feat: add mousedown and mouseup commands for press-and-hold interactions

Separates the mousePressed/mouseReleased CDP events that click bundles
together, enabling press-and-hold patterns needed for CAPTCHA solving
(e.g. Arkose Labs). Follows the same pattern as keydown/keyup.

Supports ref, CSS selector, nodeId, and x/y coordinate targeting.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: fry-lobster

cmd/pinchtab/cmd_cli_browser_actions.go

@@ -79,6 +79,10 @@ var keydownCmd = newSimpleActionCmd("keydown <key>", "Hold a key down", "keydown
 
 var keyupCmd = newSimpleActionCmd("keyup <key>", "Release a key", "keyup", cobra.ExactArgs(1))
 
+var mousedownCmd = newOptionalRefActionCmd("mousedown [ref]", "Press mouse button down", "mousedown")
+
+var mouseupCmd = newOptionalRefActionCmd("mouseup [ref]", "Release mouse button", "mouseup")
+
 var scrollintoviewCmd = newOptionalRefActionCmd("scrollintoview <selector>", "Scroll element into view and return bounding box", "scrollintoview")
 
 var dialogCmd = &cobra.Command{

cmd/pinchtab/cmd_cli_register.go

@@ -39,6 +39,8 @@ func registerBrowserCommands() {
 		keyboardCmd,
 		keydownCmd,
 		keyupCmd,
+		mousedownCmd,
+		mouseupCmd,
 		scrollintoviewCmd,
 		dialogCmd,
 		consoleCmd,
@@ -84,6 +86,8 @@ func registerBrowserCommands() {
 		keyboardCmd,
 		keydownCmd,
 		keyupCmd,
+		mousedownCmd,
+		mouseupCmd,
 		scrollintoviewCmd,
 		dialogCmd,
 		consoleCmd,
@@ -185,6 +189,8 @@ func configureBrowserFlags() {
 		keyboardInsertTextCmd,
 		keydownCmd,
 		keyupCmd,
+		mousedownCmd,
+		mouseupCmd,
 		scrollintoviewCmd,
 		networkCmd,
 		waitCmd,

internal/bridge/action_pointer.go

@@ -60,6 +60,40 @@ func (b *Bridge) actionHover(ctx context.Context, req ActionRequest) (map[string
 	return nil, fmt.Errorf("need selector, ref, nodeId, or x/y coordinates")
 }
 
+func (b *Bridge) actionMouseDown(ctx context.Context, req ActionRequest) (map[string]any, error) {
+	if req.NodeID > 0 {
+		return map[string]any{"mousedown": true}, MouseDownByNodeID(ctx, req.NodeID)
+	}
+	if req.Selector != "" {
+		node, err := firstNodeBySelector(ctx, req.Selector)
+		if err != nil {
+			return nil, err
+		}
+		return map[string]any{"mousedown": true}, MouseDownByNodeID(ctx, int64(node.BackendNodeID))
+	}
+	if req.HasXY {
+		return map[string]any{"mousedown": true, "x": req.X, "y": req.Y}, MouseDownByCoordinate(ctx, req.X, req.Y)
+	}
+	return nil, fmt.Errorf("need selector, ref, nodeId, or x/y coordinates")
+}
+
+func (b *Bridge) actionMouseUp(ctx context.Context, req ActionRequest) (map[string]any, error) {
+	if req.NodeID > 0 {
+		return map[string]any{"mouseup": true}, MouseUpByNodeID(ctx, req.NodeID)
+	}
+	if req.Selector != "" {
+		node, err := firstNodeBySelector(ctx, req.Selector)
+		if err != nil {
+			return nil, err
+		}
+		return map[string]any{"mouseup": true}, MouseUpByNodeID(ctx, int64(node.BackendNodeID))
+	}
+	if req.HasXY {
+		return map[string]any{"mouseup": true, "x": req.X, "y": req.Y}, MouseUpByCoordinate(ctx, req.X, req.Y)
+	}
+	return nil, fmt.Errorf("need selector, ref, nodeId, or x/y coordinates")
+}
+
 func (b *Bridge) actionScroll(ctx context.Context, req ActionRequest) (map[string]any, error) {
 	if req.NodeID > 0 {
 		return map[string]any{"scrolled": true}, ScrollByNodeID(ctx, req.NodeID)

internal/bridge/action_registry.go

@@ -19,6 +19,8 @@ const (
 	ActionKeyboardInsert = "keyboard-inserttext"
 	ActionKeyDown        = "keydown"
 	ActionKeyUp          = "keyup"
+	ActionMouseDown      = "mousedown"
+	ActionMouseUp        = "mouseup"
 	ActionScrollIntoView = "scrollintoview"
 )
 
