python3 scripts/anygen.py upload --file ./reference.pdf
# Output: File Token: tk_abc123

python3 scripts/anygen.py prepare \
  --message "I need a presentation about AI trends. Key content from the doc: [extracted summary]" \
  --file-token tk_abc123 \
  --save ./conversation.json

python3 scripts/anygen.py prepare \
  --input ./conversation.json \
  --message "Focus on generative AI and enterprise adoption" \
  --save ./conversation.json

python3 scripts/anygen.py create \
  --operation slide \
  --prompt "<prompt from suggested_task_params>" \
  --file-token tk_abc123 \
  --style "business formal"
# Output: Task ID: task_xxx, Task URL: https://...

sessions_spawn(
    prompt=<subagent prompt below>,
    runTimeoutSeconds=1500  # REQUIRED: 25 minutes (1500s) to cover 20-min poll + buffer
)

You are a background monitor for an AnyGen content generation task.
You MUST strictly follow every instruction below. Do not skip, reorder, or improvise any step.

Task ID: {task_id}
Task URL: {task_url}
Operation: {operation}
Script: {script_path}
Render Script: {render_script_path}
Export Format: {export_format}
Thumbnail Output: {thumbnail_output_dir}
User Language: {user_language}

IMPORTANT: This monitoring task may take up to 20 minutes. Ensure your execution timeout is at least 25 minutes.

CRITICAL RULES:
- You MUST reply in {user_language}.
- After completing ALL steps (including sending messages to the user), your FINAL reply
  MUST be EXACTLY "ANNOUNCE_SKIP" — nothing else. This prevents the main session from
  sending duplicate messages.
- Do NOT say anything beyond what is specified below. No greetings, no extra commentary.
- NEVER include technical terms like "task_id", "file_token", "poll", "sub-agent",
  "API", "script", "workspace", "downloaded to", file paths, or status labels
  in any message to the user.

Your job depends on the operation type:

## For slide / doc operations:
1. Run: python3 {script_path} poll --task-id {task_id}
   This command blocks for up to 20 minutes waiting for task completion.
   Do NOT pass --output — the file will only be downloaded when the user requests it.
2. On success:
   a. Download thumbnail:
      python3 {script_path} thumbnail --task-id {task_id} --output {thumbnail_output_dir}
   b. Send the thumbnail preview image to the user. Choose the correct method:
      - Feishu/Lark: Two-step process via OpenAPI —
        Step 1 (Upload image): POST https://open.feishu.cn/open-apis/im/v1/images
          curl -X POST 'https://open.feishu.cn/open-apis/im/v1/images' \
            -H 'Authorization: Bearer <tenant_access_token>' \
            -F 'image_type=message' \
            -F 'image=@{thumbnail_output_dir}/thumbnail.png'
          This returns an image_key.
        Step 2 (Send image message): POST https://open.feishu.cn/open-apis/im/v1/messages?receive_id_type=chat_id
          Request body:
          {
            "receive_id": "<chat_id>",
            "msg_type": "image",
            "content": "{\"image_key\":\"<image_key>\"}"
          }
      - Other platforms: Send via message tool with filePath.
   c. Send a text message to the user (in {user_language}, natural tone):
      "Your content is ready! If you'd like me to send you the file, just let me know."
   d. Reply EXACTLY: ANNOUNCE_SKIP

## For smart_draw operation:
1. Run: python3 {script_path} poll --task-id {task_id} --output ~/.openclaw/workspace/
   This command blocks for up to 20 minutes waiting for task completion.
2. On success:
   a. Get the local file path from [RESULT] Local file: line.
   b. Render to PNG:
      - For drawio: bash {render_script_path} drawio <local_file> <local_file_without_ext>.png
      - For excalidraw: bash {render_script_path} excalidraw <local_file> <local_file_without_ext>.png
   c. Send the rendered PNG image to the user. Choose the correct method:
      - Feishu/Lark: Two-step process via OpenAPI —
        Step 1 (Upload image): POST https://open.feishu.cn/open-apis/im/v1/images
          curl -X POST 'https://open.feishu.cn/open-apis/im/v1/images' \
            -H 'Authorization: Bearer <tenant_access_token>' \
            -F 'image_type=message' \
            -F 'image=@<local_file_without_ext>.png'
          This returns an image_key.
        Step 2 (Send image message): POST https://open.feishu.cn/open-apis/im/v1/messages?receive_id_type=chat_id
          Request body:
          {
            "receive_id": "<chat_id>",
            "msg_type": "image",
            "content": "{\"image_key\":\"<image_key>\"}"
          }
      - Other platforms: Send via message tool with filePath.
   d. Send a text message to the user (in {user_language}, natural tone):
      "Your diagram is ready! You can view and edit it online here: {task_url}"
   e. Reply EXACTLY: ANNOUNCE_SKIP
   f. If rendering fails:
      Send text: "The diagram has been generated but I couldn't render a preview.
       You can view and edit it here: {task_url}"
      Reply EXACTLY: ANNOUNCE_SKIP

## For chat / storybook / data_analysis / website operations:
1. Run: python3 {script_path} poll --task-id {task_id}
   This command blocks for up to 20 minutes waiting for task completion.
   No --output needed — results are viewed online.
2. On success:
   a. Send a text message to the user (in {user_language}, natural tone):
      "Your content is ready! You can view it here: {task_url}"
   b. Reply EXACTLY: ANNOUNCE_SKIP

## Common error handling (all operations):
- On failure:
  a. Send a text message to the user (in {user_language}):
     "Unfortunately the generation didn't complete successfully.
      You can check the details here: {task_url}"
  b. Reply EXACTLY: ANNOUNCE_SKIP
- On timeout (20 min):
  a. Send a text message to the user (in {user_language}):
     "The generation is taking a bit longer than expected.
      You can check the progress here: {task_url}"
  b. Reply EXACTLY: ANNOUNCE_SKIP

python3 {script_path} download --task-id {task_id} --output ~/.openclaw/workspace/