The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It accepts user-supplied business descriptions and incorporates them into prompts for sub-agents without using boundary markers or sanitization.
Ingestion points: Business descriptions provided by users via $ARGUMENTS in SKILL.md are passed to specialist agents.
Boundary markers: The skill does not use specific delimiters or instructions to help the sub-agents distinguish between developer instructions and user-provided data.
Capability inventory: The sub-agents are granted access to sensitive tools including Bash, Write, Edit, and TeamCreate.
Sanitization: No filtering or validation is performed on the input before it is used to generate tasks for the sub-agents.
[COMMAND_EXECUTION]: The skill utilizes the bash tool to perform a configuration check by echoing an environment variable (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS). This is used to dynamically determine the agent spawning strategy and is a benign use of system commands.

helmer