test: cover agent selection edge cases

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: centdix

bin/src/worktree-commands.test.ts

@@ -144,6 +144,10 @@ describe("parseAddCommandArgs", () => {
     });
   });
 
+  it("rejects empty agent ids", () => {
+    expect(() => parseAddCommandArgs(["feature/search", "--agent", "   "])).toThrow("Agent id cannot be empty");
+  });
+
   it("returns null for help", () => {
     expect(parseAddCommandArgs(["--help"])).toBeNull();
   });

frontend/src/App.test.ts

@@ -587,6 +587,23 @@ describe("App create selection", () => {
     });
   });
 
+  it("shows prefixed branch previews when multiple agents are selected", async () => {
+    vi.mocked(fetchWorktrees).mockResolvedValue([]);
+
+    render(App);
+
+    await fireEvent.click(screen.getByTitle("New Worktree (Cmd+K)"));
+    await screen.findByText("New Worktree");
+
+    await fireEvent.click(screen.getByRole("checkbox", { name: /Codex builtin chat resume/i }));
+    await fireEvent.input(screen.getByLabelText(/Branch name/i), {
+      target: { value: "feature/new" },
+    });
+
+    expect(screen.getByText("claude-feature/new")).toBeInTheDocument();
+    expect(screen.getByText("codex-feature/new")).toBeInTheDocument();
+  });
+
   it("submits multi-agent worktree creation when multiple agents are selected", async () => {
     vi.mocked(fetchWorktrees).mockResolvedValue([]);
     vi.mocked(api.createWorktree).mockResolvedValue({

to-do.md

@@ -85,24 +85,24 @@
 - [x] Update help text and tests accordingly.
 
 ## 13. Add backend tests
-- Registry tests
-- Config read/write tests for custom agents
-- Multi-agent target expansion tests
-- Launch command generation tests for custom agents
-- Validation tests for unknown/duplicate/invalid agents
-- Conversation capability tests for unsupported agents
+- [x] Registry tests
+- [x] Config read/write tests for custom agents
+- [x] Multi-agent target expansion tests
+- [x] Launch command generation tests for custom agents
+- [x] Validation tests for unknown/duplicate/invalid agents
+- [x] Conversation capability tests for unsupported agents
 
 ## 14. Add frontend tests
-- Settings agent-management flows
-- Create-worktree multi-select behavior
-- Branch preview behavior
-- Capability-driven chat visibility
-- API integration typing and state transitions
+- [x] Settings agent-management flows
+- [x] Create-worktree multi-select behavior
+- [x] Branch preview behavior
+- [ ] Capability-driven chat visibility
+- [x] API integration typing and state transitions
 
 ## 15. Add CLI tests
-- Parsing repeated or multiple agent selections
-- Validation behavior for invalid agent ids
-- Successful multi-agent worktree creation inputs
+- [x] Parsing repeated or multiple agent selections
+- [x] Validation behavior for invalid agent ids
+- [x] Successful multi-agent worktree creation inputs
 
 ## 16. Roll out incrementally
 - First land types and contracts.