feat(cli): rename studio to dashboard (ag-uh9)

Author: christso

apps/cli/src/cli.ts

@@ -4,7 +4,7 @@ import { killAllTrackedChildren } from '@agentv/core';
 import { runCli } from './index.js';
 
 // Forward SIGINT/SIGTERM to spawned provider subprocesses before exiting.
-// Without this, Studio's `child.kill('SIGTERM')` against the CLI orphans
+// Without this, Dashboard's `child.kill('SIGTERM')` against the CLI orphans
 // any in-flight `claude`/`codex`/`pi`/`copilot` subprocess. The partial
 // `index.jsonl` is already row-by-row durable, so finished tests survive.
 //

apps/cli/src/commands/eval/run-eval.ts

@@ -1460,7 +1460,7 @@ export async function runEvalCommand(
 
   // Write a stub benchmark.json before dispatching tests, carrying the planned
   // execution count so an interrupted run can still surface as resumable in
-  // Studio (results.length < planned_test_count) even when every recorded row
+  // Dashboard (results.length < planned_test_count) even when every recorded row
   // has execution_status: ok. The end-of-run write preserves this value via
   // readPlannedTestCount inside aggregateRunDir / writeArtifactsFromResults.
   // Skip on resume — we want to preserve the *original* planned count.

apps/cli/src/commands/results/eval-runner.ts

@@ -1,13 +1,13 @@
 /**
- * Studio eval runner — discovery, launch, and status tracking for eval runs
- * initiated from the Studio UI.
+ * Dashboard eval runner — discovery, launch, and status tracking for eval runs
+ * initiated from the Dashboard UI.
  *
  * Provides Hono route handlers for:
  *   - GET  /api/eval/discover  — discover eval files in the project
  *   - GET  /api/eval/targets   — list available target names
  *   - POST /api/eval/run       — launch an eval run as a child process
  *   - GET  /api/eval/status/:id — poll running eval status
- *   - GET  /api/eval/runs      — list active and recent Studio-launched runs
+ *   - GET  /api/eval/runs      — list active and recent Dashboard-launched runs
  *
  * All handlers accept a `cwd` (project root) to resolve paths against.
  * The module spawns `bun apps/cli/src/cli.ts eval run ...` and tracks
@@ -34,7 +34,7 @@ import { findRepoRoot } from '../eval/shared.js';
 
 // ── In-memory run tracker ────────────────────────────────────────────────
 
-interface StudioRun {
+interface DashboardRun {
   id: string;
   status: 'starting' | 'running' | 'finished' | 'failed';
   command: string;
@@ -50,14 +50,14 @@ interface StudioRun {
   process?: ChildProcess;
 }
 
-const activeRuns = new Map<string, StudioRun>();
+const activeRuns = new Map<string, DashboardRun>();
 
 function generateRunId(): string {
   const now = new Date();
   const pad = (n: number, w = 2) => String(n).padStart(w, '0');
   const ts = `${now.getFullYear()}${pad(now.getMonth() + 1)}${pad(now.getDate())}-${pad(now.getHours())}${pad(now.getMinutes())}${pad(now.getSeconds())}`;
   const rand = Math.random().toString(36).slice(2, 6);
-  return `studio-${ts}-${rand}`;
+  return `dashboard-${ts}-${rand}`;
 }
 
 // Keep only last 20 finished runs to prevent unbounded memory growth
@@ -73,7 +73,7 @@ function pruneFinishedRuns() {
 }
 
 /**
- * Look up the target for a Studio-launched run by its index.jsonl path.
+ * Look up the target for a Dashboard-launched run by its index.jsonl path.
  * Called by handleRuns in serve.ts when the JSONL has 0 records (run just started).
  */
 export function getActiveRunTarget(indexJsonlPath: string): string | undefined {
@@ -86,12 +86,12 @@ export function getActiveRunTarget(indexJsonlPath: string): string | undefined {
 }
 
 /**
- * Look up the in-memory status for a Studio-launched run by its index.jsonl path.
+ * Look up the in-memory status for a Dashboard-launched run by its index.jsonl path.
  * Returns 'starting' | 'running' | 'finished' | 'failed' if the run is tracked,
  * else undefined. Used by handleRuns to render a spinner for active runs in the
  * RunList instead of a misleading red ✗ derived from a 0 pass-rate.
  */
-export function getActiveRunStatus(indexJsonlPath: string): StudioRun['status'] | undefined {
+export function getActiveRunStatus(indexJsonlPath: string): DashboardRun['status'] | undefined {
   for (const run of activeRuns.values()) {
     if (run.outputDir && path.join(run.outputDir, 'index.jsonl') === indexJsonlPath) {
       return run.status;
@@ -287,7 +287,7 @@ function isCommandAvailable(cmd: string): boolean {
  *
  * The log file is the source of truth shown by the RunDetail "Run Log"
  * section after the run completes. The in-memory `stdout`/`stderr` buffers on
- * `StudioRun` remain capped for live status polling.
+ * `DashboardRun` remain capped for live status polling.
  *
  * Stream `error` events (e.g. the output dir was removed underneath us by a
  * test teardown) are swallowed so they don't surface as unhandled errors and
@@ -348,7 +348,7 @@ export function registerEvalRoutes(
   // ── Launch eval run ────────────────────────────────────────────────────
   app.post('/api/eval/run', async (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     const cwd = getCwd(c);
 
