deepscan-record

DeepScan Add Scan Record

All script paths are relative to this Skill directory.

Prerequisites: Requires being logged in, having selected a scan book, and having started a batch (refer to the deepscan-login, deepscan-task, and deepscan-session Skills).

Pre-check

Before executing, first confirm the batch exists:

python ../deepscan-session/scripts/session.py current

Exit code is 1 → First execute the deepscan-task Skill to select a scan book and start a batch, then continue.
Success → Continue adding.

Scenario 1: The user sends text

python scripts/record.py add-text "{用户文本内容}"

Example successful return:

{
  "record_id": "019caedb-69bd-...",
  "scan_result": "用户输入的文本",
  "scan_type": "text",
  "seq_number": "R1",
  "task_title": "默认扫码本"
}

Inform the user: The content "{scan_result}" has been added to the scan book "{task_title}" ({seq_number}).

Scenario 2: The user uploads an image

Use the Read tool to get the image's local path, then execute:

python scripts/record.py add-image "{图片本地路径}"

The script automatically completes:

Use zxing-cpp to decode the QR code/barcode in the image
Upload the image to OSS to obtain a URL
Add a scan record (scanned content + image URL)

Example successful return:

{
  "record_id": "019caedb-69bd-...",
  "scan_result": "https://cli.im",
  "scan_type": "qrcode",
  "image_url": "https://...",
  "seq_number": "R1",
  "task_title": "默认扫码本"
}

Inform the user: Scan record {seq_number} has been added, with content "{scan_result}".

Error Handling

No QR code in the image: Inform the user that no QR code was recognized in the image, and they can manually enter the content and add it using the text method instead.
No scan book selected / No batch: Guide the user to execute the deepscan-task Skill.
Missing dependencies: Prompt the user to execute pip install zxing-cpp Pillow。

Dependency Installation

pip install zxing-cpp Pillow