vap-media

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill uses curl to interact with the VAP API. This is the intended behavior for the skill's primary purpose and is used to perform POST and GET requests to generate and retrieve media.
  • [DATA_EXFILTRATION] (LOW): The skill transmits user-provided prompts and an environment variable (VAP_API_KEY) to api.vapagent.com. While this domain is not on the trusted source list, it is the legitimate endpoint for the service provided by the skill.
  • [PROMPT_INJECTION] (LOW): The skill possesses a surface for Indirect Prompt Injection (Category 8) by processing untrusted user data.
  • Ingestion points: The description field in the user request is used to generate the media prompt (SKILL.md).
  • Boundary markers: Absent. User input is directly interpolated into JSON payloads without delimiters or instructions to ignore embedded commands.
  • Capability inventory: Subprocess execution of curl for network operations across all generation tasks.
  • Sanitization: Absent. There is no evidence of input validation or escaping for the user-provided strings before they are sent to the external API.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:45 PM