Development & Debugging
This guide details how to create or import a skill, and manage files, use the interactive terminal, and perform debug chat in Web IDE.
1. Creating & Importing Skills
Before writing code, you need to create a development project in the skills list. The platform supports two ways to create or load a skill:

1.1 Click the Create Area to Create a Skill
Open Studio → Skill → New, then enter the name, icon, and description. This example uses Extract Emails (the screenshot above shows the demo skill title). Confirm and wait for the sandbox and files to load. The initial workspace contains a skills/ directory and .gitignore.
Once in the three-column workspace, describe your requirements in the left chat panel, or create code files in the middle File Configuration file tree and edit them in the right-hand code editor. The standard file structure contains:
skills/extract-emails/SKILL.md: Declares skill metadata and model dispatch instructions.skills/extract-emails/extract.py: Python script handling extraction and deduplication.
You can create this reference implementation directly in your workspace:
skills/extract-emails/SKILL.md:
---
name: extract-emails
description: Extract all email addresses from text, deduplicate, and preserve original order.
---
# Email Extraction Skill
Identifies email addresses from text input, deduplicating them in first-seen order.
## Execution
Invoke via Python script:
```bash
python3 skills/extract-emails/extract.py "<input_text>"
```skills/extract-emails/extract.py:
import re
import sys
def extract_emails(text: str) -> list[str]:
pattern = r'[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}'
matches = re.findall(pattern, text)
return list(dict.fromkeys(matches))
if __name__ == '__main__':
content = sys.argv[1] if len(sys.argv) > 1 else ''
emails = extract_emails(content)
for email in emails:
print(email)AI generation uses the selected model and consumes points; guide generated files toward the intended structure.
1.2 Import an Existing Skill ZIP Archive
If you have a Skill backed up or shared by others, click Import in the upper-right corner of the Skills list page (or the Import existing skill empty-state card), and upload the ZIP archive in the Import Skill dialog. The system automatically extracts and restores all code files and configurations in the background so you can resume development immediately.
2. Workspace File Management
When you open the skill details page, the system initializes and provisions an isolated workspace in a secure sandbox container, loading your project in the middle File Configuration file tree with the code editor and bottom Terminal on the right.

2.1 Multi-file Management
You can right-click or use action buttons on the middle File Configuration file tree to create, delete, rename, and move files or folders to organize your project structure.
2.2 Online Code Editing
Clicking any file in the middle file tree opens it in the right-hand code editor:
- Auto-Save: The editor automatically saves and syncs your edits to the backend sandbox container as you type.
- Real-Time Workspace Sync: The file tree automatically monitors and syncs file changes. Whether you edit files, install package dependencies in the terminal, or background processes generate new files, the tree stays updated.
- Change Isolation: Any code changes made here only take effect instantly in the debugging environment, and will not directly affect live apps. To apply the latest code to production, you must click the "Publish" button to generate an official version. For details, please see Versions & Publishing.
2.3 Two-way Interactive Terminal
The command line terminal at the bottom right connects directly to the backend sandbox:
- Running Commands: You can enter various command line operations, such as installing required code dependencies online or running custom test and debugging scripts.
- Log Feedback: The terminal streams command execution logs in real time. If code or script execution errors occur, you can view the error messages directly in the terminal output to assist with debugging.
3. Agent Debug Chat
The agent debug panel on the left provides a testing environment, allowing you to test and call your custom skill logic in real time by chatting with the agent:
Use Contact [email protected] and [email protected]; CC [email protected] as the test input. Run the generated script in the terminal. The expected output contains two unique addresses in this order:
The terminal at the lower right of the screenshot shows this actual execution. Check the generated code, output, and ordering before publishing. Generated file layouts and model responses may vary.
3.1 Immediate Effect
Every time you modify and save your code in the right editor, you don't need to manually compile, build, or redeploy. Simply send a new message in the chat box on the left, and the system will run the latest code in the background, allowing you to see the changes instantly.
3.2 Conversational Workspace Modification
You can chat with the agent in the left panel to have it create, delete, or modify workspace files. The middle file tree and right-hand editor reflect these changes in real time.
3.3 Real-time File Export
Click Export Config in the dropdown menu beside the Skill name in the upper-left corner of the page to package all code and configuration files in the current workspace into a ZIP archive and download it locally for backup or sharing.
4. Related Guides
- Directory structure & metadata reference: Skill Package Format & Runtime Conventions
- Dependency installation & preparation: Initialization Script
- Version snapshot & rollback: Versions & Publishing
- Agent binding & invocation: Using Skills in Agents
Last updated