list-tasks
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the taskmd command-line utility via the Bash tool. It constructs the command by interpolating the $ARGUMENTS variable directly into the shell string.
- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because user-provided arguments are executed in a shell without sanitization.
- Ingestion points: User input passed through the $ARGUMENTS variable in SKILL.md.
- Boundary markers: Absent; user input is not delimited or encapsulated within the shell command.
- Capability inventory: Shell command execution via the Bash tool.
- Sanitization: Absent; the skill does not escape or validate special characters in the input before execution.
Audit Metadata