@oh-my-pi/pi-coding-agent 15.12.3 → 15.13.0

package/dist/types/stt/setup.d.ts

@@ -1,18 +0,0 @@
-export interface STTDependencyStatus {
-    recorder: {
-        available: boolean;
-        tool: string | null;
-        installHint: string;
-    };
-    python: {
-        available: boolean;
-        path: string | null;
-        installHint: string;
-    };
-    whisper: {
-        available: boolean;
-        installHint: string;
-    };
-}
-export declare function checkDependencies(): Promise<STTDependencyStatus>;
-export declare function formatDependencyStatus(status: STTDependencyStatus): string;

package/src/cli/list-models.ts

@@ -1,194 +0,0 @@
-/**
- * List available models with optional fuzzy search
- */
-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";
-import { discoverAndLoadExtensions, loadExtensions } from "../extensibility/extensions";
-import { EventBus } from "../utils/event-bus";
-
-interface ProviderRow {
-	provider: string;
-	model: string;
-	context: string;
-	maxOut: string;
-	thinking: string;
-	images: string;
-}
-
-interface CanonicalRow {
-	canonical: string;
-	selected: string;
-	variants: string;
-	context: string;
-	maxOut: string;
-}
-
-function writeLine(line = ""): void {
-	process.stdout.write(`${line}\n`);
-}
-
-function renderTable<T extends Record<string, string>>(rows: T[], headers: T): void {
-	const widths = Object.fromEntries(
-		Object.keys(headers).map(key => [key, Math.max(headers[key]!.length, ...rows.map(row => row[key]!.length))]),
-	) as Record<keyof T, number>;
-
-	const headerLine = Object.keys(headers)
-		.map(key => headers[key as keyof T]!.padEnd(widths[key as keyof T]))
-		.join("  ");
-	writeLine(headerLine);
-
-	for (const row of rows) {
-		const line = Object.keys(headers)
-			.map(key => row[key as keyof T]!.padEnd(widths[key as keyof T]))
-			.join("  ");
-		writeLine(line);
-	}
-}
-
-/**
- * List available models, optionally filtered by search pattern
- */
-export async function listModels(modelRegistry: ModelRegistry, searchPattern?: string): Promise<void> {
-	const models = modelRegistry.getAvailable();
-
-	if (models.length === 0) {
-		writeLine("No models available. Set API keys in environment variables.");
-		return;
-	}
-
-	let filteredModels: Model<Api>[] = models;
-	if (searchPattern) {
-		filteredModels = fuzzyFilter(models, searchPattern, model => `${model.provider} ${model.id}`);
-	}
-
-	const filteredCanonical = modelRegistry
-		.getCanonicalModelSelections({ availableOnly: true, candidates: filteredModels })
-		.map(
-			({ record, model: selected }): CanonicalRow => ({
-				canonical: record.id,
-				selected: `${selected.provider}/${selected.id}`,
-				variants: String(record.variants.length),
-				context: formatNumber(selected.contextWindow),
-				maxOut: formatNumber(selected.maxTokens),
-			}),
-		)
-		.sort((left, right) => left.canonical.localeCompare(right.canonical));
-
-	if (filteredModels.length === 0 && filteredCanonical.length === 0) {
-		writeLine(`No models matching "${searchPattern}"`);
-		return;
-	}
-
-	filteredModels.sort((left, right) => {
-		const providerCmp = left.provider.localeCompare(right.provider);
-		if (providerCmp !== 0) return providerCmp;
-		return left.id.localeCompare(right.id);
-	});
-
-	const providerRows = filteredModels.map(model => ({
-		provider: model.provider,
-		model: model.id,
-		context: formatNumber(model.contextWindow),
-		maxOut: formatNumber(model.maxTokens),
-		thinking: model.thinking ? getSupportedEfforts(model).join(",") : model.reasoning ? "yes" : "-",
