omniroute-cli

Warn

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install a global Node.js package omniroute from the public npm registry. This package does not originate from the skill author's verified infrastructure or a recognized trusted vendor.
  • [EXTERNAL_DOWNLOADS]: The skill relies on external remote resources for its core functionality, linking to multiple SKILL.md files hosted on GitHub under a personal account (diegosouzapw). These represent unverifiable remote dependencies.
  • Evidence: Found in the 'CLI capability skills' and 'API skills (REST)' sections pointing to https://raw.githubusercontent.com/diegosouzapw/OmniRoute/main/skills/....
  • [INDIRECT_PROMPT_INJECTION]: The skill's architecture relies on ingesting remote markdown content to define its capabilities, creating an attack surface where external content can influence the agent's logic.
  • Ingestion points: Remote URL references in SKILL.md under 'CLI capability skills' and 'API skills'.
  • Boundary markers: None identified in the provided file to delimit or sanitize the external content.
  • Capability inventory: The skill facilitates command-line execution (omniroute), environment variable management, and network communication with a server.
  • Sanitization: No evidence of validation or sanitization of the remote skill content before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 5, 2026, 07:21 PM