docs: add Studio page to agentv.dev

apps/cli/src/commands/results/serve.ts

@@ -725,6 +725,61 @@ export function createApp(
     }
   });
 
+  /** Aggregate runs from all registered projects, sorted by timestamp descending. */
+  app.get('/api/projects/all-runs', (c) => {
+    const registry = loadProjectRegistry();
+    const allRuns: Array<{
+      filename: string;
+      path: string;
+      timestamp: string;
+      test_count: number;
+      pass_rate: number;
+      avg_score: number;
+      size_bytes: number;
+      target?: string;
+      experiment?: string;
+      project_id: string;
+      project_name: string;
+    }> = [];
+
+    for (const p of registry.projects) {
+      try {
+        const metas = listResultFiles(p.path);
+        for (const m of metas) {
+          let target: string | undefined;
+          let experiment: string | undefined;
+          try {
+            const records = loadLightweightResults(m.path);
+            if (records.length > 0) {
+              target = records[0].target;
+              experiment = records[0].experiment;
+            }
+          } catch {
+            // ignore enrichment errors
+          }
+          allRuns.push({
+            filename: m.filename,
+            path: m.path,
+            timestamp: m.timestamp,
+            test_count: m.testCount,
+            pass_rate: m.passRate,
+            avg_score: m.avgScore,
+            size_bytes: m.sizeBytes,
+            ...(target && { target }),
+            ...(experiment && { experiment }),
+            project_id: p.id,
+            project_name: p.name,
+          });
+        }
+      } catch {
+        // skip inaccessible projects
+      }
+    }
+
+    allRuns.sort((a, b) => b.timestamp.localeCompare(a.timestamp));
+    return c.json({ runs: allRuns });
+  });
+
   // ── Data routes (unscoped) ────────────────────────────────────────────
 
   app.get('/api/config', (c) => handleConfig(c, defaultCtx));

apps/studio/src/components/Sidebar.tsx

@@ -12,8 +12,10 @@ import { Link, useMatchRoute } from '@tanstack/react-router';
 
 import {
   isPassing,
+  useAllProjectRuns,
   useCategoryDatasets,
   useExperiments,
+  useProjectList,
   useProjectRunDetail,
   useProjectRunList,
   useRunDetail,
@@ -104,11 +106,19 @@ export function Sidebar() {
 
 function RunSidebar() {
   const matchRoute = useMatchRoute();
-  const { data } = useRunList();
+  const { data: projectData } = useProjectList();
+  const hasProjects = (projectData?.projects.length ?? 0) > 0;
 
   const isHome = matchRoute({ to: '/' });
   const runMatch = matchRoute({ to: '/runs/$runId', fuzzy: true });
 
+  // On the projects landing page, show aggregated runs from all projects
+  const useAggregated = hasProjects && isHome !== false;
+
+  const { data: localData } = useRunList();
+  const { data: aggregatedData } = useAllProjectRuns();
+  const data = useAggregated ? aggregatedData : localData;
+
   return (
     <aside className="flex w-64 flex-col border-r border-gray-800 bg-gray-900/50">
       <div className="flex items-center gap-2 border-b border-gray-800 px-4 py-4">
@@ -130,6 +140,21 @@ function RunSidebar() {
             'runId' in runMatch &&
             (runMatch as { runId: string }).runId === run.filename;
 
+          // Aggregated runs link to their project's run detail
+          if (run.project_id) {
+            return (
+              <Link
+                key={`${run.project_id}/${run.filename}`}
+                to="/projects/$projectId/runs/$runId"
+                params={{ projectId: run.project_id, runId: run.filename }}
+                className="mb-0.5 block truncate rounded-md px-2 py-1.5 text-sm text-gray-400 transition-colors hover:bg-gray-800/50 hover:text-gray-200"
+                title={run.project_name}
+              >
+                {run.filename}
+              </Link>
+            );
+          }
+
           return (
             <Link
               key={run.filename}

apps/studio/src/lib/api.ts

@@ -208,6 +208,16 @@ export function useProjectList() {
   return useQuery(projectListOptions);
 }
 