-		images: model.input.includes("image") ? "yes" : "no",
-	})) satisfies ProviderRow[];
-
-	if (filteredCanonical.length > 0) {
-		writeLine("Canonical models");
-		renderTable(filteredCanonical, {
-			canonical: "canonical",
-			selected: "selected",
-			variants: "variants",
-			context: "context",
-			maxOut: "max-out",
-		});
-		if (providerRows.length > 0) {
-			writeLine();
-		}
-	}
-
-	if (providerRows.length > 0) {
-		writeLine("Provider models");
-		renderTable(providerRows, {
-			provider: "provider",
-			model: "model",
-			context: "context",
-			maxOut: "max-out",
-			thinking: "thinking",
-			images: "images",
-		});
-	}
-}
-
-/**
- * Options for the `--list-models` command entry point.
- */
-export interface RunListModelsOptions {
-	modelRegistry: ModelRegistry;
-	cwd: string;
-	/** CLI-supplied extension paths (e.g. from `-e <path>`). */
-	additionalExtensionPaths?: string[];
-	/** Extension paths configured under `extensions:` in user settings. */
-	settingsExtensions?: string[];
-	/** Disabled extension ids from settings (`disabledExtensions`). */
-	disabledExtensionIds?: string[];
-	/** When true, skip discovery and only load `additionalExtensionPaths`. */
-	disableExtensionDiscovery?: boolean;
-	searchPattern?: string;
-}
-
-/**
- * Loads extensions (CLI `-e` paths and `settings.extensions`) and surfaces
- * any provider/model registrations on the supplied `modelRegistry` before
- * delegating to {@link listModels}. This is the single entry point used by
- * `--list-models` and exists to ensure extension-contributed providers are
- * visible in the listing (issue #905). The load is intentionally narrow:
- * no agent loop, no MCP servers, no custom-tool registration.
- */
-export async function runListModelsCommand(options: RunListModelsOptions): Promise<void> {
-	const {
-		modelRegistry,
-		cwd,
-		additionalExtensionPaths = [],
-		settingsExtensions = [],
-		disabledExtensionIds = [],
-		disableExtensionDiscovery = false,
-		searchPattern,
-	} = options;
-
-	const eventBus = new EventBus();
-	const extensionsResult = disableExtensionDiscovery
-		? await loadExtensions(additionalExtensionPaths, cwd, eventBus)
-		: await discoverAndLoadExtensions(
-				[...additionalExtensionPaths, ...settingsExtensions],
-				cwd,
-				eventBus,
-				disabledExtensionIds,
-			);
-
-	for (const { path: extPath, error } of extensionsResult.errors) {
-		process.stderr.write(`Failed to load extension: ${extPath}: ${error}\n`);
-	}
-
-	// Mirror sdk.ts: drain pending provider registrations into the registry.
-	const activeSources = extensionsResult.extensions.map(extension => extension.path);
-	modelRegistry.syncExtensionSources(activeSources);
-	for (const sourceId of new Set(activeSources)) {
-		modelRegistry.clearSourceRegistrations(sourceId);
-	}
-	for (const { name, config, sourceId } of extensionsResult.runtime.pendingProviderRegistrations) {
-		modelRegistry.registerProvider(name, config, sourceId);
-	}
-	extensionsResult.runtime.pendingProviderRegistrations = [];
-	// Discover runtime (extension) provider catalogs now that they are registered.
-	// The full refresh in main.ts ran before extensions loaded, so this is the only
-	// point where extension-contributed dynamic providers get discovered. Cache-aware
-	// so it reuses the shared 24 h model cache instead of refetching every invocation.
-	await modelRegistry.refreshRuntimeProviders("online-if-uncached");
-
-	await listModels(modelRegistry, searchPattern);
-}

