pi-free 2.2.4 → 2.2.6

package/providers/tokenrouter/tokenrouter.ts

@@ -49,7 +49,11 @@ import {
 } from "../../lib/provider-compat.ts";
 import { isFreeModel, registerWithGlobalToggle } from "../../lib/registry.ts";
 import { cleanModelName, fetchWithRetry } from "../../lib/util.ts";
-import { enhanceWithCI, setupProvider } from "../../provider-helper.ts";
+import {
+	enhanceWithCI,
+	loadCachedOrFetchModels,
+	setupProvider,
+} from "../../provider-helper.ts";
 
 const _logger = createLogger("tokenrouter");
 
@@ -564,7 +568,9 @@ export default async function tokenRouterProvider(pi: ExtensionAPI) {
 		return;
 	}
 
-	const allModels = await fetchTokenRouterModels(apiKey);
+	const allModels = await loadCachedOrFetchModels(PROVIDER_TOKENROUTER, () =>
+		fetchTokenRouterModels(apiKey),
+	);
 
 	if (allModels.length === 0) {
 		_logger.warn("[tokenrouter] No text chat models available");

package/providers/zenmux/zenmux.ts

@@ -31,7 +31,11 @@ import { safeEnrichModelsWithModelsDev } from "../../lib/model-metadata.ts";
 import { getProxyModelCompat } from "../../lib/provider-compat.ts";
 import { isFreeModel, registerWithGlobalToggle } from "../../lib/registry.ts";
 import { fetchWithRetry } from "../../lib/util.ts";
-import { createReRegister, setupProvider } from "../../provider-helper.ts";
+import {
+	createReRegister,
+	loadCachedOrFetchModels,
+	setupProvider,
+} from "../../provider-helper.ts";
 
 const _logger = createLogger("zenmux");
 
@@ -146,7 +150,9 @@ export default async function zenmuxProvider(pi: ExtensionAPI) {
 	}
 
 	// Fetch models
-	const allModels = await fetchZenmuxModels(apiKey);
+	const allModels = await loadCachedOrFetchModels(PROVIDER_ZENMUX, () =>
+		fetchZenmuxModels(apiKey),
+	);
 
 	if (allModels.length === 0) {
 		_logger.warn("[zenmux] No models available");

