pi-multikey 1.4.0 → 1.5.0

package/README.md

@@ -50,6 +50,8 @@ Built-in presets decouple "model settings" from "keys". The data comes from b.ai
 
 Endpoint `https://opencode.ai/zen/v1`; keys from [opencode.ai/auth](https://opencode.ai/auth) → workspace Keys. Context / max-output are the **Zen free-tier serving limits** (consistent across models.dev `opencode` provider + pi's built-in opencode catalog); the raw models are bigger — MiMo V2.5 = 1M ctx, Hy3 = 262K ctx. `muse-spark-1.2-contributor-free` uses the OpenAI **Responses** API endpoint; the other seven use chat completions.
 
+Requests to this endpoint automatically carry the `User-Agent` header that opencode.ai expects (`opencode/1.15.0 ai-sdk/provider-utils/4.0.23 runtime/bun/1.3.13`) — appended whenever a pool's baseUrl is `https://opencode.ai/zen/v1`, on both probes and live requests.
+
 | Model | ctx / max-out | Modalities | Supported thinking levels |
 |---|---|---|---|
 | big-pickle | 200K / 32K | text | always-on (no thinkingLevelMap, like pi's catalog) |

package/README.zh.md

@@ -56,6 +56,8 @@ DeepSeek / Tencent / 小米官方文档，并对每个 thinking 档位做过实
 原始模型更大——MiMo V2.5 = 1M ctx，Hy3 = 262K ctx。
 `muse-spark-1.2-contributor-free` 使用 OpenAI **Responses** API；其余七个使用 chat completions。
 
+对该端点的请求会自动携带 opencode.ai 期望的 `User-Agent` 头（`opencode/1.15.0 ai-sdk/provider-utils/4.0.23 runtime/bun/1.3.13`）——只要 pool 的 baseUrl 是 `https://opencode.ai/zen/v1`，探测和真实请求都会附加。
+
 | 模型 | ctx / max-out | 模态 | 生效 thinking 档位 |
 |---|---|---|---|
 | big-pickle | 200K / 32K | text | 始终思考（无 thinkingLevelMap，与 pi 内置目录一致） |

package/config.ts

@@ -339,3 +339,24 @@ export function maskKey(key: string): string {
 	if (key.length <= 10) return "…";
 	return `${key.slice(0, 6)}…${key.slice(-4)}`;
 }
+
+// ── Endpoint-required headers ────────────────────────────────────────────────
+
+/** OpenCode Zen free tier endpoint that requires a specific User-Agent header. */
+const OPENCODE_ZEN_BASE_URL = "https://opencode.ai/zen/v1";
+const OPENCODE_ZEN_USER_AGENT =
+	"opencode/1.15.0 ai-sdk/provider-utils/4.0.23 runtime/bun/1.3.13";
+
+/**
+ * Headers that must be sent to known endpoints.
+ *
+ * Currently: OpenCode Zen's free tier endpoint requires this exact User-Agent;
+ * anything else returns an empty object.
+ *
+ * The comparison is case-insensitive and tolerates a trailing slash.
+ */
+export function endpointHeaders(baseUrl: string): Record<string, string> {
+	const normalized = baseUrl.replace(/\/+$/, "").toLowerCase();
+	if (normalized === OPENCODE_ZEN_BASE_URL) return { "User-Agent": OPENCODE_ZEN_USER_AGENT };
+	return {};
+}

package/index.ts

@@ -11,7 +11,7 @@
 
 import type { ExtensionAPI, ExtensionContext } from "@earendil-works/pi-coding-agent";
 import { getApiProvider, type Api } from "@earendil-works/pi-ai";
-import { configPath, loadConfig, saveConfig, toProviderModels, type KeypoolConfig, type PoolConfig } from "./config.ts";
+import { configPath, endpointHeaders, loadConfig, saveConfig, toProviderModels, type KeypoolConfig, type PoolConfig } from "./config.ts";
 import { KeyPool } from "./pool.ts";
 import { createRotatingStreamSimple } from "./stream.ts";
 import { runManager, type ManagerHooks } from "./manage.ts";
@@ -69,7 +69,7 @@ export default function multikey(pi: ExtensionAPI) {
 			// Real keys are injected per-request by the rotating stream function.
 			apiKey: "multikey-managed",
 			api,
-			headers: pool.headers,
+			headers: { ...pool.headers, ...endpointHeaders(pool.baseUrl) },
 			models: toProviderModels(pool),
 			streamSimple: createRotatingStreamSimple(keyPool, api, notify),
 		});

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "pi-multikey",
-	"version": "1.4.0",
+	"version": "1.5.0",
 	"description": "One pi provider backed by many API keys: automatic 429 rotation, per-request key leases for concurrent subagents, and a /multikey management TUI",
 	"keywords": [
 		"pi-package",

package/probe.ts

@@ -1,3 +1,5 @@
+import { endpointHeaders } from "./config.ts";
+
 /**
  * Endpoint probing: auto-detect the auth header style and fetch the model list.
  *
@@ -57,11 +59,11 @@ function authHeaders(style: AuthStyle, key: string): Record<string, string> {
 	return style === "bearer" ? { Authorization: `Bearer ${key}` } : { "x-api-key": key };
 }
 
-async function fetchJson(url: string, headers: Record<string, string>, timeoutMs: number): Promise<{ status: number; body?: unknown }> {
+async function fetchJson(url: string, headers: Record<string, string>, timeoutMs: number, baseUrl?: string): Promise<{ status: number; body?: unknown }> {
 	try {
 		const response = await fetch(url, {
 			method: "GET",
-			headers: { Accept: "application/json", ...headers },
+			headers: { Accept: "application/json", ...endpointHeaders(baseUrl ?? url), ...headers },
 			signal: AbortSignal.timeout(timeoutMs),
 		});
 		let body: unknown;
@@ -147,7 +149,7 @@ async function chatProbe(baseUrl: string, style: AuthStyle, key: string, modelId
 	try {
 		const response = await fetch(`${trimSlash(baseUrl)}/chat/completions`, {
 			method: "POST",
-			headers: { "Content-Type": "application/json", ...authHeaders(style, key) },
+			headers: { "Content-Type": "application/json", ...endpointHeaders(baseUrl), ...authHeaders(style, key) },
 			body: JSON.stringify({ model: modelId, max_tokens: 4, messages: [{ role: "user", content: "ping" }] }),
 			signal: AbortSignal.timeout(CHAT_TIMEOUT_MS),
 		});
@@ -186,7 +188,7 @@ export async function probeEndpoint(
 			let result: { status: number; body?: unknown };
 			try {
 				emit(`GET ${url} (${style === "bearer" ? "Authorization: Bearer" : "x-api-key"})…`);
-				result = await fetchJson(url, authHeaders(style, key), MODELS_TIMEOUT_MS);
+				result = await fetchJson(url, authHeaders(style, key), MODELS_TIMEOUT_MS, baseUrl);
 			} catch (error) {
 				emit(`   network error: ${error instanceof Error ? error.message : String(error)}`);
 				continue;