimple } from "@oh-my-pi/pi-ai";
+import type { CanonicalModelVariant } from "@oh-my-pi/pi-catalog/identity";
 import { replaceTabs, truncateToWidth } from "@oh-my-pi/pi-tui";
 import { formatDuration, getProjectDir } from "@oh-my-pi/pi-utils";
 import chalk from "chalk";
-import type { CanonicalModelVariant } from "../config/model-equivalence";
 import { type CanonicalModelQueryOptions, ModelRegistry } from "../config/model-registry";
 import {
 	formatModelString,

package/src/cli/gallery-cli.ts

@@ -104,7 +104,7 @@ export async function renderGalleryState(
 	state: GalleryState,
 	width: number,
 	expanded = false,
-): Promise<string[]> {
+): Promise<readonly string[]> {
 	if (fixture.renderState) {
 		return await fixture.renderState(state, width, expanded);
 	}

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

@@ -56,7 +56,7 @@ function addGroupedReadArgs(component: ReadToolGroupComponent): void {
 	component.updateArgs({ path: groupedReadRepeatedRanges }, "read-ranges");
 }
 
-function renderReadGroupFixtureState(state: GalleryFixtureState, width: number, expanded: boolean): string[] {
+function renderReadGroupFixtureState(state: GalleryFixtureState, width: number, expanded: boolean): readonly string[] {
 	const component = new ReadToolGroupComponent();
 	component.setExpanded(expanded);
 

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

@@ -22,7 +22,11 @@ export interface GalleryFixture {
 	 * Custom gallery-only renderer for fixtures that are not one ToolExecutionComponent
 	 * (for example the read-group transcript component).
 	 */
-	renderState?: (state: GalleryFixtureState, width: number, expanded: boolean) => string[] | Promise<string[]>;
+	renderState?: (
+		state: GalleryFixtureState,
+		width: number,
+		expanded: boolean,
+	) => readonly string[] | Promise<readonly string[]>;
 	/**
 	 * Set for tools whose real `AgentTool` attaches `renderCall`/`renderResult`
 	 * directly on the instance (e.g. `task`). The harness then attaches

package/src/cli/list-models.ts

@@ -1,7 +1,8 @@
 /**
  * List available models with optional fuzzy search
  */
-import { type Api, getSupportedEfforts, type Model } from "@oh-my-pi/pi-ai";
+import type { Api, Model } from "@oh-my-pi/pi-ai";
+import { getSupportedEfforts } from "@oh-my-pi/pi-catalog/model-thinking";
 import { fuzzyFilter } from "@oh-my-pi/pi-tui";
 import { formatNumber } from "@oh-my-pi/pi-utils";
 import type { ModelRegistry } from "../config/model-registry";
@@ -64,22 +65,16 @@ export async function listModels(modelRegistry: ModelRegistry, searchPattern?: s
 	}
 
 	const filteredCanonical = modelRegistry
-		.getCanonicalModels({ availableOnly: true, candidates: filteredModels })
-		.map(record => {
-			const selected = modelRegistry.resolveCanonicalModel(record.id, {
-				availableOnly: true,
-				candidates: filteredModels,
-			});
-			if (!selected) return undefined;
-			return {
+		.getCanonicalModelSelections({ availableOnly: true, candidates: filteredModels })
+		.map(
+			({ record, model: selected }): CanonicalRow => ({
 				canonical: record.id,
 				selected: `${selected.provider}/${selected.id}`,
 				variants: String(record.variants.length),
 				context: formatNumber(selected.contextWindow),
 				maxOut: formatNumber(selected.maxTokens),
-			} satisfies CanonicalRow;
-		})
-		.filter((row): row is CanonicalRow => row !== undefined)
+			}),
+		)
 		.sort((left, right) => left.canonical.localeCompare(right.canonical));
 
 	if (filteredModels.length === 0 && filteredCanonical.length === 0) {