modelscope / modelscope/ms-agent
cannot run skill-creator on Windows
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 4.4k
- Forks
- 521
- Avg merge
- 1d
- Merged PRs (30d)
- 36
Description
<QUERY>
帮我创建一个outlook skill 可以搜索,提取邮件,总结邮件等功能
(Help me create an Outlook skill that can search, extract emails, and summarize emails)
</QUERY>
<PLAN>
Based on the skill creation process, here's the step-by-step plan:
**Step 1: Understand the Skill with Concrete Examples**
- Clarify specific use cases for the Outlook skill:
- What types of searches? (by sender, date range, subject, keywords)
- What information to extract? (attachments, specific fields, email threads)
- What summarization formats? (brief summary, action items, thread overview)
- Gather example queries like:
- "Search for emails from John about the Q4 report"
- "Extract all attachments from emails this week"
- "Summarize the email thread about project Alpha"
**Step 2: Plan Reusable Skill Contents**
- **Scripts**: Python scripts for Outlook API integration
- `scripts/search_emails.py` - Search emails with various filters
- `scripts/extract_email_data.py` - Extract email content and metadata
- `scripts/download_attachments.py` - Download and manage attachments
- **References**:
- `references/outlook_api.md` - Outlook API documentation and authentication
- `references/search_patterns.md` - Common search query patterns
- **Assets**: (if needed)
- `assets/summary_template.md` - Template for email summaries
**Step 3: Initialize the Skill**
- Run: `scripts/init_skill.py outlook-email-manager --path <output-directory>`
- This creates the skill directory structure with templates
**Step 4: Edit the Skill**
- Implement the Python scripts for Outlook integration (using Microsoft Graph API or win32com)
- Test each script to ensure they work correctly
- Create reference documentation for API usage and authentication
- Write SKILL.md with:
- Frontmatter: name and comprehensive description including triggers
- Body: Instructions for searching, extracting, and summarizing emails
- References to bundled scripts and documentation
**Step 5: Package the Skill**
- Run: `scripts/package_skill.py <path/to/outlook-email-manager>`
- This validates and creates the distributable .skill file
**Step 6: Iterate**
- Test with real Outlook queries
- Refine based on performance and user feedback
</PLAN>
<SCRIPTS>
[
{
"name": "init_skill.py",
"path": "skill-creator\\scripts\\init_skill.py",
"description": "Required to initialize the new outlook-email-manager skill directory structure with proper templates"
},
{
"name": "package_skill.py",
"path": "skill-creator\\scripts\\package_skill.py",
"description": "Required to validate and package the completed skill into a distributable .skill file"
}
]
</SCRIPTS>
<REFERENCES>
[
{
"name": "workflows.md",
"path": "skill-creator\\references\\workflows.md",
"description": "Helpful for designing the multi-step email search, extraction, and summarization workflows"
},
{
"name": "output-patterns.md",
"path": "skill-creator\\references\\output-patterns.md",
"description": "Useful for structuring email summary outputs and extraction results"
}
]
</REFERENCES>
<RESOURCES>
[]
</RESOURCES>
[INFO:ms_agent]
======== Completed Skill Plan Response ========
[INFO:ms_agent]
======== Completed Skill Plan Response ========
<QUERY>
帮我创建一个outlook skill 可以搜索,提取邮件,总结邮件等功能
(Help me create an Outlook skill that can search, extract emails, and summarize emails)
</QUERY>
<TASKS>
1. **Initialize Skill Structure**
- Run `init_skill.py outlook-email-manager --path <output-directory>` to create base directory structure
2. **Implement Core Scripts**
- Create `scripts/search_emails.py`: Search by sender, date, subject, keywords using Microsoft Graph API
- Create `scripts/extract_email_data.py`: Extract body, headers, metadata, and threads
- Create `scripts/download_attachments.py`: Download and save email attachments
- Implement authentication handler for Microsoft Graph API/OAuth
3. **Create Reference Documentation**
- Write `references/outlook_api.md`: API endpoints, authentication setup, rate limits
- Write `references/search_patterns.md`: Common query examples and filter syntax
4. **Write SKILL.md**
- Add frontmatter: skill name "outlook-email-manager" and description with triggers
- Document usage instructions for search, extract, and summarize operations
- Include example queries and expected outputs
- Reference bundled scripts and documentation
5. **Test Functionality**
- Test search with various filters (sender, date range, keywords)
- Test extraction of different email components
- Test summarization with single emails and threads
- Verify authentication flow works correctly
6. **Package Skill**
- Run `package_skill.py <path/to/outlook-email-manager>` to create distributable .skill file
- Validate all components are properly bundled
7. **Iterate & Refine**
- Test with real-world Outlook queries
- Optimize performance and error handling
- Update documentation based on feedback
</TASKS>
[INFO:ms_agent]
======== Completed Skill Tasks Response ========
[INFO:ms_agent]
======== Completed Skill Tasks Response ========
<IMPLEMENTATION>
[
{
"script": "skill-creator/scripts/init_skill.py",
"parameters": {
"skill_name": "outlook-email-manager",
"skill_name": "outlook-email-manager",
"path": "skills/public"
"path": "skills/public"
}
}
}
}
]
]
</IMPLEMENTATION>
</IMPLEMENTATION>
[INFO:ms_agent] Spec files dumped to: temp_workspace\.spec
[INFO:ms_agent] Spec files dumped to: temp_workspace\.spec
[INFO:ms_agent] Spec files dumped to: temp_workspace\.spec
```shell
python skill-creator/scripts/init_skill.py outlook-email-manager --path skills/public
```shell
python skill-creator/scripts/init_skill.py outlook-email-manager --path skills/public
python skill-creator/scripts/init_skill.py outlook-email-manager --path skills/public
```packages
```packages
[WARNING:ms_agent] Executing code block in local environment!
[WARNING:ms_agent] Executing code block in local environment!
[WARNING:ms_agent] Executing code block in local environment!
[WARNING:ms_agent] Executing code block in local environment!
[INFO:ms_agent] Installing required packages: []
[INFO:ms_agent] Installing required packages: []
[INFO:ms_agent] Installing required packages: []
[INFO:ms_agent] Installing required packages: []
[INFO:ms_agent] Installing required packages: []
** Agent skill results: [
{
"success": false,
"output": "",
"messages": "C:\\Work\\Projects\\ms-agent\\.venv\\Scripts\\python.exe: can't open file 'C:\\\\Work\\\\Projects\\\\ms-agent\\\\temp_workspaceskill-creator\\\\scripts\\\\init_skill.py': [Errno 2] No such file or directory\n"
}
]
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with skill-creator/scripts/init_skill.py and reproduce the reported command on Windows: python skill-creator/scripts/init_skill.py outlook-email-manager --path skills/public. Inspect how the script path is resolved in the shown failure, and consider the issue done when initialization succeeds from the repository on Windows.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- operating-systems, tooling
- Issue type
- Bug
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100