@brainbase-labs/cli 0.30.0 → 0.31.0

package/README.md

@@ -83,7 +83,9 @@ prompt.
 Idempotency is explicit-only: requests omit the idempotency header unless
 `--idempotency-key` is supplied. Reuse a key only for an identical retry.
 Starting a run requires `--yes`; confirming the same plan ID is replay-safe.
-Interrupting `run watch` does not cancel the remote run.
+Interrupting `run watch` does not cancel the remote run. A run that MAS pauses
+for billing stops `run watch` with exit code 1; once billing is fixed, run
+`brainbase benchmark run resume <run-id> --yes` to put it back in the queue.
 
 Use `brainbase benchmark --help`, `brainbase benchmark run --help`, or
 `brainbase benchmark <command> --help` for local command help.

package/dist/index.js

@@ -36018,7 +36018,7 @@ function padStart(s, n) {
 // package.json
 var package_default = {
   name: "@brainbase-labs/cli",
-  version: "0.30.0",
+  version: "0.31.0",
   description: "Pack, share, and install agent templates across harnesses (Claude Code, Codex, ...).",
   type: "module",
   bin: {
@@ -84745,7 +84745,7 @@ class BenchmarkApiClient {
         try {
           responseBody = text3 ? JSON.parse(text3) : null;
         } catch {}
-        throw new BenchmarkApiError(redactSensitiveText(apiErrorMessage(responseBody, response.status)), response.status, responseBody, safeCode(responseBody), response.headers.get("x-request-id") ?? response.headers.get("x-correlation-id") ?? undefined);
+        throw new BenchmarkApiError(redactSensitiveText(masApiErrorMessage(responseBody, response.status)), response.status, responseBody, safeCode(responseBody), response.headers.get("x-request-id") ?? response.headers.get("x-correlation-id") ?? undefined);
       }
       if (options.mode === "binary") {
         return {
@@ -84960,6 +84960,11 @@ class BenchmarkApiClient {
       method: "POST"
     });
   }
+  resumeRun(runId) {
+    return this.request(`/runs/${encodeURIComponent(runId)}/resume`, {
+      method: "POST"
+    });
+  }
   listDiagnoses(runId, query = {}) {
     return this.request(`/runs/${encodeURIComponent(runId)}/diagnoses`, {
       query: {
@@ -87345,6 +87350,12 @@ class BenchmarkControlOutput {
 `);
     }
   }
+  hint(text2) {
+    if (this.mode.json || this.mode.jsonl)
+      return;
+    this.writer.stdout(`${text2}
+`);
+  }
   page(page, all = false) {
     if (this.mode.jsonl) {
       for (const item of page.items)
@@ -87376,6 +87387,7 @@ class BenchmarkControlOutput {
 