@@ -42,6 +44,8 @@ func (b *Bridge) InitActionRegistry() {
 		ActionKeyboardInsert: b.actionKeyboardInsert,
 		ActionKeyDown:        b.actionKeyDown,
 		ActionKeyUp:          b.actionKeyUp,
+		ActionMouseDown:      b.actionMouseDown,
+		ActionMouseUp:        b.actionMouseUp,
 		ActionScrollIntoView: b.actionScrollIntoView,
 	}
 }

internal/bridge/cdpops/pointer.go

@@ -188,6 +188,75 @@ func DragByNodeID(ctx context.Context, nodeID int64, dx, dy int) error {
 	)
 }
 
+func MouseDownByCoordinate(ctx context.Context, x, y float64) error {
+	if x < 0 || y < 0 {
+		return fmt.Errorf("x/y coordinates must be >= 0")
+	}
+
+	return chromedp.Run(ctx, chromedp.ActionFunc(func(ctx context.Context) error {
+		return chromedp.FromContext(ctx).Target.Execute(ctx, "Input.dispatchMouseEvent", map[string]any{
+			"type":       "mousePressed",
+			"button":     "left",
+			"clickCount": 1,
+			"x":          x,
+			"y":          y,
+		}, nil)
+	}))
+}
+
+func MouseDownByNodeID(ctx context.Context, nodeID int64) error {
+	x, y, err := GetElementCenter(ctx, nodeID)
+	if err != nil {
+		return err
+	}
+
+	return chromedp.Run(ctx,
+		chromedp.ActionFunc(func(ctx context.Context) error {
+			return chromedp.FromContext(ctx).Target.Execute(ctx, "DOM.scrollIntoViewIfNeeded", map[string]any{"backendNodeId": nodeID}, nil)
+		}),
+		chromedp.ActionFunc(func(ctx context.Context) error {
+			return chromedp.FromContext(ctx).Target.Execute(ctx, "Input.dispatchMouseEvent", map[string]any{
+				"type":       "mousePressed",
+				"button":     "left",
+				"clickCount": 1,
+				"x":          x, "y": y,
+			}, nil)
+		}),
+	)
+}
+
+func MouseUpByCoordinate(ctx context.Context, x, y float64) error {
+	if x < 0 || y < 0 {
+		return fmt.Errorf("x/y coordinates must be >= 0")
+	}
+
+	return chromedp.Run(ctx, chromedp.ActionFunc(func(ctx context.Context) error {
+		return chromedp.FromContext(ctx).Target.Execute(ctx, "Input.dispatchMouseEvent", map[string]any{
+			"type":       "mouseReleased",
+			"button":     "left",
+			"clickCount": 1,
+			"x":          x,
+			"y":          y,
+		}, nil)
+	}))
+}
+
+func MouseUpByNodeID(ctx context.Context, nodeID int64) error {
+	x, y, err := GetElementCenter(ctx, nodeID)
+	if err != nil {
+		return err
+	}
+
+	return chromedp.Run(ctx, chromedp.ActionFunc(func(ctx context.Context) error {
+		return chromedp.FromContext(ctx).Target.Execute(ctx, "Input.dispatchMouseEvent", map[string]any{
+			"type":       "mouseReleased",
+			"button":     "left",
+			"clickCount": 1,
+			"x":          x, "y": y,
+		}, nil)
+	}))
+}
+
 func HoverByCoordinate(ctx context.Context, x, y float64) error {
 	if x < 0 || y < 0 {
 		return fmt.Errorf("x/y coordinates must be >= 0")

internal/bridge/cdpops_facade.go

@@ -63,6 +63,22 @@ func DragByNodeID(ctx context.Context, nodeID int64, dx, dy int) error {
 	return bridgecdpops.DragByNodeID(ctx, nodeID, dx, dy)
 }
 
+func MouseDownByCoordinate(ctx context.Context, x, y float64) error {
+	return bridgecdpops.MouseDownByCoordinate(ctx, x, y)
+}
+
+func MouseDownByNodeID(ctx context.Context, nodeID int64) error {
+	return bridgecdpops.MouseDownByNodeID(ctx, nodeID)
+}
+
+func MouseUpByCoordinate(ctx context.Context, x, y float64) error {
+	return bridgecdpops.MouseUpByCoordinate(ctx, x, y)
+}
+
+func MouseUpByNodeID(ctx context.Context, nodeID int64) error {
+	return bridgecdpops.MouseUpByNodeID(ctx, nodeID)
+}
+
 func HoverByCoordinate(ctx context.Context, x, y float64) error {
 	return bridgecdpops.HoverByCoordinate(ctx, x, y)
 }