The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted conversation data to generate HTML artifacts without explicit boundary markers or sanitization instructions in the prompt. This creates a surface where malicious instructions or scripts could be embedded in the generated visuals based on user-controlled conversation history.
[COMMAND_EXECUTION]: The skill suggests using shell commands like Bash(open ...) to display generated files. If the filename description is derived from user-controlled strings without proper sanitization, there is a risk of command injection through shell metacharacters.

show-me