\r \r

AI Parse\r

\r A skill for parsing PDF files using Large Language Models.\r \r

Capabilities\r

\r

• Extract information from PDF files\r
• Return results in JSON or Markdown format\r
• Resume processing from existing task ID\r
• Save task ID information to JSON file for reference\r \r

Parameters\r

\r | Parameter | Type | Required | Description |\r |-----------|------|----------|-------------|\r | pdf_path | string | required | Path to the PDF file to process |\r | result_path | string | required | Path to save the parsing result |\r | format | string | required | Output format: "json" or "md" |\r | task_id_path | string | required | Path to save task ID information (JSON format) |\r | --task-id | string | optional | Existing task ID to resume processing |\r \r

Usage Examples\r

\r

Normal Upload Mode\r

\r

python handler.py \x3Cpdf_path> \x3Cresult_path> \x3Cformat> \x3Ctask_id_path>\r
```\r
\r
### Resume from Existing Task or Check Status\r
\r
```bash\r
python handler.py --task-id \x3Ctask_id> \x3Cresult_path> \x3Cformat>\r
```\r
\r
## Task ID File Format\r
\r
When using normal upload mode, a task ID file will be created at `task_id_path` with the following JSON structure:\r
\r
```json\r
{\r
  "task_id": "AAFXKO",\r
  "pdf_path": "test.pdf",\r
  "submit_time": "2026-04-04 00:33:27"\r
}\r
```\r
\r
This file can be used to:\r
- Track the submitted task\r
- Retrieve the task ID later for status checking\r
- Resume processing if interrupted\r
\r
## Implementation\r
\r
Implemented by `handler.py` which:\r
- Uploads PDF files to the processing service\r
- Polls for processing completion\r
- Downloads and saves results in the requested format\r
- Supports resuming from existing task IDs\r
- Saves task ID information to JSON file\r
\r
## Environment Requirements\r
\r
- Python 3.6+\r
- requests library\r
\r
## Return Value\r
\r
The parsed result will be saved to the specified `result_path` in the requested format:\r
- **JSON format:** Structured JSON with task details and extracted content\r
- **Markdown format:** Formatted Markdown with page-by-page content\r
\r
## Notes\r
\r
- For large PDF files, processing may take multiple minutes\r
- Free users can process 30 PDF pages - visit https://api.pinocch.com/index for extra trial credits\r
- The `--task-id` parameter can be used to resume processing if interrupted\r
- Check the console output for processing progress and status updates\r
- The task ID file is created immediately after successful upload\r
- **IMPORTANT FOR AGENTS:** Before declaring a task as failed, always use the task ID to check the current status of the task. Use the `--task-id` parameter to resume or verify the task status. The task may still be processing or have completed successfully.\r