citedy-video-shorts

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation references installation via the Smithery CLI and GitHub. These are standard methods for obtaining the vendor's package.
  • [DATA_EXFILTRATION]: The skill transmits data to www.citedy.com to facilitate video generation and publishing. This is a functional requirement for the service and targets the vendor's own infrastructure.
  • [COMMAND_EXECUTION]: Includes utility scripts (scripts/register.mjs and scripts/self-test.sh) used for registration and environment validation. These scripts are intended for manual execution by the user.
  • [PROMPT_INJECTION]: The skill acts as an orchestrator that ingests user-provided text (topics, product angles) to generate video scripts, creating a surface for indirect prompt injection.
  • Ingestion points: User-supplied input in the SKILL.md orchestration flow (topics, angles, and campaign ideas).
  • Boundary markers: Absent; there are no explicit delimiters instructed for separating untrusted user data from generated prompt templates.
  • Capability inventory: The skill has capabilities to perform network requests (REST/MCP) and write to local state files.
  • Sanitization: Not present; the skill depends on the model's internal safety guardrails for the generated script content.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 07:59 PM