Trusted — Risk Score 0/100
Last scan:2 days ago Rescan
0 /100
industry_research_report
依托东方财富数据库,为指定行业生成深度研究报告
A legitimate industry research report generation skill that uses httpx to call the East Money (东方财富) financial API, saves base64-decoded PDF/DOCX output to disk, and requires only an EM_API_KEY environment variable.
Skill Nameindustry_research_report
Duration28.3s
Enginepi
Safe to install
No security concerns identified. The skill performs exactly as documented.
ResourceDeclaredInferredStatusEvidence
Network READ READ ✓ Aligned scripts/get_data.py:157 — httpx POST to ai-saas.eastmoney.com
Filesystem WRITE WRITE ✓ Aligned scripts/get_data.py:94 — writes base64-decoded PDF/DOCX files; declared in SKILL…
Environment READ READ ✓ Aligned scripts/get_data.py:66 — reads EM_API_KEY from os.environ; declared in SKILL.md
Shell NONE NONE No subprocess/shell invocation found
1 High 2 findings
🔑
High API Key 疑似硬编码凭证
API_KEY="your_api_key_here"
SKILL.md:44
🔗
Medium External URL 外部 URL
https://ai-saas.eastmoney.com/proxy/app-robo-advisor-api/assistant/write/industry/research
scripts/get_data.py:75

File Tree

2 files · 14.0 KB · 345 lines
Python 1f · 217L Markdown 1f · 128L
├─ 📁 scripts
│ └─ 🐍 get_data.py Python 217L · 9.9 KB
└─ 📝 SKILL.md Markdown 128L · 4.2 KB

Dependencies 1 items

PackageVersionSourceKnown VulnsNotes
httpx unpinned pip No Version not pinned in SKILL.md; consider pinning to a known stable release

Security Positives

✓ No shell execution or subprocess usage — pure Python with httpx and stdlib
✓ API key loaded exclusively from environment variable (EM_API_KEY), never hardcoded in code
✓ Network calls go to a well-known legitimate financial API endpoint (East Money/eastmoney.com)
✓ File writes are limited to saving PDF and DOCX reports in a predictable output directory
✓ Filename sanitization implemented (_safe_filename) to prevent path traversal
✓ Base64 decoding is used for binary file storage, a standard and safe approach
✓ DOCX text extraction is a local fallback with no network activity
✓ No iteration over environment variables, no credential harvesting
✓ No base64-to-shell, no curl|bash, no direct IP calls, no eval patterns
✓ Skill behavior is fully declared in SKILL.md with clear doc-to-code alignment