package/src/stt/setup.ts

@@ -1,52 +0,0 @@
-import { detectRecordingTools } from "./recorder";
-import { resolvePython } from "./transcriber";
-
-const isWindows = process.platform === "win32";
-
-export interface STTDependencyStatus {
-	recorder: { available: boolean; tool: string | null; installHint: string };
-	python: { available: boolean; path: string | null; installHint: string };
-	whisper: { available: boolean; installHint: string };
-}
-
-export async function checkDependencies(): Promise<STTDependencyStatus> {
-	const recorderTools = detectRecordingTools();
-	const recorderHint = isWindows
-		? "PowerShell fallback available. For better quality: install SoX or FFmpeg."
-		: "Install SoX: sudo apt install sox, or FFmpeg: sudo apt install ffmpeg";
-
-	const pythonCmd = resolvePython();
-	const pythonHint = "Install Python 3.8+ from https://python.org";
-
-	let whisperAvailable = false;
-	if (pythonCmd) {
-		const check = Bun.spawnSync([pythonCmd, "-c", "import whisper"], {
-			stdout: "pipe",
-			stderr: "pipe",
-		});
-		whisperAvailable = check.exitCode === 0;
-	}
-	const whisperHint = "Run 'omp setup stt' to auto-install, or: pip install openai-whisper";
-
-	return {
-		recorder: { available: recorderTools.length > 0, tool: recorderTools[0] ?? null, installHint: recorderHint },
-		python: { available: pythonCmd !== null, path: pythonCmd, installHint: pythonHint },
-		whisper: { available: whisperAvailable, installHint: whisperHint },
-	};
-}
-
-export function formatDependencyStatus(status: STTDependencyStatus): string {
-	const lines: string[] = ["STT Dependencies:"];
-	const check = (ok: boolean) => (ok ? "[ok]" : "[missing]");
-
-	lines.push(`  Recorder: ${check(status.recorder.available)} ${status.recorder.tool ?? "none"}`);
-	if (!status.recorder.available) lines.push(`    -> ${status.recorder.installHint}`);
-
-	lines.push(`  Python:   ${check(status.python.available)} ${status.python.path ?? "none"}`);
-	if (!status.python.available) lines.push(`    -> ${status.python.installHint}`);
-
-	lines.push(`  Whisper:  ${check(status.whisper.available)}`);
-	if (!status.whisper.available) lines.push(`    -> ${status.whisper.installHint}`);
-
-	return lines.join("\n");
-}

package/src/stt/transcribe.py

@@ -1,70 +0,0 @@
-"""Transcribe a WAV file using openai-whisper.
-
-Reads WAV directly via Python's wave module (no ffmpeg needed).
-Resamples to 16kHz mono float32 and passes to whisper as a numpy array.
-
-Usage: python transcribe.py <audio.wav> <model_name> <language>
-Prints transcribed text to stdout.
-"""
-
-import sys
-import wave
-import re
-
-
-import numpy as np
-import whisper
-
-
-def load_wav(path: str) -> np.ndarray:
-    with wave.open(path, "rb") as wf:
-        rate = wf.getframerate()
-        channels = wf.getnchannels()
-        width = wf.getsampwidth()
-        n_frames = wf.getnframes()
-        raw = wf.readframes(n_frames)
-
-    if width == 2:
-        audio = np.frombuffer(raw, dtype=np.int16).astype(np.float32) / 32768.0
-    elif width == 1:
-        audio = (np.frombuffer(raw, dtype=np.uint8).astype(np.float32) - 128.0) / 128.0
-    elif width == 4:
-        audio = np.frombuffer(raw, dtype=np.int32).astype(np.float32) / 2147483648.0
-    else:
-        raise ValueError(f"Unsupported sample width: {width}")
-
-    # Mix to mono
-    if channels > 1:
-        audio = audio.reshape(-1, channels).mean(axis=1)
-
-    # Resample to 16 kHz
-    if rate != 16000:
-        target_len = int(len(audio) * 16000 / rate)
-        audio = np.interp(
-            np.linspace(0, len(audio) - 1, target_len),
-            np.arange(len(audio)),
-            audio,
-        ).astype(np.float32)
-
-    return audio
-
-
-def main() -> None:
-    if len(sys.argv) < 2:
-        print("Usage: python transcribe.py <audio.wav> <model_name> <language>", file=sys.stderr)
-        sys.exit(1)
-    audio_path = sys.argv[1]
-    model_name = sys.argv[2] if len(sys.argv) > 2 else "base.en"
-    language = sys.argv[3] if len(sys.argv) > 3 else "en"
-    if not re.fullmatch(r"[A-Za-z]{2,3}(-[A-Za-z]{2})?", language):
-        print(f"Invalid language code: {language}", file=sys.stderr)
-        sys.exit(1)
-
-    audio = load_wav(audio_path)
-    model = whisper.load_model(model_name)
-    result = model.transcribe(audio, language=language)
-    print(result["text"].strip())
-
-
-if __name__ == "__main__":
-    main()