skills/mastra-ai/mastra/general-tasks/Gen Agent Trust Hub

general-tasks

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it defines a process for handling tasks involving external files and web content without implementing security boundaries.
  • Ingestion points: Metadata description specifies tasks involving 'files' and 'the web'.
  • Boundary markers: Not present. The instructions lack delimiters or warnings to ignore data-embedded instructions.
  • Capability inventory: The skill facilitates operations involving shell execution, file modification, and web access.
  • Sanitization: Not present. No validation or filtering of external inputs is defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 12:02 AM
Security Audit — agent-trust-hub — general-tasks