The Agent Skills Directory

[SAFE]: The skill instructions and referenced documentation (guide.md, examples.md) focus exclusively on literary analysis and story structure. No code execution, network access, or sensitive file operations are requested.
[SAFE]: Frontmatter configuration restricts the skill to the Read tool, which is appropriate for its stated purpose of analyzing provided text files.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied story text. While this is an ingestion point for potentially untrusted data, the risk is assessed as safe because the skill lacks any dangerous capabilities (like shell execution, writing files, or network requests) that could be exploited via injection. The instructions also emphasize strict adherence to the source text, reducing the likelihood of the model following instructions embedded within the story data.
[DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or network exfiltration patterns were detected in the skill or its associated references.

plot-keypoints