search-router
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs shell command execution through the
openclitool to facilitate network searches and dynamically discover command parameters using the-hflag. - [COMMAND_EXECUTION]: A Python one-liner is utilized within a shell pipeline to process and format JSON data returned by the search utility, representing a legitimate use of runtime script execution for data processing.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and processes untrusted data from diverse external search sources.
- Ingestion points: Search results retrieved from external websites and AI models via
opencli. - Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters or ignore instructions embedded within the search results.
- Capability inventory: The skill has the ability to execute shell commands and perform network queries.
- Sanitization: Absent; there are no instructions to escape, filter, or validate the content returned from external sources before processing.
Audit Metadata