file-uploads

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMPROMPT_INJECTIONNO_CODE
Full Analysis
  • Indirect Prompt Injection (MEDIUM): The skill establishes a workflow for ingesting and analyzing untrusted external content (user-provided file uploads) to perform high-capability tasks like code creation and system diagnosis.
  • Ingestion points: Processes external file uploads, cloud storage objects (S3/R2), and multipart data as described in the metadata and instructions.
  • Boundary markers: The instructions do not define delimiters (e.g., XML tags or clear separators) to prevent the agent from following instructions embedded within the uploaded content.
  • Capability inventory: The agent is empowered to create code patterns, diagnose system failures, and perform architectural reviews based on this data.
  • Sanitization: No explicit sanitization or instruction-filtering logic is provided; the skill relies solely on the agent's 'personality' to maintain security, which is insufficient against adversarial input.
  • No Code (SAFE): The provided skill content consists exclusively of markdown instructions and metadata. No executable scripts, binaries, or configuration files were found for analysis.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 03:42 AM