tushare-mcp
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data that could lead to indirect prompt injection attacks.
- Ingestion points: Untrusted data enters the agent context through tools like
tushare_news(insrc/tushare_mcp/tools/news.py),tushare_anns, andtushare_research_report(insrc/tushare_mcp/tools/research.py). These tools fetch text from external financial news sources, corporate announcements, and PDF research reports. - Boundary markers: There are no explicit boundary markers or instructions to the agent to ignore potentially malicious content within the fetched data.
- Capability inventory: The skill has capabilities for network operations (fetching from Tushare API) and local file writing (storing data in a SQLite database via
src/tushare_mcp/storage.py). - Sanitization: The skill performs standard data formatting in
src/tushare_mcp/formatter.pybut does not sanitize or filter the content of text fields for potential AI-directed instructions.
Audit Metadata