jq

Warn

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The installation scripts perform system-level modifications by acquiring elevated privileges and using package managers.
  • The script scripts/install.ps1 explicitly uses sudo to execute apt-get commands for system updates and package installation.
  • The scripts scripts/install.sh and scripts/install.ps1 invoke administrative package managers including brew, apt-get, dnf, pacman, and apk to install the jq utility.
  • [PROMPT_INJECTION]: The skill processes untrusted external JSON data, representing an indirect prompt injection attack surface.
  • Ingestion points: Untrusted JSON data enters the agent context via files or standard input (stdin) as described in references/cheatsheet.md.
  • Boundary markers: The instructions do not define delimiters or specific "ignore embedded instructions" warnings for the agent when processing JSON fields.
  • Capability inventory: The skill allows the agent to execute shell commands (jq) and run local installation scripts (scripts/install.sh, scripts/install.ps1).
  • Sanitization: There is no evidence of input validation, escaping, or filtering to prevent the agent from obeying instructions embedded within the JSON payloads.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 8, 2026, 04:18 PM
Security Audit — agent-trust-hub — jq