fastapi-project-structure

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Prompt Injection] (LOW): The MCP integration example in examples/mcp-integrated-example.md provides a surface for indirect prompt injection.
  • Ingestion points: The call_tool function within the tool definition template accepts arbitrary dictionary arguments (arguments: dict) which are processed by the application logic.
  • Boundary markers: The implementation lacks delimiters or instructions for the agent to ignore potentially malicious embedded commands in the tool's data flow.
  • Capability inventory: The tools interact with a DataService capable of reading and writing data, which could be exploited if an attacker-controlled input is processed.
  • Sanitization: No explicit sanitization, input filtering, or escaping is implemented in the provided example tool handlers.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:32 PM