Skills
Navigate
skills/funcodingdev/omniroute/omniroute-web-search/Gen Agent Trust Hub

omniroute-web-search

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's documentation directs users to an external third-party GitHub repository (diegosouzapw/OmniRoute) for setup instructions and configuration.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data in the form of search queries which are then used in shell commands. This creates a vulnerability surface where instructions embedded in search queries or results could attempt to manipulate the agent's behavior.
  • Ingestion points: The query parameter used in the curl examples in SKILL.md.
  • Boundary markers: None; the query is interpolated directly into a JSON payload within a shell command without specific boundary delimiters.
  • Capability inventory: Shell command execution via curl and data processing via jq as demonstrated in SKILL.md.
  • Sanitization: The instructions do not include any logic for sanitizing or escaping user-provided queries before they are passed to the shell environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 07:22 PM