The Agent Skills Directory

[SAFE]: The skill operates entirely on local text data and files within the user's current directory for the purpose of writing style analysis and application.\n- [SAFE]: No indicators of data exfiltration, credential harvesting, or remote code downloads are present in the skill's instructions or template.\n- [SAFE]: There is a potential surface for indirect prompt injection as the skill processes untrusted text samples provided by the user (Ingestion points: text input and ./blog-posts/*.md files mentioned in SKILL.md). However, the skill's capabilities are limited to reading and writing local text files (Capability inventory: File-write to ./STYLE_PROFILE.md in SKILL.md) and do not include dangerous tools like shell access or network requests. Boundary markers and input sanitization are not explicitly defined, but the lack of exploitable tools makes the risk negligible.

style-profile