mcp-copilot-studio-server-generator

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface where untrusted data provided as context variables is used to generate implementation files.
  • Ingestion points: Context variables such as 'Server Purpose', 'Tools Needed', and 'Target APIs' are ingested to drive code generation in SKILL.md.
  • Boundary markers: The generation template does not include delimiters or instructions to treat user-provided variables as data rather than instructions.
  • Capability inventory: The skill facilitates the creation of C# logic (script.csx) and server-side code (Node.js, FastAPI, Azure Functions) with potential for external network and filesystem access.
  • Sanitization: There are no documented validation or escaping processes for the variables before they are interpolated into the generated scripts.
  • [SAFE]: No instances of hardcoded credentials, malicious persistence, or unauthorized data exfiltration were found in the analyzed skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 05:26 AM