@@ -378,7 +378,7 @@ export function registerEvalRoutes(
     // Determine the output directory for this run. When the caller provides
     // an explicit --output (resume/rerun), use that path. Otherwise generate
     // the default path now so we can pass it via --output and later correlate
-    // the filesystem run with this in-memory StudioRun (needed to show the
+    // the filesystem run with this in-memory DashboardRun (needed to show the
     // target in the sidebar before any results have been written).
     const outputDir = body.output?.trim()
       ? path.resolve(cwd, body.output.trim())
@@ -389,7 +389,7 @@ export function registerEvalRoutes(
     const command = buildCliPreview(args);
     const runId = generateRunId();
 
-    const run: StudioRun = {
+    const run: DashboardRun = {
       id: runId,
       status: 'starting',
       command,
@@ -479,7 +479,7 @@ export function registerEvalRoutes(
   // before exiting.
   app.post('/api/eval/run/:id/stop', (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     const id = c.req.param('id');
     const run = activeRuns.get(id ?? '');
@@ -570,7 +570,7 @@ export function registerEvalRoutes(
 
   app.post('/api/projects/:projectId/eval/run', async (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     const cwd = getCwd(c);
 
@@ -605,7 +605,7 @@ export function registerEvalRoutes(
     const command = buildCliPreview(args);
     const runId = generateRunId();
 
-    const run: StudioRun = {
+    const run: DashboardRun = {
       id: runId,
       status: 'starting',
       command,
@@ -668,7 +668,7 @@ export function registerEvalRoutes(
 
   app.post('/api/projects/:projectId/eval/run/:id/stop', (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     const id = c.req.param('id');
     const run = activeRuns.get(id ?? '');

