dsh-livebench-panel 0.1.2 → 0.1.3

package/lib/index.js

@@ -403,7 +403,7 @@ function apply(ctx) {
 	/** The single run slot. */
 	let run = null;
 
-	const startRun = (body) => {
+	const startRun = async (body) => {
 		const layout = livebenchLayout();
 		if (!layout.available) {
 			return { status: 409, payload: { ok: false, error: `LiveBench venv not found under ${layout.root}` } };
@@ -473,11 +473,23 @@ function apply(ctx) {
 			PATH: `${join(layout.root, ".venv", "Scripts")}${delimiter}${process.env.PATH ?? ""}`,
 		};
 		// Only OpenAI-compatible providers can be routed with --api-base; for
-		// those, hand the key over via env (never the command line).
+		// those, hand the key over via env (never the command line). The key is
+		// resolved through the harness credential seam when available (values may
+		// live encrypted in .credentials.yaml rather than in the process env),
+		// falling back to the plain environment variable.
 		if (provider && provider.api === "openai-completions" && provider.baseURL) {
 			args.push("--api-base", provider.baseURL);
-			if (provider.keyEnv && process.env[provider.keyEnv]) {
-				env.LIVEBENCH_API_KEY = process.env[provider.keyEnv];
+			if (provider.keyEnv) {
+				let key;
+				try {
+					const credentials = ctx.get ? ctx.get("credentials") : undefined;
+					if (credentials && typeof credentials.resolve === "function") {
+						const resolved = await credentials.resolve(provider.keyEnv);
+						if (resolved && typeof resolved.value === "string" && resolved.value.length > 0) key = resolved.value;
+					}
+				} catch { /* credential seam unavailable — fall through to env */ }
+				if (!key && process.env[provider.keyEnv]) key = process.env[provider.keyEnv];
+				if (key) env.LIVEBENCH_API_KEY = key;
 			}
 		}
 
@@ -581,7 +593,7 @@ function apply(ctx) {
 				sendJson(res, 409, { ok: false, error: "another evaluation is already running" });
 				return;
 			}
-			const result = startRun(body);
+			const result = await startRun(body);
 			sendJson(res, result.status, result.payload);
 		},
 	}), `${name}: start route`);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "dsh-livebench-panel",
-  "version": "0.1.2",
+  "version": "0.1.3",
   "description": "DSH web plugin: a LiveBench tab in the Trajectory view (right of 对话/轨迹). Run LiveBench evaluations against every model configured in the DeepSeek Harness — pick provider/model, category, task, release and question range from dropdowns, watch progress, and read scores in place.",
   "license": "MIT",
   "type": "module",