 // src/cli/benchmark-control.ts
 var TERMINAL_RUN_STATUSES = new Set(["cancelled", "failed", "succeeded"]);
+var PAUSED_RUN_STATUS = "paused";
 var MAX_MANIFEST_IMPORT_BYTES = 2 * 1024 * 1024;
 var MAX_CSV_IMPORT_BYTES = 10 * 1024 * 1024;
 var MAX_PAGINATION_PAGES = 1e4;
@@ -87552,7 +87564,7 @@ function commandHelp() {
     "",
     `  ${import_picocolors60.default.cyan("init | list | show | create | validate | pull | push")}`,
     `  ${import_picocolors60.default.cyan("publish | revisions | import | export | archive | restore")}`,
-    `  ${import_picocolors60.default.cyan("run plan|start|watch|list|show|cancel")}`,
+    `  ${import_picocolors60.default.cyan("run plan|start|watch|list|show|cancel|resume")}`,
     `  ${import_picocolors60.default.cyan("results | diagnoses | attempt | artifacts | export-results")}`,
     `  ${import_picocolors60.default.cyan("history | baseline show|set|clear")}`,
     "",
@@ -87952,7 +87964,7 @@ function runPlanInput(options, revisionId) {
     revision_id: revisionId,
     variants: variants.length > 0 ? variants.map(parseVariant) : [{ key: "default" }],
     repetitions: integerOption(options.parsed, "repetitions", { min: 1, max: 10 }) ?? 1,
-    concurrency: integerOption(options.parsed, "concurrency", { min: 1, max: 10 }) ?? 5
+    concurrency: integerOption(options.parsed, "concurrency", { min: 1, max: 50 }) ?? 5
   };
   if (cases.length > 0)
     input.case_keys = cases;
@@ -87998,7 +88010,15 @@ async function runStart(cwd2, argv, deps) {
   const { client, output } = context(cwd2, "benchmark run start", options, deps);
   output.result(json(await client.confirmRunPlan(planId)));
 }
-async function watchRun(client, output, runId, interval, deps) {
+function pausedRunHint(runId, run, agentId) {
+  const reason = run.paused_reason;
+  const message = reason && typeof reason === "object" && !Array.isArray(reason) ? reason.message : undefined;
+  const first = typeof message === "string" && message.trim() ? `Run paused: ${message.trim()}` : "Run paused.";
+  const agent = agentId ? ` --agent ${agentId}` : "";
+  return `${first}
+Resume with: brainbase benchmark run resume ${runId} --yes${agent}`;
+}
+async function watchRun(client, output, runId, interval, agentId, deps) {
   let interrupted = false;
   const controller = new AbortController;
   const onInterrupt = () => {
@@ -88020,9 +88040,16 @@ async function watchRun(client, output, runId, interval, deps) {
       const digest2 = output.digest(latest);
       if (digest2 !== previous) {
         const status = typeof latest.status === "string" ? latest.status : "unknown";
-        output.event(TERMINAL_RUN_STATUSES.has(status) ? "run.completed" : "run.updated", latest, typeof latest.updated_at === "string" ? latest.updated_at : undefined);
+        const terminal = TERMINAL_RUN_STATUSES.has(status);
+        const paused = status === PAUSED_RUN_STATUS;
+        output.event(terminal ? "run.completed" : paused ? "run.paused" : "run.updated", latest, typeof latest.updated_at === "string" ? latest.updated_at : undefined);
         previous = digest2;
-        if (TERMINAL_RUN_STATUSES.has(status)) {
+        if (paused) {
+          output.hint(pausedRunHint(runId, latest, agentId));
+          deps.setExitCode(1);
+          return { run: latest, interrupted: false };
+        }
+        if (terminal) {
           if (status !== "succeeded")
             deps.setExitCode(1);
           return { run: latest, interrupted: false };
@@ -88044,7 +88071,7 @@ async function runWatch(cwd2, argv, deps) {
   const watched = await watchRun(client, output, options.positionals[0], integerOption(options.parsed, "interval", {
     min: 100,
     max: 60000
-  }) ?? 2000, deps);
+  }) ?? 2000, options.agent, deps);
   if (options.json) {
     output.result({
       run: watched.run,
@@ -88084,6 +88111,15 @@ async function runCancel(cwd2, argv, deps) {
   const { client, output } = context(cwd2, "benchmark run cancel", options, deps);
   output.result(json(await client.cancelRun(options.positionals[0])));
 }
+async function runResume(cwd2, argv, deps) {
+  const options = parseControlOptions(argv, ["yes"]);
+  expectPositionals(options.positionals, 1, 1, "brainbase benchmark run resume <run-id>");
+  if (!bool(options.parsed, "yes")) {
+    throw usageError("--yes is required to resume a benchmark run");
+  }
+  const { client, output } = context(cwd2, "benchmark run resume", options, deps);
+  output.result(json(await client.resumeRun(options.positionals[0])));
+}
 async function runOneShot(cwd2, argv, deps) {
   const options = parseControlOptions(argv, [
     ...RUN_PLAN_OPTIONS,
@@ -88123,7 +88159,7 @@ async function runOneShot(cwd2, argv, deps) {
   const watched = await watchRun(client, output, runId, integerOption(options.parsed, "interval", {
     min: 100,
     max: 60000
-  }) ?? 2000, deps);
+  }) ?? 2000, options.agent, deps);
   if (options.json) {
     output.result({
       plan,
@@ -88149,6 +88185,8 @@ async function runRunRouter(cwd2, argv, deps) {
     case "cancel":
     case "stop":
       return await runCancel(cwd2, rest2, deps);
+    case "resume":
+      return await runResume(cwd2, rest2, deps);
     case undefined:
     case "help":
     case "--help":
@@ -88422,10 +88460,10 @@ function commandLevelHelp(sub, argv) {
       if (action === "show") {
         return usage("brainbase benchmark run show <run-id>");
       }
-      if (action === "cancel" || action === "stop") {
+      if (action === "cancel" || action === "stop" || action === "resume") {
         return usage(`brainbase benchmark run ${action} <run-id> --yes`);
       }
-      return usage("brainbase benchmark run <benchmark-id> --yes [--revision <revision-id>] [--watch] [--idempotency-key <key>]", "brainbase benchmark run plan <benchmark-id> --revision <revision-id>", "brainbase benchmark run start <plan-id> --yes", "brainbase benchmark run watch <run-id> [--interval-ms <n>]", "brainbase benchmark run list [benchmark-id] [--status <status>] [--limit <n>] [--cursor <cursor>] [--all]", "brainbase benchmark run show <run-id>", "brainbase benchmark run cancel <run-id> --yes");
+      return usage("brainbase benchmark run <benchmark-id> --yes [--revision <revision-id>] [--watch] [--idempotency-key <key>]", "brainbase benchmark run plan <benchmark-id> --revision <revision-id>", "brainbase benchmark run start <plan-id> --yes", "brainbase benchmark run watch <run-id> [--interval-ms <n>]", "brainbase benchmark run list [benchmark-id] [--status <status>] [--limit <n>] [--cursor <cursor>] [--all]", "brainbase benchmark run show <run-id>", "brainbase benchmark run cancel <run-id> --yes", "brainbase benchmark run resume <run-id> --yes");
     case "results":
       return usage("brainbase benchmark results <run-id> [--baseline-target <target-id>]");
     case "diagnoses":
@@ -88458,7 +88496,7 @@ function automationCommand(sub, argv) {
     const action = argv[0];
     if (action === "stop")
       return "benchmark run cancel";
-    if (action && ["plan", "start", "watch", "list", "show", "cancel"].includes(action)) {
+    if (action && ["plan", "start", "watch", "list", "show", "cancel", "resume"].includes(action)) {
       return `benchmark run ${action}`;
     }
     return "benchmark run";

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@brainbase-labs/cli",
-  "version": "0.30.0",
+  "version": "0.31.0",
   "description": "Pack, share, and install agent templates across harnesses (Claude Code, Codex, ...).",
   "type": "module",
   "bin": {