apps/cli/src/commands/results/report-template.ts

@@ -10,7 +10,7 @@
  * { "tags": ["baseline", "v2-prompt"], "updated_at": "2026-04-10T00:00:00.000Z" }
  * ```
  *
- * Used by the Studio compare API so users can retroactively tag runs
+ * Used by the Dashboard compare API so users can retroactively tag runs
  * without changing the eval YAML or the run manifest itself. This mirrors
  * the Langfuse / W&B / GitHub `tags` pattern — a mutable multi-valued
  * list of free-form labels that lives alongside the immutable run_id.

apps/cli/src/commands/results/run-tags.ts

@@ -1,12 +1,12 @@
 /**
- * `agentv studio` / `agentv serve` — starts the AgentV Studio server, a React SPA for
+ * `agentv dashboard` / `agentv serve` — starts the AgentV Dashboard server, a React SPA for
  * reviewing evaluation results.
  *
  * The server uses Hono for routing and @hono/node-server to listen.
- * The Studio SPA is served from a pre-built dist directory.
+ * The Dashboard SPA is served from a pre-built dist directory.
  *
  * API endpoints:
- *   - GET /           — Studio SPA (React app)
+ *   - GET /           — Dashboard SPA (React app)
  *   - GET /api/runs   — list available run workspaces with metadata
  *   - GET /api/runs/:filename — load results from a specific run workspace
  *   - GET /api/runs/:filename/log — stream the captured console.log for a run
@@ -386,13 +386,13 @@ async function handleRuns(c: C, { searchDir, agentvDir }: DataContext) {
           avgScore = records.reduce((sum, r) => sum + r.score, 0) / records.length;
         } else {
           // Run is in-progress with 0 results written yet — fall back to the
-          // in-memory target stored when the Studio launched this run.
+          // in-memory target stored when the Dashboard launched this run.
           target = getActiveRunTarget(m.path);
         }
       } catch {
         // ignore enrichment errors
       }
-      // Surface live status for Studio-launched runs that are still starting
+      // Surface live status for Dashboard-launched runs that are still starting
       // or running so the RunList can render a spinner instead of the
       // pass/fail dot derived from a 0% pass rate.
       const liveStatus = getActiveRunStatus(m.path);
@@ -447,7 +447,7 @@ async function handleRunDetail(c: C, { searchDir }: DataContext) {
   try {
     const loaded = await loadManifestResultsForMeta(searchDir, meta);
     // Surface run_dir + suite_filter for local runs so the UI can launch a
-    // Studio-side resume against this exact run. Remote runs live in the
+    // Dashboard-side resume against this exact run. Remote runs live in the
     // results-repo cache and cannot be resumed in place, so omit both fields.
     const resumeMeta = meta.source === 'local' ? deriveResumeMeta(searchDir, meta.path) : {};
     const liveStatus = meta.source === 'local' ? getActiveRunStatus(meta.path) : undefined;
@@ -467,7 +467,7 @@ async function handleRunDetail(c: C, { searchDir }: DataContext) {
  * Compute `run_dir` (relative to cwd, snake_case) and `suite_filter` (the
  * eval file path stored in benchmark.json metadata) for a local run manifest.
  * Returns whatever fields could be resolved — both are best-effort and only
- * needed by the Studio "Resume run" / "Rerun failed" actions.
+ * needed by the Dashboard "Resume run" / "Rerun failed" actions.
  */
 function deriveResumeMeta(
   cwd: string,
@@ -1104,11 +1104,11 @@ export function createApp(
     });
   }
 
-  // ── Studio configuration ──────────────────────────────────────────────
+  // ── Dashboard configuration ──────────────────────────────────────────────
 
   app.post('/api/config', async (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     try {
       const body = await c.req.json<Partial<StudioConfig>>();
@@ -1174,7 +1174,7 @@ export function createApp(
 
   app.post('/api/projects', async (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     try {
       const body = await c.req.json<{ path: string }>();
@@ -1268,7 +1268,7 @@ export function createApp(
 
   app.delete('/api/projects/:projectId', (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     const removed = removeProject(c.req.param('projectId') ?? '');
     if (!removed) return c.json({ error: 'Project not found' }, 404);
@@ -1289,13 +1289,13 @@ export function createApp(
   app.get('/api/runs', (c) => handleRuns(c, defaultCtx));
   app.put('/api/runs/:filename/tags', (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     return handleRunTagsPut(c, defaultCtx);
   });
   app.delete('/api/runs/:filename/tags', (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     return handleRunTagsDelete(c, defaultCtx);
   });
@@ -1321,7 +1321,7 @@ export function createApp(
 
   app.post('/api/feedback', async (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     let body: unknown;
     try {
@@ -1416,13 +1416,13 @@ export function createApp(
   app.get('/api/projects/:projectId/runs', (c) => withProject(c, handleRuns));
   app.put('/api/projects/:projectId/runs/:filename/tags', (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     return withProject(c, handleRunTagsPut);
   });
   app.delete('/api/projects/:projectId/runs/:filename/tags', (c) => {
     if (readOnly) {
-      return c.json({ error: 'Studio is running in read-only mode' }, 403);
+      return c.json({ error: 'Dashboard is running in read-only mode' }, 403);
     }
     return withProject(c, handleRunTagsDelete);
   });
@@ -1465,11 +1465,13 @@ export function createApp(
     { readOnly },
   );
 
-  // ── Static file serving for Studio SPA ────────────────────────────────
+  // ── Static file serving for Dashboard SPA ────────────────────────────────
 
   const studioDistPath = options?.studioDir ?? resolveStudioDistDir();
   if (!studioDistPath || !existsSync(path.join(studioDistPath, 'index.html'))) {
-    throw new Error('Studio dist not found. Run "bun run build" in apps/studio/ to build the SPA.');
+    throw new Error(
+      'Dashboard dist not found. Run "bun run build" in apps/studio/ to build the SPA.',
+    );
   }
 
   app.get('/', (c) => {
@@ -1546,8 +1548,8 @@ function resolveStudioDistDir(): string | undefined {
 // ── CLI command ──────────────────────────────────────────────────────────
 
 export const resultsServeCommand = command({
-  name: 'studio',
-  description: 'Start AgentV Studio — a local dashboard for reviewing evaluation results',
+  name: 'dashboard',
+  description: 'Start AgentV Dashboard — a local dashboard for reviewing evaluation results',
   args: {
     source: positional({
       type: optional(string),
@@ -1583,7 +1585,7 @@ export const resultsServeCommand = command({
     }),
     readOnly: flag({
       long: 'read-only',
-      description: 'Disable write operations and launch Studio in read-only leaderboard mode',
+      description: 'Disable write operations and launch Dashboard in read-only leaderboard mode',
     }),
   },
   handler: async ({ source, port, dir, single, add, remove, readOnly }) => {
@@ -1614,7 +1616,7 @@ export const resultsServeCommand = command({
     }
 
     // ── Version check ────────────────────────────────────────────────
-    // Enforce `required_version` from .agentv/config.yaml so Studio/serve
+    // Enforce `required_version` from .agentv/config.yaml so Dashboard/serve
     // match `agentv eval` behavior. Same prompt in TTY, warn+continue
     // otherwise. Single-project scope only — when one agentv instance
     // serves multiple repos with differing version requirements, a