feat(vom): actions refactor

packages/dsh-plugin-browserskill/src/tools.ts

@@ -581,9 +581,15 @@ function defineBrowserOperations(deps: ToolDeps, register: DefinitionRegistrar):
           type: "integer",
           description: "Number of consecutive presses (double-click = 2).",
         },
+        modifiers: {
+          type: "array",
+          items: { type: "string", enum: ["alt", "ctrl", "meta", "shift"] },
+          description: "Keyboard modifiers held during the click.",
+        },
         captureId: {
           type: "string",
-          description: "Short-lived Surface capture id returned by browser_screenshot.",
+          description:
+            "Short-lived Surface capture id returned by browser_inspect action=screenshot.",
         },
         imageX: {
           type: "number",
@@ -626,6 +632,9 @@ function defineBrowserOperations(deps: ToolDeps, register: DefinitionRegistrar):
         const cmdArgs = ["click", "--session", sessionId];
         if (args.button !== undefined) cmdArgs.push("--button", args.button);
         if (args.clickCount !== undefined) cmdArgs.push("--click-count", String(args.clickCount));
+        if (args.modifiers !== undefined && args.modifiers.length > 0) {
+          cmdArgs.push("--modifiers", args.modifiers.join(","));
+        }
         if (args.captureId !== undefined) {
           cmdArgs.push(
             "--capture",

packages/dsh-plugin-browserskill/tests/tools.test.ts

@@ -478,12 +478,12 @@ describe("interaction tools", () => {
     emulate: { tab_id: 7, cleared: false },
   };
 
-  it("interact.click maps target, button, and click count", async () => {
+  it("interact.click maps target, button, click count, and modifiers", async () => {
     const { tools, calls } = setup(responses);
     await startSession(tools);
     const click = tools.get("interact.click");
     const value = (await click?.execute(
-      { target: "@e1", button: "right", clickCount: 2 },
+      { target: "@e1", button: "right", clickCount: 2, modifiers: ["ctrl", "shift"] },
       makeExec(),
     )) as { session: string; x: number; y: number };
     expect(value).toMatchObject({ session: "s1", x: 10, y: 20 });
@@ -495,6 +495,8 @@ describe("interaction tools", () => {
       "right",
       "--click-count",
       "2",
+      "--modifiers",
+      "ctrl,shift",
       "@e1",
     ]);
   });
@@ -973,7 +975,7 @@ describe("inspect.screenshot", () => {
       },
     });
     await startSession(tools);
-    const screenshot = tools.get("browser_screenshot");
+    const screenshot = tools.get("inspect.screenshot");
     const value = await screenshot?.execute({ ref: "@e1" }, makeExec());
 
     expect(value).toMatchObject({