supabase-storage

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (HIGH): The skill provides an interface for the agent to interact with storage objects using data that may originate from untrusted sources (e.g., user-uploaded file names). \n
  • Ingestion points: Path and name parameters used in upload(), remove(), move(), and copy() methods within SKILL.md. \n
  • Boundary markers: No boundary markers, delimiters, or instructions to ignore embedded commands are present to protect the agent from path-based attacks. \n
  • Capability inventory: High-privilege file operations are exposed, including file deletion (remove), file movement (move), and data overwriting via upsert: true. \n
  • Sanitization: There is no evidence of sanitization, validation, or path-normalization logic to prevent an attacker from manipulating the agent into performing unauthorized operations via malicious input strings.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 10:37 AM