diff --git a/extensions/cli/src/ui/MarkdownRenderer.tsx b/extensions/cli/src/ui/MarkdownRenderer.tsx
index 8cc2d2223da..eb451c59001 100644
--- a/extensions/cli/src/ui/MarkdownRenderer.tsx
+++ b/extensions/cli/src/ui/MarkdownRenderer.tsx
@@ -1,4 +1,4 @@
-import { Text } from "ink";
+import { Box, Text } from "ink";
 import React from "react";
 
 import {
@@ -84,6 +84,17 @@ const MarkdownRenderer: React.FC<MarkdownRendererProps> = React.memo(
           </Text>
         ),
       },
+      {
+        // Match URLs - must come after markdown patterns to avoid conflicts
+        // Matches http://, https://, and angle-bracket URLs like <https://...>
+        // Uses a capturing group to match URL content, but we'll use match[0] for full text
+        regex: /<?(https?:\/\/[^\s<>]+)>?/g,
+        render: (content, key) => (
+          <Text key={key} color="cyan">
+            {content}
+          </Text>
+        ),
+      },
     ];
 
     const renderMarkdown = (text: string | null | undefined) => {
@@ -121,6 +132,7 @@ const MarkdownRenderer: React.FC<MarkdownRendererProps> = React.memo(
         index: number;
         length: number;
         content: string;
+        fullMatch: string;
         render: (content: string, key: string) => React.ReactNode;
       }> = [];
 
@@ -138,10 +150,14 @@ const MarkdownRenderer: React.FC<MarkdownRendererProps> = React.memo(
           );
 
           if (!isInCodeBlock) {
+            // For URLs, use the full match to preserve angle brackets
+            // For other patterns, use capture groups as before
+            const isUrlPattern = pattern.regex.source.includes("https?");
             allMatches.push({
               index: match.index,
               length: match[0].length,
-              content: match[2] || match[1], // Use second capture group for headings, first for others
+              content: isUrlPattern ? match[0] : match[2] || match[1], // Use full match for URLs, capture groups for others
+              fullMatch: match[0],
               render: pattern.render,
             });
           }
@@ -154,6 +170,7 @@ const MarkdownRenderer: React.FC<MarkdownRendererProps> = React.memo(
         length: number;
         type: "code" | "other";
         content?: string;
+        fullMatch?: string;
         render?: (content: string, key: string) => React.ReactNode;
         language?: string;
         code?: string;
@@ -170,6 +187,7 @@ const MarkdownRenderer: React.FC<MarkdownRendererProps> = React.memo(
           length: match.length,
           type: "other" as const,
           content: match.content,
+          fullMatch: match.fullMatch,
           render: match.render,
         })),
       ];
@@ -227,7 +245,32 @@ const MarkdownRenderer: React.FC<MarkdownRendererProps> = React.memo(
       return parts;
     };
 
-    return <Text>{renderMarkdown(content)}</Text>;
+    const parts = renderMarkdown(content);
+
+    // Check if content has URLs by looking for cyan-colored Text elements
+    const hasUrls = parts.some(
+      (part) =>
+        React.isValidElement(part) && (part as any).props?.color === "cyan",
+    );
+
+    // Use Box layout when URLs are present to prevent them from breaking
+    // across lines. Each element (URL or text) becomes a separate flex item.
+    if (hasUrls) {
+      return (
+        <Box flexDirection="row" flexWrap="wrap">
+          {parts.map((part, idx) =>
+            React.isValidElement(part) ? (
+              part
+            ) : (
+              <Text key={`text-${idx}`}>{part}</Text>
+            ),
+          )}
+        </Box>
+      );
+    }
+
+    // No URLs: use original Text wrapping for backward compatibility
+    return <Text>{parts}</Text>;
   },
 );
 
diff --git a/extensions/cli/src/ui/MarkdownRenderer.url.test.tsx b/extensions/cli/src/ui/MarkdownRenderer.url.test.tsx
new file mode 100644
index 00000000000..3e507df3c65
--- /dev/null
+++ b/extensions/cli/src/ui/MarkdownRenderer.url.test.tsx
@@ -0,0 +1,93 @@
+import { render } from "ink-testing-library";
+import React from "react";
+import { describe, expect, it } from "vitest";
+
+import { MarkdownRenderer } from "./MarkdownRenderer.js";
+
+describe("MarkdownRenderer - URL handling", () => {
+  it("should render URLs without breaking", () => {
+    const content =
+      "Check out this link: https://github.com/continuedev/continue/discussions/8240";
+
+    const { lastFrame } = render(<MarkdownRenderer content={content} />);
+    const frame = lastFrame();
+
+    // URL should be present and complete
+    expect(frame).toContain(
+      "https://github.com/continuedev/continue/discussions/8240",
+    );
+    // Frame should contain the prefix text
+    expect(frame).toContain("Check out this link:");
+  });
+
+  it("should handle angle-bracket URLs", () => {
+    const content =
+      "expectation would be the link <https://github.com/continuedev/continue/discussions/8240>";
+
+    const { lastFrame } = render(<MarkdownRenderer content={content} />);
+    const frame = lastFrame();
+
+    // URL should be extracted and rendered, with angle brackets preserved
+    expect(frame).toContain(
+      "https://github.com/continuedev/continue/discussions/8240",
+    );
+  });
+
+  it("should handle multiple URLs in the same content", () => {
+    const content =
+      "Check https://example.com and also https://github.com/continuedev/continue";
+
+    const { lastFrame } = render(<MarkdownRenderer content={content} />);
+    const frame = lastFrame();
+
+    expect(frame).toContain("https://example.com");
+    expect(frame).toContain("https://github.com/continuedev/continue");
+  });
+
+  it("should handle URLs with markdown text", () => {
+    const content =
+      "This is **bold text** and this is a link: https://continue.dev and more text.";
+
+    const { lastFrame } = render(<MarkdownRenderer content={content} />);
+    const frame = lastFrame();
+
+    expect(frame).toContain("bold text");
+    expect(frame).toContain("https://continue.dev");
+  });
+
+  it("should handle http URLs", () => {
+    const content = "Visit http://localhost:3000/api/endpoint";
+
+    const { lastFrame } = render(<MarkdownRenderer content={content} />);
+    const frame = lastFrame();
+
+    expect(frame).toContain("http://localhost:3000/api/endpoint");
+  });
+
+  it("should use Box layout when URLs are present", () => {
+    const content = "Link: https://example.com";
+
+    const { lastFrame } = render(<MarkdownRenderer content={content} />);
+    const frame = lastFrame();
+
+    // Should contain the URL
+    expect(frame).toContain("https://example.com");
+    // Should render without errors
+    expect(frame).toBeTruthy();
+  });
+
+  it("should not break URLs across lines", () => {
+    // This is more of an integration test - the actual line breaking
+    // depends on terminal width, but we can verify the structure
+    const content =
+      "A very long URL: https://github.com/continuedev/continue/discussions/8240/with/many/path/segments";
+
+    const { lastFrame } = render(<MarkdownRenderer content={content} />);
+    const frame = lastFrame();
+
+    // The URL should be present as a complete unit
+    expect(frame).toContain(
+      "https://github.com/continuedev/continue/discussions/8240/with/many/path/segments",
+    );
+  });
+});