fix(gui): harden multimodal image attachments

packages/ui/acp/tests/codec.spec.ts

@@ -1,4 +1,5 @@
 import { describe, expect, it } from 'vitest'
+import { AttachmentId } from '@deepseek-ai/dsh-attachment'
 import type { ContentBlock } from '@deepseek-ai/dsh-llm'
 import type { TurnEndReason } from '@deepseek-ai/dsh-session'
 import { SessionId } from '@deepseek-ai/dsh-session'
@@ -35,6 +36,17 @@ describe('turnEndToStopReason', () => {
 describe('harnessBlockToAcpContent', () => {
   it('maps a text block to ACP text content', () => {
     expect(harnessBlockToAcpContent({ type: 'text', text: 'hi' })).toEqual({ type: 'text', text: 'hi' })
+    const attachmentId = AttachmentId(`sha256:${'a'.repeat(64)}`)
+    expect(harnessBlockToAcpContent({
+      type: 'image',
+      attachment: {
+        attachmentId,
+        mediaType: 'image/png',
+        bytes: 1,
+        width: 1,
+        height: 1,
+      },
+    })).toEqual({ type: 'text', text: `[image attachment ${attachmentId}]` })
   })
 
   it('returns undefined for non-text blocks (reasoning / plugin-added)', () => {

packages/ui/tui/README.md

@@ -124,6 +124,6 @@ Append-only; newly visible content follows the reusable request prefix and does
 ## Known Limitations and Deferred Work
 
 - **One configured session owns the transcript and editor** — questions from other agents can still use the shared overlay provider, but session rendering and prompt input remain bound to `sessionId`.
-- **Tool cards are text terminal presentations** — terminal, diff, and generic cards use tool-owned titles/content, but session content currently has no image block for inline image rendering.
+- **Images use explicit text markers** — the terminal cannot render inline raster images, so user, assistant, tool-result, and streaming image blocks render as `[image attachment <id>]` instead of disappearing.
 - **Non-TTY operation is intentionally unsupported** — app bundles that need automation must compose a one-shot or server front door (`dsh-cli-demo`, `dsh-acp`) rather than expecting an internal fallback.
 - **Manual `/skill:` invocation always reloads the full skill body** — the TUI does not detect a skill already present in the conversation, so repeated invocations append its instructions again.

packages/ui/tui/src/index.ts

@@ -424,6 +424,9 @@ function contentText(content: readonly ContentBlock[]): string {
       case 'tool-result':
         parts.push(contentText(block.content))
         break
+      case 'image':
+        parts.push(`[image attachment ${block.attachment.attachmentId}]`)
+        break
       default: {
         const rawType = (block as { type?: unknown }).type
         parts.push(`[${typeof rawType === 'string' ? rawType : 'content'}]`)
@@ -648,7 +651,14 @@ class AssistantMessageComponent extends Container {
   constructor(content: readonly ContentBlock[], showReasoning: boolean, palette: Palette, mdTheme: MarkdownTheme) {
     super()
     const reasoning = displayText(textBlocks(content, 'reasoning').trim())
-    const text = displayText(textBlocks(content, 'text').trim())
+    const text = displayText(content
+      .flatMap(block => block.type === 'text'
+        ? [block.text]
+        : block.type === 'image'
+          ? [`[image attachment ${block.attachment.attachmentId}]`]
+          : [])
+      .join('\n\n')
+      .trim())
     if (reasoning && showReasoning) {
       this.addChild(new Spacer(1))
       this.addChild(new Text(palette.italic(palette.muted('Reasoning')), 1, 0))
@@ -668,6 +678,7 @@ class AssistantMessageComponent extends Container {
 interface StreamingBlock {
   type: string
   text: string
+  block?: ContentBlock
 }
 