package/providers/codestral/codestral.ts

@@ -1,128 +0,0 @@
-/**
- * Codestral Provider Extension
- *
- * Codestral is Mistral AI's code-focused model. This provider registers it
- * through the Codestral-specific endpoint (codestral.mistral.ai) using
- * the Mistral SDK (api: "mistral-conversations") — separate from the built-in
- * "mistral" provider which uses api.mistral.ai.
- *
- * NOTE: Do NOT use api: "openai-completions" here. Codestral's API is
- * Mistral-format (camelCase fields, maxTokens, no stream_options/store).
- * The OpenAI completions adapter sends OpenAI-specific fields that Mistral
- * rejects with HTTP 422 "Extra inputs are not permitted".
- *
- * Free tier (Experiment plan):
- *   - 2 req/min, 500K tokens/min, 1B tokens/month
- *   - No credit card — phone verification only
- *   - Sign up at https://console.mistral.ai/codestral
- *
- * Endpoints:
- *   Chat:  https://codestral.mistral.ai/v1/chat/completions
- *   FIM:   https://codestral.mistral.ai/v1/fim/completions (not used by pi)
- *
- * Setup:
- *   1. Get API key from https://console.mistral.ai/codestral
- *   2. Set CODESTRAL_API_KEY env var (or MISTRAL_API_KEY as fallback)
- *
- * Usage:
- *   pi install git:github.com/apmantza/pi-free
- *   # Set CODESTRAL_API_KEY env var
- *   # Models appear in /model selector as "codestral/codestral-latest"
- */
-
-import type {
-	ExtensionAPI,
-	ProviderModelConfig,
-} from "@earendil-works/pi-coding-agent";
-import {
-	getCodestralApiKey,
-	getCodestralShowPaid,
-	getMistralApiKey,
-} from "../../config.ts";
-import { BASE_URL_CODESTRAL, PROVIDER_CODESTRAL } from "../../constants.ts";
-import { createLogger } from "../../lib/logger.ts";
-import { registerWithGlobalToggle } from "../../lib/registry.ts";
-import { enhanceWithCI, setupProvider } from "../../provider-helper.ts";
-
-const _logger = createLogger("codestral");
-
-// =============================================================================
-// Model Definition
-// =============================================================================
-
-const CODESTRAL_MODEL: ProviderModelConfig = {
-	id: "codestral-latest",
-	name: "Codestral",
-	reasoning: false,
-	input: ["text"],
-	cost: {
-		input: 0.3,
-		output: 0.9,
-		cacheRead: 0,
-		cacheWrite: 0,
-	},
-	contextWindow: 256_000,
-	maxTokens: 4_096,
-};
-
-// =============================================================================
-// Extension Entry Point
-// =============================================================================
-
-export default async function codestralProvider(pi: ExtensionAPI) {
-	// Try CODESTRAL_API_KEY first, fall back to MISTRAL_API_KEY
-	const apiKey = getCodestralApiKey() || getMistralApiKey();
-
-	if (!apiKey) {
-		_logger.info(
-			"[codestral] Skipping — neither CODESTRAL_API_KEY nor MISTRAL_API_KEY set",
-		);
-		return;
-	}
-
-	const keySource = getCodestralApiKey()
-		? "CODESTRAL_API_KEY"
-		: "MISTRAL_API_KEY";
-	_logger.info(`[codestral] Using key from ${keySource}`);
-
-	const allModels = [CODESTRAL_MODEL];
-	const freeModels = allModels; // All $0.30/$0.90 — still accessible via Experiment free tier
-	const stored = { free: freeModels, all: allModels };
-
-	// Re-register function — uses mistral-conversations API (Mistral SDK)
-	// NOT openai-completions: Codestral uses the same API format as Mistral
-	// and rejects OpenAI-specific fields (stream_options, store, max_completion_tokens) with 422.
-	const reRegister = (models: typeof freeModels) => {
-		pi.registerProvider(PROVIDER_CODESTRAL, {
-			baseUrl: BASE_URL_CODESTRAL,
-			apiKey,
-			api: "mistral-conversations" as const,
-			models: enhanceWithCI(models, PROVIDER_CODESTRAL),
-		});
-	};
-
-	// Register with global toggle
-	registerWithGlobalToggle(PROVIDER_CODESTRAL, stored, reRegister, true);
-
-	// Setup provider (toggle command, status bar, error handling)
-	setupProvider(
-		pi,
-		{
-			providerId: PROVIDER_CODESTRAL,
-			initialShowPaid: getCodestralShowPaid(),
-			skipToggle: true, // Only one model, no toggle needed
-			reRegister: (models) => {
-				stored.free = models;
-				stored.all = models;
-				reRegister(models);
-			},
-		},
-		stored,
-	);
-
-	// Initial registration — uses mistral-conversations API (Mistral SDK)
-	reRegister(freeModels);
-
-	_logger.info(`[codestral] Registered codestral-latest via ${keySource}`);
-
-}

package/providers/qoder/encoding.ts

@@ -1,48 +0,0 @@
-/**
- * Qoder WAF bypass body encoding.
- *
- * Qoder's API uses a custom base64 variant with a scrambled alphabet and
- * rearranged segments to evade WAF detection. This is the same encoding
- * used by the official qodercli binary.
- *
- * The algorithm:
- *   1. Take the standard base64 of the input
- *   2. Rearrange segments: last N/3 chars + middle N-2N/3 chars + first N/3 chars
- *   3. Substitute each character through a custom alphabet
- *   4. Replace '=' padding with '$'
- */
-
-const QODER_CUSTOM_ALPHABET =
-	"_doRTgHZBKcGVjlvpC,@aFSx#DPuNJme&i*MzLOEn)sUrthbf%Y^w.(kIQyXqWA!";
-const QODER_STD_ALPHABET =
-	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
-
-/**
- * Encode a body string or buffer using Qoder's custom WAF-bypass encoding.
- *
- * @param plaintext - String or Buffer to encode
- * @returns Encoded string ready for transmission
- */
-export function qoderEncodeBody(plaintext: string | Buffer): string {
-	const std = Buffer.isBuffer(plaintext)
-		? plaintext.toString("base64")
-		: Buffer.from(plaintext).toString("base64");
-	const n = std.length;
-	const a = Math.floor(n / 3);
-	const rearranged = std.slice(n - a) + std.slice(a, n - a) + std.slice(0, a);
-	let out = "";
-	for (let i = 0; i < n; i++) {
-		const c = rearranged[i];
-		if (c === "=") {
-			out += "$";
-		} else {
-			const idx = QODER_STD_ALPHABET.indexOf(c);
-			if (idx >= 0) {
-				out += QODER_CUSTOM_ALPHABET[idx];
-			} else {
-				out += c;
-			}
-		}
-	}
-	return out;
-}