The Agent Skills Directory

[SAFE]: The skill performs legitimate file generation tasks without any detected security risks. Its execution flow is restricted to gathering user input and writing it into a local markdown template.\n- [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion point for untrusted data (user-provided homework instructions) that is interpolated into a markdown file. However, this is considered safe as the skill does not have any dangerous capabilities (such as shell execution, dynamic code loading, or network exfiltration) that could be triggered by the generated content. Boundary markers are not explicitly used, but the output is restricted to a structured document format.

my-homework-new