 class StreamingAssistantComponent extends Container {
@@ -691,6 +702,8 @@ class StreamingAssistantComponent extends Container {
       this.blocks.set(chunk.index, block)
     } else if (chunk.type === 'block-end' && (chunk.block.type === 'text' || chunk.block.type === 'reasoning')) {
       this.blocks.set(chunk.index, { type: chunk.block.type, text: chunk.block.text })
+    } else if (chunk.type === 'block-end' && chunk.block.type === 'image') {
+      this.blocks.set(chunk.index, { type: 'image', text: '', block: chunk.block })
     }
     this.rebuild()
   }
@@ -707,6 +720,7 @@ class StreamingAssistantComponent extends Container {
       .flatMap<ContentBlock>(([, block]) => {
         if (block.type === 'text') return [{ type: 'text', text: block.text }]
         if (block.type === 'reasoning') return [{ type: 'reasoning', text: block.text }]
+        if (block.type === 'image' && block.block?.type === 'image') return [block.block]
         return []
       })
     const component = new AssistantMessageComponent(content, this.showReasoning, this.palette, this.mdTheme)

packages/ui/tui/tests/tui.spec.ts

@@ -434,6 +434,29 @@ describe('pi-tui chat lifecycle and transcript', () => {
       step: 1,
       chunk: { type: 'tool-call-delta', index: 2, id: 'stream-tool' as never, argumentsDelta: '{}' },
     })
+    result.session.append('assistant/chunk', {
+      turn: 3,
+      step: 1,
+      chunk: { type: 'block-start', index: 3, blockType: 'image' },
+    })
+    result.session.append('assistant/chunk', {
+      turn: 3,
+      step: 1,
+      chunk: {
+        type: 'block-end',
+        index: 3,
+        block: {
+          type: 'image',
+          attachment: {
+            attachmentId: 'sha256:stream-image' as never,
+            mediaType: 'image/png',
+            bytes: 1,
+            width: 1,
+            height: 1,
+          },
+        },
+      },
+    })
     result.session.append('assistant/chunk', {
       turn: 3,
       step: 1,
@@ -441,6 +464,7 @@ describe('pi-tui chat lifecycle and transcript', () => {
     })
     await tick()
     expect(result.terminal.output).toContain('live thought')
+    expect(result.terminal.output).toContain('sha256:stream-image]')
     result.terminal.send('\x12')
     await tick()
     appendAssistant(
@@ -729,6 +753,16 @@ describe('pi-tui chat lifecycle and transcript', () => {
             { type: 'text', text: '# Heading\n\n[link](https://example.com) `code`\n\n```ts\nconst x = 1\n```\n\n> quote\n\n---\n\n- item\n\n**bold** *italic* ~~strike~~' },
             { type: 'tool-call', id: 'nested' as never, name: 'nested_tool', arguments: '{}' },
             { type: 'tool-result', toolCallId: 'nested' as never, content: [{ type: 'reasoning', text: 'nested result' }] },
+            {
+              type: 'image',
+              attachment: {
+                attachmentId: 'sha256:user-image' as never,
+                mediaType: 'image/png',
+                bytes: 1,
+                width: 1,
+                height: 1,
+              },
+            },
             { type: 'future-block' } as never,
             {} as never,
           ],
@@ -737,6 +771,16 @@ describe('pi-tui chat lifecycle and transcript', () => {
         appendAssistant(session, [
           { type: 'reasoning', text: 'styled reasoning' },
           { type: 'text', text: 'styled answer' },
+          {
+            type: 'image',
+            attachment: {
+              attachmentId: 'sha256:assistant-image' as never,
+              mediaType: 'image/png',
+              bytes: 1,
+              width: 1,
+              height: 1,
+            },
+          },
         ], { inputTokens: 2_000_000, outputTokens: 1_500_000 })
         session.append('todo/write', { todos: [
           { content: 'done', status: 'completed' },
@@ -756,6 +800,8 @@ describe('pi-tui chat lifecycle and transcript', () => {
     expect(result.terminal.output).toContain('Heading')
     expect(result.terminal.output).toContain('nested_tool({})')
     expect(result.terminal.output).toContain('nested result')
+    expect(result.terminal.output).toContain('sha256:user-image]')
+    expect(result.terminal.output).toContain('[image attachment sha256:assistant-image]')
     expect(result.terminal.output).toContain('[future-block]')
     expect(result.terminal.output).toContain('[content]')
     expect(result.terminal.output).toContain('↑2.0m ↓1.5m')