+export const allProjectRunsOptions = queryOptions({
+  queryKey: ['projects', 'all-runs'],
+  queryFn: () => fetchJson<RunListResponse>('/api/projects/all-runs'),
+  refetchInterval: 5_000,
+});
+
+export function useAllProjectRuns() {
+  return useQuery(allProjectRunsOptions);
+}
+
 export async function addProjectApi(projectPath: string): Promise<ProjectEntry> {
   const res = await fetch('/api/projects', {
     method: 'POST',

apps/studio/src/lib/types.ts

@@ -15,6 +15,8 @@ export interface RunMeta {
   size_bytes: number;
   target?: string;
   experiment?: string;
+  project_id?: string;
+  project_name?: string;
 }
 
 export interface RunListResponse {

apps/web/src/content/docs/docs/tools/studio.mdx

@@ -0,0 +1,94 @@
+---
+title: Studio
+description: Visual dashboard for reviewing evaluation results
+sidebar:
+  order: 6
+---
+
+import { Image } from 'astro:assets';
+import studioRuns from '../../../../assets/screenshots/studio-runs.png';
+import studioProjects from '../../../../assets/screenshots/studio-projects.png';
+
+The `studio` command launches a web-based dashboard for browsing evaluation runs, inspecting individual test results, and reviewing scores.
+
+<Image src={studioRuns} alt="AgentV Studio showing evaluation runs with pass rates and scores" />
+
+## Usage
+
+```bash
+agentv studio
+```
+
+Studio auto-discovers results from `.agentv/results/` in the current directory and opens at `http://localhost:3117`.
+
+You can also point it at a specific results file:
+
+```bash
+agentv studio .agentv/results/runs/2026-03-30T11-45-56-989Z/index.jsonl
+```
+
+## Options
+
+| Option | Description |
+|--------|-------------|
+| `--port`, `-p` | Port to listen on (flag > `PORT` env var > 3117) |
+| `--dir`, `-d` | Working directory (default: current directory) |
+| `--multi` | Launch in multi-project dashboard mode |
+| `--add <path>` | Register a project by path |
+| `--remove <id>` | Unregister a project by ID |
+| `--discover <path>` | Scan a directory tree for repos with `.agentv/` |
+
+## Features
+
+- **Recent Runs** -- table of all evaluation runs with target, experiment, timestamp, test count, pass rate, and mean score
+- **Experiments** -- group and compare runs by experiment name
+- **Targets** -- group runs by target (model/agent)
+- **Run Detail** -- drill into a run to see per-test results, scores, and evaluator output
+- **Human Review** -- add feedback annotations to individual test results
+
+## Multi-Project Dashboard
+
+By default, Studio shows results for the current directory. The multi-project mode lets you view results across multiple repositories from a single dashboard.
+
+### Registering Projects
+
+Register projects one at a time:
+
+```bash
+agentv studio --add /path/to/project-a
+agentv studio --add /path/to/project-b
+```
+
+Each path must contain a `.agentv/` directory. Projects are stored in `~/.agentv/projects.yaml`.
+
+### Auto-Discovery
+
+Scan a parent directory to find and register all projects:
+
+```bash
+agentv studio --discover /path/to/repos
+```
+
+This recursively searches (up to 2 levels deep) for directories containing `.agentv/` and registers them.
+
+### Launching the Dashboard
+
+Once projects are registered, launch the multi-project dashboard:
+
+```bash
+agentv studio --multi
+```
+
+If you have any registered projects, `--multi` is automatically enabled. The landing page shows a card for each project with run count, pass rate, and last run time. Click a project to view its runs.
+
+<Image src={studioProjects} alt="AgentV Studio multi-project dashboard showing project cards with pass rates" />
+
+### Removing Projects
+
+Unregister a project by its ID:
+
+```bash
+agentv studio --remove my-project
+```
+
+Project IDs are derived from the directory name (e.g., `/home/user/repos/my-project` becomes `my-project`).