recipe-task
Pass
Audited by Gen Agent Trust Hub on Jul 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted user input ($ARGUMENTS) by interpolating it directly into instructions for a sub-agent (rule-advisor) without sanitization or boundary markers.
- Ingestion points: User input enters via the
$ARGUMENTSvariable inSKILL.md. - Boundary markers: Absent. The user input is placed directly inside a prompt string for the sub-agent:
"Analyze the task and select appropriate rules for: $ARGUMENTS...". - Capability inventory: The skill has the capability to spawn agents (
spawn_agent) and execute implementation steps based on the potentially influenced output of the sub-agent. - Sanitization: No evidence of input validation, escaping, or explicit instruction delimiters to prevent the agent from obeying instructions embedded within the task description.
Audit Metadata