diff --git a/.changeset/strong-lions-tan.md b/.changeset/strong-lions-tan.md
new file mode 100644
index 00000000000..16c6a5a8308
--- /dev/null
+++ b/.changeset/strong-lions-tan.md
@@ -0,0 +1,5 @@
+---
+'@primer/mcp': minor
+---
+
+Adds review_alt_text tool to the Primer MCP
diff --git a/packages/mcp/src/server.ts b/packages/mcp/src/server.ts
index 07f179c4ece..e33c5a20e17 100644
--- a/packages/mcp/src/server.ts
+++ b/packages/mcp/src/server.ts
@@ -629,4 +629,61 @@ The following list of coding guidelines must be followed:
   },
 )
 
+// -----------------------------------------------------------------------------
+// Accessibility
+// -----------------------------------------------------------------------------
+
+/**
+ * The `review_alt_text` tool is experimental and may be removed in future versions.
+ *
+ * The intent of this tool is to assist products like Copilot Code Review and Copilot Coding Agent
+ * in reviewing both user- and AI-generated alt text for images, ensuring compliance with accessibility guidelines.
+ * This tool is not intended to replace human-generated alt text; rather, it supports the review process
+ * by providing suggestions for improvement. It should be used alongside human review, not as a substitute.
+ *
+ *
+ **/
+server.tool(
+  'review_alt_text',
+  'Evaluates image alt text against accessibility best practices and context relevance.',
+  {
+    surroundingText: z.string().describe('Text surrounding the image, relevant to the image.'),
+    alt: z.string().describe('The alt text of the image being evaluated'),
+    image: z
+      .union([
+        z.instanceof(File).describe('The image src file being evaluated'),
+        z.string().url().describe('The URL of the image src being evaluated'),
+        z.string().describe('The file path of the image src being evaluated'),
+      ])
+      .describe('The image file, file path, or URL being evaluated'),
+  },
+  async ({surroundingText, alt, image}) => {
+    // Call the LLM through MCP sampling
+    const response = await server.server.createMessage({
+      messages: [
+        {
+          role: 'user',
+          content: {
+            type: 'text',
+            text: `Does this alt text: '${alt}' meet accessibility guidelines and describe the image: ${image} accurately in context of this surrounding text: '${surroundingText}'?\n\n`,
+          },
+        },
+      ],
+      sampling: {temperature: 0.4},
+      maxTokens: 500,
+    })
+
+    return {
+      content: [
+        {
+          type: 'text',
+          text: response.content.type === 'text' ? response.content.text : 'Unable to generate summary',
+        },
+      ],
+      altTextEvaluation: response.content.type === 'text' ? response.content.text : 'Unable to generate summary',
+      nextSteps: `If the evaluation indicates issues with the alt text, provide more meaningful alt text based on the feedback. DO NOT run this tool repeatedly on the same image - evaluations may vary slightly with each run.`,
+    }
+  },
+)
+
 export {server}